Certification

Understand the Sun Certified Java Developer Exam Process

2009-06-20T23:33:00.000-07:00

OK, so now you know everything about the language. But can you actually build

something in it? You’ll hear that argument from some who’ve never taken (or passed)

the programmer’s exam. Obviously, they don’t understand how darn difficult the

programmer’s exam actually is, but nonetheless there is something to the claim that,

“just because you know how the compiler and VM work does not mean you can

develope software.” The Developer exam, which is unique in the IT exam world,

lets you answer that question (most often posed by a prospective employer).

In the Developer exam, you get to put your code where your mouth is by developing

a software application. In fact, the Developer exam isn’t even a multiple-choice test

but rather a project that you build, given a (somewhat sketchy) specification. You’re

told what to build, with some general guidelines, and then it’s up to you to implement

and deliver the program. You have an unlimited amount of time in which to finish

the project (as of this writing), but there is a short follow-up essay exam (taken at an

authorized testing center, just as the Programmer exam is). The follow-up questions

are largely used to verify that it was you (not your hotshot programmer brother-in-law

who owed you big time) who did the work. In other words, the follow-up exam asks

essay questions that only the project developer could answer (for example, “Justify

your design choice on…”).

First, we’ll lay out the facts of the exam—how it works, how you do it, etc., and

then we’ll dive into what you need to know to pass it. Keep in mind that the actual

knowledge you need to pass cannot be stuffed into a book this size, even if we made

the book big enough to crush a car. Being a programmer is one thing, but being a

developer is quite another. And you can’t become a developer just by memorizing

some facts. Study and memorization can work for passing the Programmer’s exam—

but that’s OK because the programmer’s exam is designed to verify that you’re smart

and that you really know the language. A prospective employer doesn’t have to train

you in Java if you’ve passed the programmer’s exam. But if your employer wants

to verify that you can follow a spec and implement a well-designed, maintainable,

correct application, then you need either previous experience successfully building

one or more Java applications or you need to pass the SCJD.

The next seven chapters (in other words, the rest of the book) show you what

you’ll need to know and do to pass the exam, but it’s up to you to do the heavy

lifting. And unless you’re already well-versed in some of the topics (Swing, Threads,

RMI, etc.) then you’ll need to do some outside reading and practice in those

technologies. We’re focusing here on what the exam assessors are looking for in

your finished project.

How Does It Work?

The exam has two parts, The Assignment and The Essay. You must successfully pass

both parts to become certified.

The Assignment

Once you register for the Developer’s exam, you’re given instructions for downloading

your assignment. There are many possible assignments that you might get. The

assignment is a 9- or 10-page document with instructions for completing the project.

Instructions include both the application specification and requirements for

implementation and delivery. It also includes notes about how the application

will be marked (evaluated, graded, assessed).

The Essay

Once you’ve submitted your assignment, you should immediately register for the

essay portion of the certification. You can’t register until after you’ve submitted your

completed assignment, but the sooner the better once you have submitted it. You

really want to take the essay portion while the application you just completed is still

fresh in your mind. The essay portion will feel somewhat familiar to you—it takes

place in an authorized Prometric testing center, just as the Programmer’s exam does.

You have 90 minutes to complete the essay portion, and it normally involves just a

handful (about five) questions.

The Assessment

Once you’ve submitted both your assignment and the follow-up essay, the two

pieces will be sent to the assessor for grading. It might be four weeks or so before

you learn the results.

Understand the Sun Certified Java Developer Exam Process

Are You a Good Candidate?

If you haven’t yet passed the Programmer’s (SCJP) exam, then stop right now and go

get certified. You must pass the Programmer’s exam before you’re allowed to register

for the Developer exam. So we figure that you’ll read the first part of the book, take

the Programmer’s exam (passing, of course), then come back at some point and start

reading this part. That means by the time you’re reading this part, this book should

already be dog-eared, marked-up, scratched, bent, and possibly dusty (from that dry

spell between taking the Programmer’s exam and going for the Developer exam).

If you got to this paragraph, then we assume you’re already a Sun Certified Java

Programmer. But are you ready for the Developer exam? Well, the good news is that

you don’t need to be ready when you register for the exam. You’ve got plenty of time

to complete the assignment once you download it. So unlike the Programmer’s

exam, you don’t have to wait until you’re at top form for passing the exam. You can

download the assignment, analyze what you’ll need to learn to complete it, and then

get to work. Sun (and most candidates) estimates that it takes between 90 and 120

hours of solid work to complete the exam, and that assumes you’re already familiar

with all the necessary technologies (networking, database, threads/locking, Swing,

etc.). Some people work for three weeks straight, as if the project were their full-time

job. Others work on it when they can, in their spare time, and might take several

months to actually finish it. Of course, there’s always the chance that you download

it and discover you’re way over your head and unlikely to get up-to-speed within a

year. But if you’ve passed the Programmer’s exam and you’re willing to commit the

time to work on it (plus whatever additional time you need to learn any required

technologies you’re not familiar with), then we say go for it…if you’ve got the

money (we’ll get to that in the next section).

Having said all that, we don’t recommend registering until you’ve read the rest of

this book. It’ll give you a better idea of what’s really involved, and you might decide

to wait a while if you’re still a beginner at some of these technologies. But if you’re

comfortable with at least three of the following, chances are you’re ready to at least

download the assignment:

■ Swing and GUI design

■ Networking issues: sockets and RMI

■ Database issues: searching and record-locking

■ Writing clear, maintainable code

■ OO design and development

How Much Does It Cost?

All you need is $250 (US dollars) and you’re in business…for the first part. The

SCJD is in two parts, remember: the exam assignment (the specification that you

download, implement, and submit) and the follow-up essay. The follow-up exam is

an additional $150. So you’re looking at $400 total to get your certification. There’s

no partial certification, so submitting your exam doesn’t get you anywhere unless

you successfully take the follow-up exam. In other words, you can’t be certified

without spending the $400.

How Long Does It Take?

As of this writing, there is no time limit specified for completing the assignment

once you’ve downloaded it, but we don’t advise waiting more than a year, as the

requirements could change. Plus, there’s a new requirement (although these

requirements could change at any time so check the Sun website frequently at

http://suned.sun.com for updates) that you must not use a version of Java that is

deemed “out of date.” The current definition of out of date is that your version must

not have been superceded by a new production version for more than 18 months by

the time you make your submission. What that means is that if your version has

been out for less than 18 months, you’re fine. If your version is older than 18

months (in other words, its official public release was more than 18 months ago),

then the version released directly after your version must be less than 18 months old.

So don’t take forever is what we’re saying, or you could find yourself rewriting your

application. It’s not good enough for your program to run on newer versions; you

need to indicate in your exam which version you’ve compiled and tested on.

What’s the Exam Deliverable?

Chapter 17 covers this in picky detail, but the short version is: a JAR file. As of this

writing, you must submit the entire application, including compiled working classes,

source code, and documentation in a single JAR file. Your assignment instructions

will specify exactly how you must submit it and the most important rule is that you

must not deviate in any way from the submission instructions.

Can I Develop with an IDE?

You can, but everything you submit must be your own creation. In other words,

no auto-generated code. So use an IDE as an editor but not as a GUI-building tool

Understand the Sun Certified Java Developer Exam Process

or something that implements your networking. And whatever you do, be sure to

test on a machine other than your development machine! When using an IDE

(or not, but there’s more of a danger when using an IDE) you can end up being

sheltered and protected from things like classpath issues, which allow your program

to run fine on your machine and then blow up (OK, just fail to run) at runtime on

another system.

How Is It Graded?

Once you’ve completed both the assignment and the essay exam, an assessor takes

both pieces and performs the assessment. Your project is first assumed to be correct

and is given a starting point value (currently 155 points, but this could change).

Then points are deducted through a variety of audits. For example, you might get

12 points deducted for issues with your coding conventions, and perhaps another 15

(out of a possible, say, 18 points) for problems with your record-locking or search

algorithm. That subtracts 27 from your starting total of 155, and leaves you with

128 points. Currently, the exam requires 124 points to pass, so you’d be good with

128. Your instructions will give you an idea of the relative importance of certain

evaluation (audit) criteria, but you won’t know the specific values for specific

violations. DISCLAIMER: the point values mentioned here are merely examples of

how the exam is graded; they are not the actual point values used in the assessment.

The only thing you will know with certainty is the relative importance of different

aspects of your project. We’ll give you one clue right now, though: code readability/

clarity and threading/locking will be extremely important. You’ll almost certainly

find these two areas carrying the most weight on your assignment instructions.

What Are the Exam Assessors Thinking?

OK, we aren’t mind readers, and for all we know the assessors are thinking about

last night’s Bellbottom Bowling party as they mark your exam. But we do know one

thing: they aren’t looking to see how clever an algorithm designer you are! If anything,

it’s just the opposite. When you think of the Developer exam, don’t think Lone

Ranger. Instead, think Team Player. And don’t even think about showing off your

programming prowess by revising the specification to do something even better and

cooler than what’s asked for. There’s a saying we have in the software world, and it

will serve you well to remember it while building your Developer project: “Code as

if the next guy to maintain it is a homicidal maniac who knows where you live.”

The exam assessors aren’t thinking like my-algorithm-is-bigger-than-yours

code mavericks. They aren’t looking for the next great breakthrough in

record-locking. They aren’t even looking for new, creative implementations.

They are looking for development that says, “I’m a thoughtful programmer. I

care about maintainability. Readability is very important to me. I want everyone

to understand what’s going on with the least amount of effort. I write careful,

correct code. My code might not be the snappiest, and it might even incur an

extra few bytes in order to make something less complex, but my logic is simple,

my code and my design are clear and implement the specification perfectly, I

didn’t reinvent the wheel anywhere, and gosh—wouldn’t you just love to have

me on your team?” If your project submission says all that about you, you’re

in great shape.

The exam assessor looks at your code first from an entirely selfish perspective by

asking, “Is this easy for me to evaluate?” Chapters 16 and 17 offer insight into what

you need to do to make the assessor’s job easier. Trust us on this one—they’d rather

be at the beach (or skiing, mountain-biking, taking a Martha Stewart crafts workshop)

than spending unnecessary time figuring out how to get your assignment working.

Beginning with the “refreshed” exam assignments at the end of 2002, the requirements

changed to make the submission rules much more strict, in order to benefit the assessor.

If at any time you neglect to follow even a single submission requirement—say, the

directory structure of your project puts the user documentation in a different

folder—you’ll be failed on the spot. The assessor won’t make any allowances for

misplaced files, even if the program still runs perfectly. Don’t make them go looking

for something.

Another aspect of making the assessor’s life easier is what you’ll learn in Chapters

11 and 12. The little things really matter! For example, while you might think—if

you indent your code four spaces—that an occasional three-space indentation here

and there is OK, what’s the harm in that? The harm is in readability, and while a

couple of inconsistencies in indentations might not be a big deal, adhering to the

Java Coding Conventions is absolutely crucial for others looking at your code…

especially the assessor.

We’ve seen people fail the exam because they put the curly braces on the line

below the method declaration rather than immediately following the declaration

(on the same line), violating the official Java Coding Conventions. While this

infraction alone probably might not cause you to fail, the points deducted for code

convention violations might be the ones that sink you where you otherwise might

have squeaked by. You don’t get to make very many mistakes in this exam. Just

because your manager or your co-workers are tolerant of a little sloppiness here and

there, the assessor won’t be. Reread the preceding Exam Watch, copy it down on a

post-it note, and stick it onto your bathroom mirror. Each morning, say it to

yourself, “I’m a thoughtful programmer. I care…” (except say the whole thing).

What Are the Exam Assessors NOT Thinking?

If your solution works—according to the spec—then even if the algorithms might

be tweaked just a little more for efficiency, you probably won’t be marked down—

especially if the code is clear, maintainable, and gets the job done correctly. They’re

also not looking for one particular solution. There is no one right way to implement

your assignment. There are a gazillion wrong ways, however, and we’ll be looking at

some of those throughout the rest of the book. But here’s one that’s guaranteed to

kill you (both on the exam and in the real world): deadlock. Remember, we talked

about threads in Chapter 8, and you’d better take it all very seriously. If there’s even

a chance that your design could lead to deadlock (it doesn’t have to actually cause

deadlock right before the assessor’s eyes) then you can probably kiss that $400 goodbye.

The bottom line is that they’re not looking for The Perfect Solution. But they’re

also not looking for innovative new approaches, regardless of how clever, when

well-known patterns or other solutions exist. They’re especially not looking for you

to reinvent the wheel or write your own, say, new and improved set of classes to

replace the perfectly working core library packages.

What’s the Assignment Like?

We can’t give you a real assignment from the actual exam, of course, but here are a

couple of examples to give you the flavor of what the specification might look like.

And don’t be thinking these are outlandish examples; wait ‘til you see the real ones.

WindRider Horse Cruises

WindRider Horse Cruises (WHC) offers a variety of unique vacation trips, all on

horseback. Copying the cruise ship model, WHC has 4-day, 7-day, and 14-day

cruises that include all the food, drinks, and partying you can handle. (Which, after

four straight days on a horse won’t be much.) WindRider has grown steadily from a

two-person outfit offering one cruise a month to a busy operation with several cruises

running simultaneously in different parts of the world. But while the business has

Understand the Sun Certified Java Developer Exam Process grown, their cruise booking software hasn’t kept pace. The WindRider CEO is

acting as the company’s IT director, but he has some quirks. He insists on keeping

the entire application—including the database server—homegrown. In other words,

he doesn’t want to buy or use a database server written by anyone but his trusted

friend Wilbur. Sadly, Wilbur sustained an injury while fulfilling his other

WindRider duties (training horses to tolerate the disco music) and that’s where you

come in. Your job is to build the new WindRider booking software. One restriction

is that you must use the WindRider’s existing data file format. All cruise records

must stay in that format because the accounting part of the company still has

software that requires that format, and you’re only updating the booking software.

Customers must be able to call in to one of the four booking offices and request

a cruise. A customer service representative then uses the new booking application

(the one you’re going to write) to search for and then book an appropriate cruise to

meet that customer’s needs. Although the data file lives on one machine, back at the

head office, the three other booking offices (and possibly more in the future) need

to be able to access it over a standard TCP/IP network. So there’s a danger that two

customer service agents could be trying to book the same cruise slot at the same

time. (A cruise slot is like a ‘cabin’ on a real seafaring cruise. So any given cruise

might have anywhere between 6 to 12 slots, and each slot represents a record in the

data file.) You’ll have to make sure that this doesn’t happen! Overbooking would be

a Really Bad Thing. (Especially for the horse.)

So the people who interact with the actual software are the customer service

agents. But it’s the actual cruise customers who are making the requests. For example, a

customer might phone up and say, “I’d like a 4- or 7-day Horse Cruise sometime in

August 2003, in the United States.” The customer service agent must then use the

system to perform a search for that customer’s needs. The application needs to provide

a list of all possible matching cruises and then also allow the agent to reserve (book)

a cruise slot for that customer.

You must use a Swing GUI, and WindRider’s CEO just happens to be dating a

Computer-Human Interaction specialist, so you can bet she’ll be looking for all the

right characteristics of a usable GUI.

For networking, you have the choice between RMI and using regular old Java

TCP sockets (with serialized objects). It’s really up to you to make that decision, but

you’d better be prepared to explain why you chose what you chose.

The data file format is a little ugly, not comma-delimited or anything, just a

bunch of fixed-length fields. And you must stick to this data file exactly. We’ll send

it to you so you can see exactly how it’s formatted and start working with it in your

development. You can’t change a thing about what goes into a record. You can’t add

a field, can’t reformat the data…nothing. Your job is simply to build the actual

database server that accesses the data file to allow for searching, booking, adding new

cruises, etc. Oh, and don’t forget about those concurrent user issues—you must lock

these records in some way during use.

From his hospital bed, Wilbur sketched out what the database interface should

be, and you need to follow this exactly (although you can add more, but you must at

least provide these two methods in your public interface to the database server).

public void updateRecord(String[] recordData, int whichRecord) throws

LockedRecordException, NoSuchRecordException;

public int[] findByCustomerCriteria(Criteria criteriaObject);

But then you still need to add the methods for deleting, locking, etc. And you’ll

have to create the custom Exceptions and decide what should go in the Criteria class

(the thing you’re going to use to search the database).

Your job, ultimately, is to deliver the following:

■ The customer service GUI application that they use to search and book

records in the database.

■ The actual database server application—the thing that actually gets into the

data file and takes care of locking, etc. This is most likely the piece the GUI

interacts with.

■ Networking functionality so that multiple users can access this, remotely.

Confused?

That’s part of the idea. You need to think through the problems, think about new

problems not addressed in this spec, and figure out how to solve them, even in the

face of incomplete information. The real world isn’t perfect. Specs never seem to be

complete. And the person you need to ask for clarification never seems to be at his

desk when you call. Oh, and there’s nobody—and we do mean nobody—who will

reassure you that you’re on the right track by implementing a particular solution.

You’re just going to have to roll your sleeves up and answer your own “what about

?” questions.

And boy oh boy are there issues. Both raised and unraised by this specification.

The majority of the rest of this book raises those issues and gives you a lot to think

about. We can’t give you solutions—there aren’t any right solutions, remember—and

it wouldn’t be ethical to work out all the issues here. That’s the whole point of the

Developer exam! The actual coding is quite straightforward and fairly simple. It’s not

like you’re writing the world’s greatest neural network or artificial life program. But

thinking about the true business issues—about what the customer might need, what

the customer service agents need, and what the business itself needs, and then planning

and implementing a solution—are what this certification is all about. You’ll thank us

one day. And don’t forget, if you get frustrated, just remember how much you like us

for getting you through the Programmer certification. Which we did, or of course

you wouldn’t be reading this far into the book!

Overview of the Developer Exam Chapters

We’re going to cover a lot of ground here, some at a high level and some a little

lower. The high-level areas are the places where you need to design solutions and

discover potential problems. Locking issues, for example, are handled at a high level.

We’ll raise issues to get you thinking, but you’ll have to come up with your own

designs—after all, we have no way of knowing what your exact assignment will be.

The lower-level areas are reserved for things you must do throughout your entire

application—such as coding standards, OO design, documentation, etc., and for

tools such as javadoc and Jar. We also cover GUI usability in some depth, but it will

be up to you to work out the implementations. The following is a chapter-by-chapter

look at what we cover in the rest of the book:

MCP 70-536 Recent Exam questions

2009-06-20T19:30:00.000-07:00

You are writing a custom dictionary. The custom-dictionary class is named MyDictionary. You need to ensure that the dictionary is type safe.

Which code segment should you use?

class MyDictionary : Dictionary

class MyDictionary : HashTable

class MyDictionary : IDictionary

class MyDictionary { ... }

Dictionary t = new Dictionary();

MyDictionary dictionary = (MyDictionary)t;

You write a class named Employee that includes the following code segment.

public class Employee {

string employeeId, employeeName, jobTitleName;

public string GetName() { return employeeName;

}

public string GetTitle() {

return jobTitleName;

}

You need to expose this class to COM in a type library. The COM interface must also facilitate forward-compatibility across new versions of the Employee class. You need to choose a method for generating the COM interface.

What should you do?

Add the following attribute to the class definition.

[ClassInterface(ClassInterfaceType.None)]

public class Employee {

Add the following attribute to the class definition.

[ClassInterface(ClassInterfaceType.AutoDual)]

public class Employee {

Add the following attribute to the class definition.

[ComVisible(true)]

public class Employee {

Define an interface for the class and add the following attribute to the class definition.

[ClassInterface(ClassInterfaceType.None)]

public class Employee : IEmployee

{

You are working on a debug build of an application.

You need to find the line of code that caused an exception to be thrown. Which property of the Exception class should you use to achieve this goal?

Data

Message

StackTrace

Source

You need to write a code segment that performs the following tasks:

* Retrieves the name of each paused service.

* Passes the name to the Add method of Collection1.

Which code segment should you use?

ManagementObjectSearcher searcher =

new ManagementObjectSearcher(

"Select * from Win32_Service where State = ‘Paused’");

for each (ManagementObject svc in searcher.Get()) {

Collection1.Add(svc["DisplayName"]);

}

ManagementObjectSearcher searcher =

new ManagementObjectSearcher(

"Select * from Win32_Service", "State = ‘Paused’");

for each (ManagementObject svc in searcher.Get()) {

Collection1.Add(svc["DisplayName"]);

}

ManagementObjectSearcher searcher =

new ManagementObjectSearcher(

"Select * from Win32_Service");

for each (ManagementObject svc in searcher.Get()) {

if ((String) svc["State"] == "'Paused'") {

Collection1.Add(svc["DisplayName"]);

}}

ManagementObjectSearcher searcher = new ManagementObjectSearcher();

searcher.Scope = new ManagementScope("Win32_Service");

for each (ManagementObject svc in searcher.Get()) {

if ((String)svc["State"] == "Paused") {

Collection1.Add(svc["DisplayName"]);

}}

You are developing an application that dynamically loads assemblies from an application directory.

You need to write a code segment that loads an assembly named Braindumps1.dll into the current application domain. Which code segment should you use?

AppDomain domain = AppDomain.CurrentDomain;

string myPath = Path.Combine(domain.BaseDirectory, "Braindumps1.dll");

Assembly asm = Assembly.LoadFrom(myPath);

AppDomain domain = AppDomain.CurrentDomain;

string myPath = Path.Combine(domain.BaseDirectory, "Braindumps1.dll");

Assembly asm = Assembly.Load(myPath);

AppDomain domain = AppDomain.CurrentDomain;

string myPath = Path.Combine(domain.DynamicDirectory, "Braindumps1.dll");

Assembly asm = AppDomain.CurrentDomain.Load(myPath);

AppDomain domain = AppDomain.CurrentDomain;

Assembly asm = domain.GetData("Braindumps1.dll");

You are testing a newly developed method named PersistToDB. This method accepts a parameter of type EventLogEntry. This method does not return a value. You need to create a code segment that helps you to test the method. The code segment must read entries from the application log of local computers and then pass the entries on to the PersistToDB method. The code block must pass only events of type Error or Warning from the source MySource to the PersistToDB method. Which code segment should you use?

EventLog myLog = new EventLog("Application", ".");

for each (EventLogEntry entry in myLog.Entries) {

if (entry.Source == "MySource") {

PersistToDB(entry);

}}

EventLog myLog = new EventLog("Application", ".");

myLog.Source = "MySource";

for each (EventLogEntry entry in myLog.Entries) {

if (entry.EntryType == (EventLogEntryType::Error &

EventLogEntryType::Warning)) {

PersistToDB(entry);

}}

EventLog myLog = new EventLog("Application", ".");

for each (EventLogEntry entry in myLog.Entries) {

if (entry.Source == "MySource") {

if (entry.EntryType == EventLogEntryType::Error ||

entry.EntryType == EventLogEntryType::Warning) {

PersistToDB(entry);

}

}}

EventLog myLog = new EventLog("Application", ".");

myLog.Source = "MySource";

for each (EventLogEntry entry in myLog.Entries) {

if (entry.EntryType == EventLogEntryType::Error ||

entry.EntryType == EventLogEntryType::Warning) {

PersistToDB(entry);

}}

You are developing an application for a client residing in Hong Kong.

You need to display negative currency values by using a minus sign. Which code segment should you use?

NumberFormatInfo culture = new CultureInfo("zh-HK")::NumberFormat;

culture.NumberNegativePattern = 1;

return numberToPrint.ToString("C", culture);

NumberFormatInfo culture = new CultureInfo("zh-HK")::NumberFormat;

culture.CurrencyNegativePattern = 1;

return numberToPrint.ToString("C", culture);

CultureInfo culture = new CultureInfo("zh-HK");

return numberToPrint.ToString("-(0)", culture);

CultureInfo culture = new CultureInfo("zh-HK");

return numberToPrint.ToString("()", culture);

Your application uses two threads, named threadOne and threadTwo.

You need to modify the code to prevent the execution of threadOne until threadTwo completes

execution.

What should you do?

Configure threadOne to run at a lower priority.

Configure threadTwo to run at a higher priority.

Use a WaitCallback delegate to synchronize the threads.

Call the Sleep method of threadOne.

Call the SpinLock method of threadOne.

You are writing an application that uses SOAP to exchange data with other applications.

You use a Department class that inherits from ArrayList to send objects to another application.

The Department object is named dept.

You need to ensure that the application serializes the Department object for transport by using SOAP. Which code should you use?

SoapFormatter formatter = new SoapFormatter();

byte[] buffer = new byte[dept.Capacity];

MemoryStream stream = new MemoryStream(buffer);

Foreach (object o in dept) {

formatter.Serialize(stream, o);

}

SoapFormatter formatter = new SoapFormatter();

byte[] buffer = new byte[dept.Capacity];

MemoryStream stream = new MemoryStream(buffer);

formatter.Serialize(stream, dept);

SoapFormatter formatter = new SoapFormatter();

MemoryStream stream = new MemoryStream();

foreach (object o in dept) {

Formatter.Serialize(stream, o);

}

SoapFormatter formatter = new SoapFormatter();

MemoryStream stream = new MemoryStream();

formatter.Serialize(stream, dept);

10)

You need to write a code segment that will create a common language runtime (CLR) unit of isolation within an application. Which code segment should you use?

AppDomainSetup mySetup = AppDomain::CurrentDomain::SetuupInformation;

mySetup.ShadowCopyFiles = "true";

System::Diagnostics::Process myProcess;

myProcess = new System::Diagnostics::Process();

AppDomain domain;

domain = AppDomain::CreateDomain("BraindumpsDomain");

System::ComponentModel::Component myComponent;

myComponent = new System::ComponentModel::Component();

11)

You are developing an application that will use custom authentication and role-based security.

You need to write a code segment to make the runtime assign an unauthenticated principal object to each running thread.

Which code segment should you use?

AppDomain domain = AppDomain::CurrentDomain;

domain.SetPrincipalPolicy(PrincipalPolicy::WindowsPrincipal);

AppDomain domain = AppDomain::CurrentDomain;

domain.SetThreadPrincipal(new WindowsPrincipal(nullptr));

AppDomain domain = AppDomain::CurrentDomain;

Domain.SetAppDomainPolicy(PolicyLevel::CreateAppDomainLevel());

AppDomain domain = AppDomain::CurrentDomain;

Domain.SetPrincipalPolicy(PrincipalPolicy::UnauthenticatedPrincipal);

12)

You need to write a code segment that transfers the contents of a byte array named dataToSend by using a NetworkStream object named netStream. You need to use a cache of size 8,192 bytes. Which code segment should you use?

MemoryStream memStream = new MemoryStream(8192);

memStream.Write(dataToSend, 0, (int) netStream.Length);

MemoryStream memStream = new MemoryStream(8192);

netStream.Write(dataToSend, 0, (int) memStream.Length);

BufferedStream bufStream = new BufferedStream(netStream, 8192);

bufStream.Write(dataToSend, 0, dataToSend.Length);

BufferedStream bufStream = new BufferedStream(netStream);

bufStream.Write(dataToSend, 0, 8192);

13)

You are writing a method to compress an array of bytes. The bytes to be compressed are passed to the method in a parameter named document.

You need to compress the contents of the incoming parameter.

Which code segment should you use?

MemoryStream inStream = new MemoryStream(document);

GZipStream zipStream = new GZipStream(inStream, CompressionMode.Compress);

byte[] result = new Byte[document.Length];

zipStream.Write(result, 0, result.Length);

return result;

MemoryStream Stream = new MemoryStream(document);

GZipStream zipStream = new GZipStream(stream, CompressionMode.Compress);

zipStream.Write(document, 0, document.Length);

zipStream.Close();

return stream.ToArray();

MemoryStream outStream = new MemoryStream();

GZipStream zipStream = new GZipStream(outStream, CompressionMode.Compress);

zipStream.Write(document, 0, document.Length);

zipStream.Close();

return outStream.ToArray();

MemoryStream inStream = new MemoryStream(document);

GZipStream zipStream = new GZipStream(inStream, CompressionMode.Compress);

MemoryStream outStream = new MemoryStream();

int b;

while ((b = zipStream.ReadByte()) != -1) {

outStream.WriteByte((byte)b);

}

return outStream.ToArray();

14)

You need to select a class that is optimized for key-based item retrieval from both small and large collections. Which class should you choose?

OrderedDictionary class

HybridDictionary class

ListDictionary class

Hashtable class

15)

You create a class library that contains the class hierarchy defined in the following code segment.

(Line numbers are included for reference only.)

01. public ref class Employee {

02.

03. public :

04. String Name;

05. };

06.

07. public ref class Manager : public Employee {

08.

09.public :

10.int Level;

11. };

12.

13. public ref class Group {

14.

15. public :

16. array Employees;

17. };

You create an instance of the Group class. You populate the fields of the instance. When you attempt to serialize the instance by using

the Serialize method of the XmlSerializer class, you receive InvalidOperationException.

You also receive the following error message: "There was an error generating the XML document."

You need to modify the code segment so that you can successfully serialize instances of the Group class by using the XmlSerializer class. You also need to ensure that the XML output contains an element for all public fields in the class hierarchy. What should you do?

Insert the following code between lines 14 and 15 of the code segment:

[XmlArrayItem(Type = __typeof(Employee))]

[XmlArrayItem(Type = __typeof(Manager))]

Insert the following code between lines 14 and 15 of the code segment:

[XmlElement(Type = __typeof(Employees))]

Insert the following code between lines 14 and 15 of the code segment:

[XmlArray(ElementName="Employees")]

Insert the following code between lines 3 and 4 of the code segment:

[XmlElement(Type = __typeof(Employee))]

andInsert the following code segment between lines 8 and 9 of the code segment:

[XmlElement(Type = __typeof(Manager))]

16)

You create an application that stores information about your customers who reside in various regions. You are developing internal utilities for this application.

You need to gather regional information about your customers in Canada.

Which code segment should you use?

for each (CultureInfo culture in CultureInfo::GetCultures(CultureTypes::SpecificCultures)) {

// Output the region information...}

CultureInfo cultureInfo = new CultureInfo("CA");

// Output the region information...

RegionInfo regionInfo = new RegionInfo("CA

// Output the region information...

RegionInfo regionInfo = new RegionInfo("");

if(regionInfo.Name == "CA"){

// Output the region information...}

17)

You need to generate a report that lists language codes and region codes.

Which code segment should you use?

for each (CultureInfo culture in

CultureInfo::GetCultures(CultureTypes::SpecificCultures)) {

// Output the culture information...}

CultureInfo culture = new CultureInfo("");

CultureTypes types = culture.CultureTypes;

// Output the culture information...

for each (CultureInfo culture in

CultureInfo::GetCultures(CultureTypes::NeutralCultures)) {

// Output the culture information...}

for each (CultureInfo culture in

CultureInfo::GetCultures(CultureTypes::ReplacementCultures)) {

// Output the culture information...}

18)

You need to identify a type that meets the following criteria: ?

Is always a number.?

Is not greater than 65,535. Which type should you choose?

System.UInt16

int

System.String

System.IntPtr

19)

You are writing code for user authentication and authorization. The username, password, and roles are stored in your application data store.

You need to establish a user security context that will be used for authorization checks such as IsInRole. You write the following code segment to authorize the user.

if (!TestPassword(userName, password))

throw new Exception("could not authenticate user");

String[] userRolesArray =

LookupUserRoles(userName);

You need to complete this code so that it establishes the user

security context. Which code segment should you use?

GenericIdentity ident = new GenericIdentity(userName);

GenericPrincipal currentUser = New GenericPrincipal(ident, userRolesArray);

Thread.CurrentPrincipal = currentUser;

WindowsIdentity ident = new WindowsIdentity(userName);

WindowsPrincipal currentUser = new WindowsPrincipal(ident);

Thread.CurrentPrincipal = currentUser;

NTAccount userNTName = new NTAccount(userName);

GenericIdentity ident = new GenericIdentity(userNTName.Value);

GenericPrincipal currentUser = new GenericPrincipal(ident, userRolesArray);

Thread.CurrentPrincipal = currentUser;

IntPtr token = IntPtr.Zero;

token = LogonUserUsingInterop(username, encryptedPassword);

WindowsImpersonationContext ctx = WindowsIdentity.Impersonate(token);

20)

You are creating a strong-named assembly named Braindumps1 that will be used in multiple applications. Braindumps1 will be rebuilt frequently during the development cycle. You need to ensure that each time the assembly is rebuilt it works correctly with each application that uses it.

You need to configure the computer on which you develop Braindumps1 such that each application uses the latest build of Braindumps1.

Which two actions should you perform? (Each correct answer presents part of the solution. Choose two.)

Create a DEVPATH environment variable that points to the build output directory for the strong-named assembly.

Add the following XML element to the machine configuration file:

publicKeyToken="32ab4ba45e0a69a1"

language="en-US" version="*.*.*.*" />

Add the following XML element to the configuration file of each application that uses the strong-named assembly:

publicKeyToken="32ab4ba45e0a69a1"

language="en-US" version="*.*.*.*" />

21)

You need to return the contents of an isolated storage file as a string. The file is machine-scoped and is named Settings.dat. Which code segment should you use?

IsolatedStorageFileStream isoStream;

isoStream = new IsolatedStorageFileStream("Settings.dat", FileMode::Open);

String result = (new StreamReader(isoStream)).ReadToEnd();

IsolatedStorageFile isoFile;

isoFile = IsolatedStorageFile::GetMachineStoreForAssembly();

IsolatedStorageFileStream isoStream;

isoStream = new IsolatedStorageFileStream("Settings.dat", FileMode::Open, isoFile);

String result = (new StreamReader(isoStream)).ReadToEnd();

IsolatedStorageFileStream isoStream;

isoStream = new IsolatedStorageFileStream("Settings.dat", FileMode::Open);

String result = isoStream.ToString();

IsolatedStorageFile isoFile;

isoFile = IsolatedStorageFile::GetMachineStoreForAssemblly();

IsolatedStorageFileStream isoStream;

isoStream = new IsolatedStorageFileStream("Settings.dat", FileMode::Open, isoFile);

String result = isoStream.ToString();

22)

You are defining a class named BraindumpsClass that contains several child objects.

BraindumpsClass contains a method named ProcessChildren that performs actions on the child objects. BraindumpsClass objects will be serializable.

You need to ensure that the ProcessChildren method is executed after the BraindumpsClass object and all its child objects are reconstructed.

Which two actions should you perform? (Each correct answer presents part of the solution. Choose two.)

Apply the OnDeserializing attribute to the ProcessChildren method.

Specify that BraindumpsClass implements the IDeserializationCallback interface.

Specify that BraindumpsClass inherits from the ObjectManager class.

Apply the OnSerialized attribute to the ProcessChildren method.

Create a GetObjectData method that calls ProcessChildren.

Create an OnDeserialization method that calls ProcessChildren.

23)

You develop a service application that needs to be deployed. Your network administrator creates a specific user account for your service application. You need to configure your service application to run in the context of this specific user account. What should you do?

Prior to installation, set the StartType property of the ServiceInstaller class.

Prior to installation, set the Account, Username, and Password properties of the ServiceProcessInstaller class.

Use the CONFIG option of the net.exe command-line tool to install the service.

Use the installutil.exe command-line tool to install the service.

24)

You need to write a multicast delegate that accepts a DateTime argument and returns a Boolean value. Which code segment should you use?

public delegate int PowerDeviceOn(bool, DateTime);

public delegate bool PowerDeviceOn(Object, EventArgs);

public delegate void PowerDeviceOn(DateTime);

public delegate bool PowerDeviceOn(DateTime);

25)

You are writing an application that uses isolated storage to store user preferences. The application uses multiple assemblies. Multiple users will use this application on the same computer. You need to create a directory named Preferences in the isolated storage area that is

scoped to the current Microsoft Windows identity and assembly.

Which code segment should you use?

IsolatedStorageFile store;

store = IsolatedStorageFile::GetUserStoreForAssembly();

store.CreateDirectory("Preferences");

IsolatedStorageFile store;

store = IsolatedStorageFile::GetMachineStoreForAssembly();

store.CreateDirectory("Preferences");

IsolatedStorageFile store;

store = IsolatedStorageFile::GetUserStoreForDomain();

store.CreateDirectory("Preferences");

IsolatedStorageFile store;

store = IsolatedStorageFile::GetMachineStoreForApplication();

store.CreateDirectory("Preferences");

26)

You are using the Microsoft Visual Studio 2005 IDE to examine the output of a method that returns a string. You assign the output of the method to a string variable named fName. You need to write a code segment that prints the following on a single line The message: "Test Failed: " The value of fName if the value of fName does not equal "Braindumps" You also need to ensure that the code segment simultaneously facilitates uninterrupted execution of the application. Which code segment should you use?

Debug::Assert(fName == "Braindumps", "Test Failed: ", fName);

Debug::WriteLinelf(fName != "Braindumps", fName, "Test Failed");

if (fName != "Braindumps") {

Debug::Print("Test Failed: ");

Debug::Print(fName);

}

if (fName != "Braindumps") {

Debug::WriteLine("Test Failed: ");

Debug::WriteLine(fName);

}

27)

You are developing a method that searches a string for a substring. The method will be localized to Italy.

Your method accepts the following parameters: The string to be searched, which is named searchListThe string for which to search, which is named searchValue You need to write the code. Which code segment should you use?

return searchList.IndexOf(searchValue);

CompareInfo comparer = new CultureInfo("it-IT")::CompareInfo;

return comparer.Compare(searchList, search Value);

CultureInfo comarer = new CultureInfo("it-IT");

if (searchList.IndexOf(searchValue) > 0) {

return true;

} else {

return false;

}

CompareInfo comparer = new CultureInfo("it-IT")::CompareInfo;

if (comparer.IndexOf(searchList, searchValue) > 0) {

return true;

} else {

return false;

28)

You are creating an application that retrieves values from a custom section of the application configuration file. The custom section uses XML as shown in the following block.

You need to write a code segment to define a class named Role. You need to ensure that the Role class is initialized with values that are retrieved from the custom section of the configuration file. Which code segment should you use?

public class Role : ConfigurationElement {

internal string_ElementName = "name";

[ConfigurationProperty("role")]

public string Name {

get {

return ((string)base["role"]);

}

public class Role : ConfigurationElement {

internal string_ElementName = "role";

[ConfigurationProperty("name", RequiredValue = true)]

public string Name {

get {

return ((string)base["name"]);

}

public class Role : ConfigurationElement {

internal string_ElementName = "role";

private String_name;

[ConfigurationProperty("name")]

public string Name {

get {

return_name;

}

public class Role : ConfigurationElement {

internal string_ElementName = "name";

private String_name;

[ConfigurationProperty("role", RequiredValue = true)]

public string Name {

get {

return_name;

}

29)

You are developing a method to encrypt sensitive data with the Data Encryption Standard (DES) algorithm. Your method accepts the following parameters:

The byte array to be encrypted, which is named messageAn encryption key, which is named keyAn initialization vector, which is named iv

You need to encrypt the data. You also need to write the encrypted data to a MemoryStream object.

Which code segment should you use?

DES des = new DESCryptoServiceProvider();

des.BlockSize = message.Length;

ICryptoTransform crypto = des.CreateEncryptor(key, iv);

MemoryStream cipherStream = new MemoryStream();

CryptoStream cryptoStream = new CryptoStream(cipherStream,

crypto, CryptoStreamMode.Write);

cryptoStream.Write(message, 0, message.Length);

DES des = new DESCryptoServiceProvider();

ICryptoTransform crypto = des.CreateDecryptor(key, iv);

MemoryStream cipherStream = new MemoryStream();

CryptoStream cryptoStream = new CryptoStream(cipherStream,

crypto, CryptoStreamMode.Write);

cryptoStream.Write(message, 0, message.Length);

DES des = new DESCryptoServiceProvider();

ICryptoTransform crypto = des.CreateEncryptor();

MemoryStream cipherStream = new MemoryStream();

CryptoStream cryptoStream = new CryptoStream(cipherStream,

crypto, CryptoStreamMode.Write);

cryptoStream.Write(message, 0, message.Length);

DES des = new DESCryptoServiceProvider();

ICryptoTransform crypto = des.CreateEncryptor(key, iv);

MemoryStream cipherStream = new MemoryStream();

CryptoStream cryptoStream = new CryptoStream(cipherStream,

crypto, CryptoStreamMode.Write);

cryptoStream.Write(message, 0, message.Length);

30)

You are developing a utility screen for a new client application. The utility screen displays a thermometer that conveys the current status of processes being carried out by the application.

You need to draw a rectangle on the screen to serve as the background of the thermometer as shown in the exhibit. The rectangle must be filled with gradient shading.

Which code segment should you choose?

Exhibit:

Rectangle rectangle = new Rectangle(10, 10, 450, 25);

LinearGradientBrush rectangleBrush = new LinearGradientBrush(rectangle, Color.AliceBlue, Color.CornflowerBlue, LinearGradientMode.ForwardDiagonal);

Pen rectanglePen = new Pen(rectangleBrush);

Graphics g = this.CreateGraphics();

g.DrawRectangle(rectanglePen, rectangle);

Rectangle rectangle = new Rectangle(10, 10, 450, 25);

LinearGradientBrush rectangleBrush = new LinearGradientBrush(rectangle, Color.AliceBlue, Color.CornflowerBlue,

LinearGradientMode.ForwardDiagonal);

Pen rectanglePen = new Pen(rectangleBrush);

Graphics g = this.CreateGraphics();

g.FillRectangle(rectangleBrush, rectangle);

RectangleF rectangle = new RectangleF(10f, 10f, 450f, 25f);

Point[] points = new Point[] {new Point(0, 0), new Point(110, 145)};

LinearGradientBrush rectangelBrush = new LinearGradientBrush(rectangle, Color.AliceBlue, Color.CornflowerBlue, LinearGradientMode.ForwardDiagonal);

Pen rectanglePen = new Pen(rectangleBrush);

Graphics g = this.CreateGraphics();

g.DrawPolygon(rectanglePen, points);

RectangleF rectangle = new RectangleF(10f, 10f, 450f, 25f);

SolidBrush rectangleBrush = new SolidBrush(Color.AliceBlue);

Pen rectanglePen = new Pen(rectangleBrush);

Graphics g = this.CreateGraphics();

g.DrawRectangle(rectangleBrush, rectangle);

31)

You write the following code to implement the BraindumpsClass.MyMethod function.

public class BraindumpsClass {

public int MyMethod(int arg) {

return arg;

}}

You need to call the BraindumpsClass.MyMethod function dynamically from an unrelated class in your assembly. Which code segment should you use?

BraindumpsClass myClass = new BraindumpsClass();

Type t = BraindumpsClass::typeid;

MethodInfom = t.GetMethod("MyMethod");

int i = (int)m.Invoke(this, new array {1});

BraindumpsClass myClass = new BraindumpsClass();

Type t = BraindumpsClass::typeid;

MethodInfom = t.GetMethod("MyMethod");

int i = (int)m.Invoke(myClass, new array {1});

BraindumpsClass myClass = new BraindumpsClass();

Type t = BraindumpsClass::typeid;

MethodInfom = t.GetMethod("BraindumpsClass.MyMethod");

int i = (int)m.Invoke(myClass, new array {1});

Type t =Type::GetType("BraindumpsClass");

MethodInfom = t.GetMethod("MyMethod");

int i = (int)m.Invoke(this, new array {1});

32)

You are creating an application that lists processes on remote computers. The application requires a method that performs the following tasks: Accept the remote computer name as a string parameter named strComputer.Return an ArrayList object that contains the names of all processes that are running on that computer. You need to write a code segment that retrieves the name of each process that is running on the remote computer and adds the name to the ArrayList object. Which code segment should you use?

ArrayList al = new ArrayList();

array procs = Process::GetProcessesByName(StrComputer);

for each (Process proc in procs) {

al.Add(proc);

}

ArrayList al = new ArrayList();

array procs = Process::GetProcesses(StrComputer);

for each (Process proc in procs) {

al.Add(proc);

}

ArrayList al = new ArrayList();

array procs = Process::GetProcessesByName(StrComputer);

for each (Process proc in procs) {

al.Add(proc.ProcessName);

}

ArrayList al = new ArrayList();

array procs = Process::GetProcesses(StrComputer);

for each (Process proc in procs) {

al.Add(proc.ProcessName);

}

33)

You write the following custom exception class named CustomException.

public class CustomException : ApplicationException {

public static int COR_E_ARGUMENT = unchecked((int)0x80070057);

public CustomException(string msg) : base(msg) {

HResult = COR_E_ARGUMENT;

}}

You need to write a code segment that will use the CustomException class to immediately return control to the COM caller. You also need to ensure that the caller has access to the error code. Which code segment should you use?

return Marshal.GetExceptionForHR(CustomException.COR_E_ARGUMENT);

return CustomException.COR_E_ARGUMENT;

Marshal.ThrowExceptionForHR(CustomException.COR_E_ARGUMENT);

throw new CustomException("Argument is out of bounds");

BASICS OF SAS Certifications

2009-06-20T05:17:00.000-07:00

SAS, short for "Statistical Analysis System," is a widely-used computer package for data

management and analysis, written and licensed by the SAS Institute. It is available on a wide variety of computers including mainframes, minicomputers, workstations, and DOS (IBM-compatible) personal computers. This handout summarizes a few very basic features of SAS useful in elementary econometric work. SAS is only one of many statistical packages currently available. (See the article by MacKie- Mason listed in the references below for a comparison of some popular packages.)

SAS is especially good at data manipulation: reading data in a wide variety of formats, merging data sets in elaborate ways, modifying the data, and writing new data sets of virtually unlimited size. However, these capabilities are too complex to be documented in this brief handout. To find out more about SAS, consult the voluminous documentation published by the SAS Institute. The name "SAS" is a trademark of the SAS Institute.

1. SAS jobs

Every SAS job, or program, consists of a sequence of DATA steps and PROC steps. The DATA steps create and modify data sets, while the PROC steps display or analyze the data. A step ends when either another step begins or the end of the job is reached. For readability, most SAS programmers used indentation or skipped lines to divide steps from each other. Every SAS step consists of a sequence of statements, each ending in a semicolon (;). A statement can spill across several lines, if necessary, and several statements can appear on the same line. SAS ignores case. You can type a SAS program in lower case, upper case, or a combination. In the examples in this handout, for clarity, standard SAS words will be written all-upper case, while user-specified names

will be written in lower case. Note, however, that user-specified names must begin with a letter and consist of no more than eight characters. Italics will be used to indicate user-specified phrases\.

SAS jobs produce two kinds of output. The SAS log is an annotated list of all statements in the job. The annotations report good news (the size of data sets created or analyzed, the number of missing values, etc.) and bad news (syntax errors, problems in reading data, etc.). The SAS listing gives the results of SAS PROC steps (estimates, standard errors, etc.). On some computers the log and the listing are output separately, on other computers together. A few statements can be used anywhere in a SAS job. These include the following:

Example of SAS DATA step statement Meaning

TITLE1 'Great Statistical Results'; Prints the phrase "Great Statistical Results" as the first title

line at the top of SAS listing output for the current and

subsequent PROCs. Phrases can be up to 60 characters

long, but must not include a quote mark. To change the

title in mid-job, just include another TITLE1 statement.

TITLEn 'May Be the Result of Faked

Data';

Prints the phrase "May Be the Result of Faked Data" as the

nth title line at the top of SAS listing output for the current

and subsequent PROCs. Here n can be an integer from 1 to

10, indicating the row of the title. Changing the title for a

particular row deletes all titles beneath it (having higher n).

HANDOUT: BASICS OF SAS Page 2 of 14

Example of SAS DATA step statement Meaning

* Now compute average annual earnings; Any statement beginning with an asterisk (*) is treated as a

comment and ignored by SAS. Good programmers sprinkle

their programs liberally with comments, describing what

their SAS jobs are supposed to do.

ENDSAS; Stops SAS processing. Later statements are ignored. This

can be handy for debugging.

2. The DATA step

The DATA step begins with a DATA statement, usually followed immediately by statements specifying the source for the input data. The most common ways to specify input data use either the SET statement (for an existing SAS dataset created earlier in the same job) or the pair of statements INFILE and INPUT (for an external file residing on a disk drive). These are usually followed by assignment statements, which create new variables out of the old ones. Here are examples of statements used in the DATA step.

Example SAS DATA step statement Meaning

DATA mydata; Begins a DATA step which will create the SAS dataset

entitled "mydata."

INFILE mystuff;

INPUT myvar1 myvar2 myvar3;

Reads input data from external file "mystuff." (The location

of this external file must be described at the beginning of

the SAS job in a form which depends on the computer. For

example, MVS computers require that "mystuff" be

described in a DD statement in the job control language.)

The data in this file are arranged in three columns and will

be henceforth called "myvar1," "myvar2," and "myvar3."

SET myold; Reads input data from existing SAS dataset "myold"

created earlier in the same SAS job.

myx = myvar1 + myvar2 * myvar3; Creates a new variable to be called "myx," equal to the sum

of myvar1 and the product of myvar2 and myvar3. The

standard order of operation applies: multiplication and

division precede addition and subtraction. (Statements

creating new variables, or assigning new values to old

ones, are called "assignment statements." Spaces, shown

here for clarity, are not required in assignment statements.)

IF myx LE 23 THEN myy = myvar3; Creates a new variable to be called "myy," equal to myvar3

if myx is less than or equal to than 23. If myx is greater than

23 then myy is recorded as a missing value in the new

dataset. Note that in this example, the clause following

"THEN" is an assignment statement.

IF myvar2 GT 5 THEN myz =

SQRT(myvar2); ELSE myz = 64;

Creates a new variable to be called "myz," equal to the

square root of myvar2 if myvar2 is greater than 5, and 64

otherwise. The new variable will thus have no missing

values.

HANDOUT: BASICS OF SAS Page 3 of 14

Example SAS DATA step statement Meaning

IF myvar1 NE 17; Includes in the new dataset only those observations for

which myvar1 is not equal to 17. Other observations are

discarded. (This is the so-called "subsetting IF" statement,

whose function is very different than the "IF-THEN"

statement described above.)

Assignment statements are the most common statements in a DATA step. SAS permits the usual arithmetic operators to be used on the right-hand side (+ for addition, - for subtraction, * for multiplication, / for division) as well as ** for exponentiation. SAS also supplies a huge number of built-in functions. A few very useful ones are given below. The arguments of functions can be algebraic expressions, not just a single variable as shown here, except for the lag functions. SAS function Meaning

LOG(myvar) Natural logarithm (logarithm to base e = 2.71828... Produces missing value if

argument is nonpositive.

EXP(myvar) Exponential function. (Inverse of LOG.)

SQRT(myvar) Square root. Produces missing value if argument is negative.

LAG1(myvar) First lag: previous observation of myvar. (Do not use any lag function in an IF

or ELSE statement. Strange results may occur.)

LAG5(myvar) Fifth lag: observation of myvar, five observations back.

PROBNORM(myvar)

Cumulative standard normal probability distribution function:

So-called comparison operators are used in "IF-THEN" and "subsetting IF" statements. The most widely-used comparison operators in SAS are the following. All can be written either as two-letter mnemonics or as symbols, but the "not-equal" symbols should be avoided because they vary across computers (use "NE" instead). (For comparison purposes, any missing value is treated as being less than the smallest negative number that SAS can store.)

Two-letter

mnemonic

Symbol Meaning

EQ = Equal to.

NE ^= or ~= Not equal to.

GT > Greater than.

LT <>

HANDOUT: BASICS OF SAS Page 4 of 14

Two-letter

mnemonic

Symbol Meaning

GE >= Greater than or equal to.

LE <= Less than or equal to.

EXAMPLE: DATA grbgout;

SET grbgin;

* This step creates SAS file "grbgout" out of SAS file "grbgin,"

which contains data on earnings, age, schooling, and gender;

lwage = LOG(earnings);

* Have no data on work experience, so impute it using age and

years of schooling;

exper = age - school - 6;

* Create dummy (zero-one) numerical variable for gender,

called "gendum," using existing character variable "gender";

IF gender = 'F' THEN gendum = 1;

ELSE gendum = 0;

3. The PROC step

The PROC step begins with a PROC statement, taking the following general form:

PROC GEEWHIZZ DATA=mydata (WHERE=(condition)) options;

The options are, of course, optional, and vary with the particular PROC used. The WHERE expression

in parentheses (one of SAS's many "dataset options") modifies the "DATA=" and is not required either. It serves to limit the observations included in the analysis to those which satisfy the condition, which might be take a form like "myvar1 LT 4" or "myvar2 EQ myvar3." The "DATA=" expression indicates which

SAS dataset is to be used. It is required if a dataset option such as WHERE is used. If no "DATA="

expression is used, then the PROC uses the SAS dataset created most recently.

Of course, "GEEWHIZZ" is not a real SAS PROC. Some useful real PROCs are described below.

4. Some PROCs for displaying data

4.1 PROC PRINT prints all the data in the dataset. If a statement like

VAR myvar1 myvar2;

is included in the PROC, then only the variables myvar1 and myvar2 are printed.

EXAMPLE: PROC PRINT DATA=mydata;

TITLE1 'Look At These Numbers!';

VAR wage school exper gender;

HANDOUT: BASICS OF SAS Page 5 of 14

4.2 PROC MEANS computes and prints descriptive statistics (means, variances, minima, maxima, etc.)

for all variables in the dataset. Just as with PROC PRINT, inclusion of a VAR statement limits the computation to the specified variables. This PROC takes very little computer time, and it is a good idea to include it before any more elaborate statistical analysis is performed.

4.3 PROC PLOT produces crude scatter plots on a line printer. It requires at least one PLOT statement (not to be confused with PROC PLOT, which begins the PROC) of the following form:

PLOT mvert*myhor;

where myvert is the variable plotted on the vertical axis and myhor is the variable on the horizontal axis.

Many plots can be requested in a single PLOT statement, as shown by the following examples:

PLOT myvert1*myhor1 myvert2*myhor2 myvert3*myhor3;

PLOT myvert1*(myhor1 myhor2);

PLOT (myvert1 myvert2)*(myhor1 myhor2);

The first statement requests three plots with three different pairs of variables. The second statement

requests two plots, both having myvert1 on the vertical axis. The third statement requests four plots, using alternately myvert1 and myvert2 on the vertical axis, and myhor1 and myhor2 on the horizontal axis.

EXAMPLE: PROC PLOT DATA=mydata;

TITLE1 'Scatter Plot of Earnings Against Age';

TITLE2 'To Give You Something to Look Forward To';

PLOT earn*age;

5. Some PROCs for estimating linear and qualitative models

Many different PROCs exist for estimating linear and qualitative econometric models. Most require at least one model statement of the following form:

mylabel: MODEL mydepvar = myvar1 myvar2 myvar3 / options;

The label clause ("mylabel:") is optional but useful if more than one model is to be estimated. The label appears at the top of the results listing for that model. The variable to the left of the equal sign is the explained variable, while the variables to the right are the explanatory variables. A constant term is present by default unless excluded by the option "NOINT." The entire estimated variance-covariance matrix of the coefficient estimates is printed if the option "COVB" is present. Many other options are available,

specific to the particular PROC. More than one model statement can be included in the same PROC if more than one model is to be estimated on the same data.

Several other optional statements, common to the PROCs described in this section, are shown in the following table.

Example SAS PROC step statement Meaning

WEIGHT mywtvar; Specifies a variable "mywtvar" in the data set

used to weight each observation. For weighted

least-squares, the value of mywtvar should be

equal or proportional to the reciprocal of the

variance of the error term. (This statement is not

available in PROC AUTOREG.)

OUTPUT OUT=newfile P=myfitted R=myresid; Outputs a new SAS dataset called "newfile,"

having all the original data, plus the predicted or

fitted values (given the name myfitted here) and

the residuals (given the name myresid here).

Multiple OUTPUT statements are permitted.

Each applies only to the preceding MODEL

statement. If the "OUT=" clause is omitted, the

output dataset replaces and takes the name of

the input dataset. (For PROC SYSLIN, the

"OUT=" clause appears as an option on the

PROC statement, not here.) The "P=" and "R="

clauses are optional. (The "R=" clause is not

available in PROC PROBIT.)

TEST myvar2=5; Computes an F-statistic for the hypothesis that

the coefficient of myvar2 equals five. (This

statement is not available in PROC AUTOREG or

PROC PROBIT.)

TEST myvar2=2*myvar3,

3*INTERCEPT+myvar3=0;

Computes an F-statistic for the joint hypothesis

that both: (1) the coefficient of myvar2 equals

twice the coefficient of myvar3, and (2) 3 times

the intercept, plus the coefficient of myvar3,

equals zero. (This statement is not available in

PROC AUTOREG or PROC PROBIT.)

5.1 PROC REG computes ordinary least squares regression. Useful options on the MODEL statement

for this PROC are shown in the following table.

MODEL

statement option

for PROC REG

Meaning

CLI Prints predicted values and 95% confidence intervals for individual prediction for all

observations in the sample with nonmissing explanatory variables.

CLM Prints predicted values and 95% confidence intervals for mean prediction for all

observations in the sample with nonmissing explanatory variables.

DW Prints Durbin-Watson statistic.

ACOV Prints White's alternative estimate of the variance-covariance matrix for the coefficients

which is consistent in the presence of heteroskedasticity. The square roots of the

diagonal elements are White's alternative standard errors.

SPEC Prints White's test statistic for heteroskedasticity.

EXAMPLE: PROC REG DATA=ohiodata (WHERE=(AGE>25));

HANDOUT: BASICS OF SAS Page 7 of 14

TITLE1 'Standard Human Capital Equation';

eqwage: MODEL lwage = school exper /SPEC;

* Suspect heteroskedasticity here;

5.2 PROC GLM is similar to PROC REG, but has some different capabilities, such as the ability tohandle discrete regressors represented as character (non-numeric) data, which are automatically interpreted as dummy variables. However, PROC GLM allows only one MODEL statement, so in general PROC REG is preferable for econometric work.

5.3 PROC AUTOREG computes linear regression corrected for serial correlation. Useful options on

the MODEL statement for this PROC are shown in the following table.

MODEL statement

option for PROC

AUTOREG

Meaning

METHOD=ML Uses the maximum-likelihood method.

METHOD=ULS Uses the "unconditional least squares" method, that is, nonlinear least squares.

METHOD=YW Uses the Yule-Walker method, a generalization of the Prais-Winsten method for

autocorrelation of arbitrary order. If no METHOD option is present, the Yule-Walker

method is used by default.

NLAG=n Specifies the order of the autocorrelation to be correct for. If n=1 the error term is

assumed to be AR(1), while if n=2 the error term is assumed to be AR(2), etc. If this

option is not present, then ordinary least squares estimates are computed.

ITER Specifies that the Yule-Walker estimates are to be iterated until convergence. If this

option is not present, then no iteration takes place and two-step estimates are

computed. This option is ignored if METHOD=ML or METHOD=ULS is specified.

5.4 PROC SYSLIN computes simultaneous-equation estimates using all equations together. The method to be used is specified as an option on the PROC statement. Some useful options on the PROC statement are shown in the following table.

PROC statement option for PROC

SYSLIN

Meaning

OUT=newfile Outputs a new SAS dataset called "newfile," having all the original data, plus any

variables (such as residuals or predicted values) specified in OUTPUT statements later

in the PROC.

2SLS Uses the two-stage least-squares method.

3SLS Uses the three-stage least-squares method.

LIML Uses the limited-information maximum-likelihood method.

FIML Use the full-information maximum-likelihood method.

HANDOUT: BASICS OF SAS Page 8 of 14

PROC statement

option for PROC

SYSLIN

Meaning

SUR Use Zellner's "seemingly unrelated regressions" method, also called joint generalized

least squares.

FIRST Prints the first-stage regression results for 2SLS or 3SLS.

If no estimation method is selected, this PROC will produce ordinary least squares estimates. If 2SLS,

3SLS, LIML or FIML estimates are desired then two additional statements must be present:

Statements for PROC SYSLIN Meaning

ENDOGENOUS myend1

myend2;

Lists the endogenous regressors in the model, i.e. the variables which

appear on the left-hand side in the first stage.

INSTRUMENTS myins1

myins2;

Lists all the predetermined variables in the model, i.e. the variables

which appear on the right-hand side in the first stage.

EXAMPLE: The following statements compute two-stage least-squares estimates for the regression of q

on p and y (a demand curve), where p is endogenous, and w and r are additional instruments (factor prices

from the supply curve):

PROC SYSLIN 2SLS;

TITLE1 'Estimate of Demand Curve";

demandeq: MODEL q = p y;

ENDOGENOUS p;

INSTRUMENTS y w r;

5.5 PROC PROBIT estimates a probit and/or logit model. Three types of statements are most commonly used in this PROC: the CLASS statement, the MODEL statement, and the OUTPUT statement.

The CLASS statement flags categorical data. Like PROC GLM, PROC PROBIT can handle character as well as numeric data. All character variables and the dependent variable (if it is a dummy [zero-one] variable) must first be listed in a CLASS statement.

The MODEL statement, as usual, lists the dependent and independent variables. With micro-data, the dependent data is typically a dummy variable, so it should have already been listed in the preceding CLASS statement. However, PROC PROBIT can also handle aggregate data, where one (integer) variable gives the number of successes and another (integer) variable gives the number of trials. In this case both variables are listed on the left-hand-side of the equal sign, separated by a "/". Useful options on the

MODEL statement for this PROC are shown in the following table.

HANDOUT: BASICS OF SAS Page 9 of 14

MODEL

statement option

for PROC

PROBIT

Meaning

D=LOGISTIC Estimates a logit model.

D=NORMAL Estimates a probit model. If no "D=" option is present, a probit model is estimated by

default.

The OUTPUT statement as usual lists the results to be saved and their destination. Three

commonly used options are given below. Note that the "P=" option saves the predicted probability, while

the XBETA saves the estimated coefficients times the independent variables.

OUTPUT

statement option

for PROC

PROBIT

Meaning

OUT=newfile Outputs a new SAS dataset called "newfile."

P=probs Includes predicted probabilities in output file as a new variable called "probs."

Warning: if the dependent variable is a dummy (zero-one) variable, the predictions are

for the probability that it equals zero!

XBETA=z Includes the inner product of the estimated coefficients and the independent variables

in the output file as a new variable called "z."

It should be noted that if the dependent variable is a dummy (zero-one) variable, the PROC PROBIT

coefficient estimates correspond to the probability that the dependent variable equals zero. That is, the

signs are reversed from usual econometric convention.

EXAMPLE 1: Micro-data on individual workers, dummy dependent variable.

PROC PROBIT DATA=people;

TITLE1 'Probit & Logit Models of Labor Force Participation: Micro-Data';

CLASS working;

* Note: the variable 'working' = 1 if working, = 0 otherwise;

labor1: working = school health income /D=NORMAL;

OUTPUT OUT=myfile P=ppredict XBETA=zprobit;

labor2: working = school health income /D=LOGISTIC;

OUTPUT OUT=myfile P=lpredict XBETA=zlogit;

EXAMPLE 2: Aggregate data, dependent variables show number of success and number of trials in each

observation (here a state).

PROC PROBIT DATA=states;

TITLE1 'Probit Model of Cancer Deaths: State-Level Data';

cancer: deaths/pop = avginc avgage povrate/D=NORMAL;

* Note: The variables 'deaths' and 'population' are state totals;

* The variables 'avginc,' 'avgage,' and 'povrate' are state averages;

HANDOUT: BASICS OF SAS Page 10 of 14

6. PROC NLIN for nonlinear least squares

PROC NLIN estimates models which are nonlinear in their parameters, using nonlinear least squares.

6.1 Specifying the equation to be estimated: A PROC NLIN step looks very different from a PROC for estimating linear models, like PROC REG. For one thing, the MODEL statement does not merely list the variables. It specifies the model explicitly. The general form of the model statement is:

MODEL mydepvar = expression;

Obviously, explicit expressions (not just lists of regressors) are necessary in PROC NLIN, since the functional form cannot be assumed to be linear, as in PROC REG. However, in reading these expressions SAS must distinguish somehow between parameters to be estimated and data variables to be read from the DATA set. To that end, SAS requires a PARAMETERS statement to name the parameters, whose typical form is illustrated as follows.

PARAMETERS myparm1 myparm2 myparm3;

Sometimes MODELS can be quite complicated to write explicitly. PROC NLIN therefore permits assignment statements before the PARAMETERS and MODEL statements to define intermediate or temporary variables. Both assignment statements and MODEL statements can use almost any SAS operator or function which is permissible in a DATA step (except lag functions).

6.2 Specifying the computational algorithm: Yet another difference between PROC NLIN and PROC REG reflects the fact finding the estimates is more difficult. Estimates for nonlinear models are solutions to systems of nonlinear equations. These estimates cannot be computed in one step. Instead, they must be found through some iterative technique that gets closer and closer to the exact solution with each iteration. An iteration rule is called an algorithm. The algorithm used for computing the estimates is

specified by an option on the PROC NLIN statement.

PROC statement option for

PROC NLIN

Meaning

METHOD=GAUSS Uses Gauss-Newton algorithm. If no method is specified, this one is

used by default.

METHOD=MARQUARDT Uses Marquardt algorithm.

METHOD=NEWTON Uses Newton-Raphson algorithm.

METHOD=GRADIENT Uses "Steepest descent" algorithm.

METHOD=DUD Uses the "doesn't use derivatives" algorithm of Ralston and Jennrich.

Does not require "DER." statements (see below). Instead, computes

estimates of derivatives from previous iterations.

Most computational algorithms use the partial derivatives of the objective function (the sum of squared residuals) with respect to the parameters to find the estimates. (Methods using or approximating these partial derivatives are collectively called "gradient algorithms" and generally converge much faster to

(1)

(2)

the estimates than alternative methods.) Although some statistical packages can differentiate the objective function automatically, SAS needs a little help. In particular, SAS requires the user to supply partial derivatives of the model as "DER." statements, whose general form is as follows.

DER.myparm = expression;

There must be one "DER." statement for each parameter to be estimated.

PROC NLIN also permits OUTPUT statements similar to those used in PROC REG."

EXAMPLE: Suppose we wish to estimate the following model by the Newton-Raphson method:

where yt and xt are data and $ is an unknown parameter. The derivative of the model with respect to $

is:

(check this). To estimate this model in PROC NLIN we would use the following statements:

PROC NLIN DATA=grbgin METHOD=NEWTON;

TITLE1 'Estimates of FNLM';

TITLE2 '(FNLM = funny non-linear model)';

PARAMETERS beta;

MODEL y = beta**x;

DER.beta = x * beta**(x-1);

OUTPUT OUT=grbgout P=yhat R=uhat;

Alternatively, we could use an assignment statement so that exponentiation is performed only once, saving a little computation time:

PROC SYSLIN DATA=gooddata METHOD=NEWTON;

TITLE1 'Estimates of FNLM';

TITLE2 '(FNLM = funny non-linear model)';

PARAMETERS beta;

btothex = beta**x;

MODEL y = btothex;

DER.beta = x * btothex/beta;

OUTPUT OUT=grbgout P=yhat R=uhat;

7. PROC ARIMA for Box-Jenkins time series analysis

Box and Jenkins (1976) propose three steps for analysis of data thought to be generated by an

ARIMA (auto-regressive integrated moving average) statistical process, namely

HANDOUT: BASICS OF SAS Page 12 of 14

(1) identification (also called specification)

(2) estimation

(3) forecasting

Corresponding to these three steps in the Box-Jenkins methodology are three statements used in PROC

ARIMA.

7.1 Identification: The IDENTIFY statement computes and plots the sample autocovariances, inverse

autocovariances, and partial autocovariances for a series. Typical examples of the IDENTIFY statement

are:

IDENTIFY VAR=myvar;

IDENTIFY VAR=myvar(1);

IDENTIFY VAR=myvar(1,1);

The first statement analyzes the raw data. The second and third statements analyze the first-differenced data and the second-differenced data, respectively. (The "1"s could be replaced by higher integers if seasonal factors were thought to be present.) Typically, an initial computer run using PROC ARIMA includes several IDENTIFY statements using various degrees of differencing to determine the level of differencing required to produce a stationary series. The estimates and plots from the stationary series are

then used to make guesses as to the number of autoregressive parameters (p) and the number of movingaverage parameters (q).

7.2 Estimation: The ESTIMATE statement then computes estimates of a particular model. Typical

examples of the ESTIMATE statement are:

ESTIMATE P=1 PLOT;

ESTIMATE Q=2 PLOT;

ESTIMATE P=2 Q=1 PLOT MAXIT=25;

The first statement estimates an AR(1) model. The second statement estimates an MA(2) model. The third statement estimates an ARMA(2,1) model. The PLOT option is not required but recommended. It causes plots of the autocorrelations and partial autocorrelations of the residuals to be plotted. These should be close to white noise for a good model. The MAXIT option is also optional, but recommended for models

of moderate order (p+q>2). It permits numerical iterations beyond the default number of 15. The position of the ESTIMATE statement determines the level of differencing. The level of differencing used is simply that of the last preceding IDENTIFY statement. Typically, the second computer run using PROC ARIMA inserts ESTIMATE statements after that IDENTIFY statement which specifies the minimum level of differencing apparently required to produce a stationary series. In addition to parameter estimates, the ESTIMATE statement computes standard errors, tstatistics,

and goodness-of-fit criteria that can help determine the best model.

7.3 Forecasting: The FORECAST statement computes forecasts of the series for as many periods into

the future as desired. Typical examples of the FORECAST statement are:

FORECAST;

HANDOUT: BASICS OF SAS Page 13 of 14

FORECAST LEAD=12;

The first statement forecasts the default number of periods (24). The second statement forecasts 12 periods. The output always includes within-sample forecasts as well.

The position of the FORECAST statement determines which model is to be used. The model used is simply that of the last preceding ESTIMATE statement. Thus a forecast cannot be produced without all three statements: IDENTIFY, ESTIMATE, and FORECAST. Typically, the third computer run using PROC ARIMA inserts FORECAST statements after the ESTIMATE statements which specify the most

promising model. EXAMPLE: The following example computes autocorrelations and partial autocorrelations for the S&P 500 index of stock prices, estimates AR(1), MA(1), and ARMA(1,1) models for the first-differenced data, and forecasts on the basis of each model. Indentation is used to emphasize the hierarchy of IDENTIFY,

ESTIMATE, and FORECAST statements.

PROC ARIMA DATA=indices;

TITLE1 'Forecasting Project to Get Rich Quick';

IDENTIFY SP500;

IDENTIFY SP500 (1);

ESTIMATE P=1 PLOT;

FORECAST LEAD=6;

ESTIMATE Q=1 PLOT;

FORECAST LEAD=6;

ESTIMATE P=1 Q=1 PLOT;

FORECAST LEAD=6;

IDENTIFY SP500 (1,1);

Oracle certifications

2008-08-05T03:44:00.000-07:00

Oracle Certification Program Candidate Guide

Benefits of Oracle certifications

The demand for professionals in the information technology (IT) industry is high, and the competition for jobs is intense. Individuals, experienced or new to the profession, need to know what skills make them attractive to employers. Employers look for ways to select prospective employees who have the solid foundation of skills needed for effective performance.

The Oracle Certification Program helps the IT industry establish a standard of competence in key entry-level and professional job roles.

An Oracle Certification is a valuable, industry-recognized credential that signifies a proven level of knowledge and ability. Each higher level of Oracle certification brings a higher standard of benchmarked skill and ability, which leads to greater opportunities and higher pay.

"Technical certifications have evolved from a hiring tool to a screening tool: If you don’t have them, you aren’t viewed as a serious candidate." ¹

Benefits to the Technical Professional

An Oracle Certified Associate (OCA) demonstrates a solid understanding of the foundation skills of a given job role, which can be applied at an apprentice or entry level.

By earning an OCA certification, you can have increased entry-level job opportunities. The OCA is the stepping-stone to starting a successful career as an Oracle professional.

Beyond OCA, by becoming an Oracle Certified Professional (OCP) you demonstrate your understanding of the full range of skills required by Oracle professionals in your chosen job role. An OCP is in high demand in today's marketplace, and the level of demand is expected to grow with each new installation of Oracle technologies around the world. An Oracle Certification helps raise your visibility and increases your access to the industry's most challenging opportunities.

"Oracle’s certification strength in a declining economy is due in part to the fact tha more complex, high-level certifications appear to be less vulnerable" ²t

An Oracle Certified Master (OCM) certification is the highest credential that an Oracle professional can earn in the Oracle Certification Program. An Oracle DBA Certified Master is a top-level Oracle expert who has in-depth technical knowledge of the Oracle data server and industry experience in managing mission-critical Oracle database systems and applications. OCMs are looked upon as experts within their organizations, among their peers, and across the industry.

The true value of earning an Oracle Certification credential is increased opportunity. With more opportunity come career growth and higher pay.

"Given the key importance of data as an organizational asset, it should come to no surprise that DBAs remain in high demand and that related training certification programs are popular even in this time of economic trouble... By the numbers, Oracle leads the DBMS/DBA certification area.” ³

Benefits to the IT Employer

The Oracle Certification Program is also valuable to hiring managers who want to distinguish among candidates for critical IT positions. For companies that send employees for annual IT training, certification helps ensure a return on the training investment by validating the knowledge and understanding gained during training sessions. Companies can also combine certification with an employee development program to enhance employee loyalty and performance on the job. Hiring certified professionals can have a direct impact on a company's success.

1 Source: Employers Raise the Bar on Certification, Information Week, 2002

2 Source: IDC Certification Report and Forecast 2002–2006

3 Source: Certified Expert: Working as a Database Administrator, CertMag, January 2003

Oracle9i DBA Certification Overview

The expertise of Oracle database administrators (DBAs) is integral to the success of today's increasingly complex system environments. Their critical work requires a broad understanding of the architecture and processes of the Oracle database, as well as plenty of hands-on experience in solving problems. The best DBAs know that they stand between optimal performance and an event that could bring their company to a standstill.

“Currently, Oracle skills remain the most sought-after database skills in the IT marketplace.” ¹

Oracle9i Database Administrator Certified Associate (OCA)

An Oracle9i Database Administrator Certified Associate (OCA) certification is an entry-level credential for candidates who have a proven foundation of basic knowledge, which they can build upon as they work toward a career as an Oracle database administrator. An OCA can next earn Oracle9i DBA Certified Professional (OCP) and Oracle9i DBA Certified Master (OCM) credentials.

Oracle9i Database Administrator Certified Professional (OCP)

By moving upward and earning OCP status, Oracle professionals demonstrate the complete set of skills that are required for working independently as an expert DBA. The OCP credential provides candidates increased opportunities and higher pay. The OCP DBA path requires four tests two for OCA certification plus two more. Additionally, Oracle now requires (as of September 1, 2002) all newly starting OCP candidates to complete one Oracle University hands-on course in preparation for their exams and for future roles as an Oracle technologist. An OCP certification helps you translate your knowledge and skills into increased visibility through the market's most highly valued database benchmark program.

View the chart on page 5 for a list of required tests and course requirements to earn the Oracle9i Database OCP credential.

Oracle9i DBA Certified Master (OCM)

Beyond OCP, highly experienced DBAs who are ready for a new challenge can earn the highly respected Oracle9i DBA Certified Master (OCM) credential, which can help their careers develop further and open doors to senior DBA and consulting opportunities. The Oracle9i DBA Certified Master Program requires Oracle University coursework on advanced DBA topics and hands-on demonstration of DBA skills in a live application environment. For more information about this offering, consult the Oracle Certification Program Web site.

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 3

Upgrading Your Current OCP Credential to Oracle9i

Single Release Upgrade Exams

Generally, one exam is required for OCPs to upgrade to each new release of an Oracle DBA track. For example, one exam upgrades the Oracle7.3 credential to Oracle8, one exam from Oracle8 to Oracle8i, one exam from Oracle8i to Oracle9i and one exam from Oracle9i to Oracle 10g.

Oracle7.3 to Oracle8 DBA OCP

Exam #1Z0-010 Oracle8: New Features for Administrators.

Oracle8 to Oracle8i DBA OCP

Exam #1Z0-020 Oracle8i: New Features for Administrators.

Oracle8i to Oracle9i DBA OCP

Exam #1Z0-030 Oracle9i Database: New Features for Administrators.

Bridged Release Upgrade Exams

OCPs with older credentials may skip intervening releases and upgrade to a current technology with a single exam. For example, one exam upgrades Oracle7.3 and Oracle8 DBAs directly to Oracle9i, and one exam upgrades Oracle8i DBAs directly to Oracle Database 10g.

Oracle7.3 and Oracle8 DBA OCPs to Oracle9i

Exam #1Z0-035 Oracle9i DBA: New Features for Oracle7.3 and Oracle8 OCPs.

Candidate Qualifications

Most candidates for DBA certification combine up-to-date training with some level of on-the-job experience. There is no "typical" candidate. Many of the questions on the OCP tests are based on real job scenarios. In addition to the appropriate training, you need hands-on experience with the software. Trial software versions and hands-on labs are included in most Oracle University training products.

1 Source: IDC Certification Report 2002

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 4

Required Exams and Selected Hands-On Courses for Oracle9i Database Certification Paths

Oracle customers and business partners demand hands-on experience—with all aspects of Oracle database—from their Oracle Certified Professionals. In order to meet our commitments to our customers and constituents, Oracle University has recently made a significant investment toward building more validity and quality into its Certification Program including scenario-based exam questions.

Exams Required

Oracle9i Database Administrator Certified Associate Oracle9i Database Administrator Certified Professional

Oracle9i Database: Fundamentals I

Exam # 1Z0-031

Oracle9iDatabase Administrator Certified Associate

Oracle9iDatabase: Fundamentals II

Exam # 1Z0-032

Introduction to Oracle: SQL^®

Exam # 1Z0-007

Introduction to Oracle: SQL^®and PL/SQL^™Exam # 1Z0-001

Exam # 1Z0-007

Introduction to Oracle9i: SQL can be taken over the Internet; (see Section 5 of this guide).

Oracle9i Database: Performance Tuning

Exam # 1Z0-033

Oracle9iDatabase Administrator Certified Professional

Oracle University Hands-On Course Requirement

To earn your Oracle9i Certified Professional credential, you must pass the exams listed above and complete one of the selected Oracle University courses listed below.

Note the additional courses that have been added to the list of those that fulfil the Oracle9i DBA OCP Hands-On Course Requirement. Beginning June 21, 2004 candidates may now attend any one of the courses below to meet the requirement.

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 5

Current OCP Candidates Tested BEFORE September 1, 2002

OCP candidates who have already completed at least one exam toward earning their OCP credential before September 1, 2002, will be permitted to earn their OCP credential without having to complete one of the hands-on courses listed above.

New OCP Candidates Tested AFTER September 1, 2002

All new OCP candidates will continue to be permitted to enroll in the required exams before fulfilling the hands-on course requirement, but will not be certified until completing both the exams and course requirement.

Note that candidates must first complete the OCA certification before obtaining an OCP credential.

OAI, OIA, and Workforce Development Students

Note that students completing their Oracle study with an Oracle Academic Initiative (OAI) program, Oracle Internet Academy (OIA), or an Oracle Workforce Development school do NOT need to complete one of the specific courses listed on the previous page, since these students are already completing this curriculum with their program of study.

Certified OCP DBAs

Currently certified OCP DBAs are not required to complete any additional hands-on courses; they are required only to pass the upgrade exam.

The courses listed on the previous page are offered at Oracle University centers around the world as well as at many Oracle Authorized Education Centers, and Oracle Authorized Education Partners. Instructor-led courses are also now available online, enabling candidates to complete this requirement from their home or office if travel or time is a constraint.

To find more information about these courses, or to enroll, visit the Oracle University Web site at http://www.oracle.com/education/ and select the Oracle9i DBA Learning Path.

* This title is not available online.

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 6

Oracle9i Database Upgrade Certification Paths

Follow the paths in the diagram below to upgrade your current Oracle DBA OCP credential.

Oracle9i Database Certified Master Certification Path

Follow the path in the diagram below to earn your Oracle9i DBA Certified Master credential.

Oracle7.3 Database Administrator Certified Professional

Oracle9i New Features for Administrators

Exam # 1Z0-030

Oracle9iDBA New Features for Oracle7.3 and Oracle8 OCPs

Exam # 1Z0-035

Oracle8i Database Administrator Certified Professional

Oracle8 New Features for Administrators

Exam # 1Z0-010

Oracle8 Database Administrator Certified Professional

Oracle8i New Features for Administrators

Exam # 1Z0-020

Oracle9iDatabase Administrator Certified Professional

Oracle9iDatabase Master Practicum (details are on page 11 of this guide)

Oracle9i Database Administrator Certified Professional

Attend two of the selected advanced OU courses listed on page 10 of this guide.

Oracle9iDatabase Administrator Certified Master

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 7

Oracle9i Database Administrator Certified Master Requirements

To earn your Oracle9i Database Administrator Certified Master credential, you must meet the following requirements.

Oracle9i DBA Certified Master (OCM)

An OCM certification is the highest credential that an Oracle professional can earn in the Oracle Certification Program. An Oracle9i Database Administrator Certified Master is a top-level Oracle expert who has in-depth technical knowledge of the Oracle data server and industry experience in managing mission-critical Oracle database systems and applications. OCMs are looked upon as experts within their organizations, among their peers, and across the industry.

Oracle9i Database OCM Practicum

To earn the Oracle9i Database Administrator Certified Master credential, candidates must pass the practicum exam.

The Oracle9i DBA Certified Master practicum is a two-day live application event where participants are required to complete a series of scenarios and resolve technical problems that will test their expertise in an Oracle9i Database environment. Participants are scored on their ability to successfully complete the assigned tasks.

The practicum is conducted at an Oracle University facility in each global region. Contact your local Oracle University representative to find the Practicum location nearest you. Visit http://www.oracle.com/education/contact_info.html to view your local Oracle University contact information.

Practicum Registration Prerequisites

Prior to registering for the Oracle Certified Master practicum, candidates must complete two requirements.

1. They must have completed their Oracle9i Database Administrator Certified Professional (OCP) credential.

2. They must have taken two advanced DBA courses from Oracle University. Advanced courses are taught by top Oracle University instructors and offer hands-on experience in topics that may be addressed during the practicum. Advanced courses that are offered in an instructor-led online format satisfy the requirement for candidacy to the OCM.

The following is a list of Oracle University courses that meet the advanced course requirement.

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 8

Oracle9i Database OCM Practicum Course Prerequisite

The OCM Practicum candidates must attend two of the following courses listed below.

January 2005 - Oracle Certification Program Candidate Guide: Oracle9i Database Administrator Certified Associate, Professional, and Master Paths 9

Oracle9i DBA Certified Master Practicum

An OCM candidate must successfully pass the Oracle9i DBA Certified Master practicum after completing the required OU advanced learning requirements listed on the previous page.

Minimum Practicum Participant Skills

1. Proficiency in Oracle9i SQL

2. Working knowledge of Linux command language (United Linux 1.0) including:

a. Formatting and executing basic OS commands

b. Creating directories and navigating through directory structures

c. File management using copy, move, and delete

d. Linux environment text editors

e. Setting environment variables

3. Ability to locate and launch Oracle executables that include:

a. RMAN utility

b. Oracle Net Manager

c. Oracle Net Configuration Assistant

d. OEM

e. Listener utility

f. OMS

g. Oracle Password utility

h. Database Creation Assistant

4. Proficiency in using Oracle Enterprise Manager

5. Proficiency in using Oracle Net Manager and Oracle Net Configuration Assistant to configure networking

6. Advanced knowledge and use of Oracle9i (v9.0.1.3) Enterprise Server technology and features

7. Familiarity with navigating through online Oracle documentation

8. Proficiency in using Konqueror 2.2 browser software

Practicum Environment

The practicum is conducted in a dedicated Oracle University classroom environment. An Oracle instructor is responsible for proctoring the event, distributing scenarios, and collecting results. Participants are randomly assigned a seat on the day of the event.

Participants are not allowed to use personal documentation or notes during the practicum nor are they permitted to collaborate with other participants. Cellular phones, pagers, and Personal Digital Assistants are not allowed in the practicum classroom.

Each participant is provided with a dedicated Linux server with Advanced Server 2.1 installed and the base Oracle server software. A complete online Oracle documentation set is available to each participant.

Scoring

Participants are scored on their ability to complete the assigned tasks and to recover from various database failure scenarios. Information from each participant's environment is collected at various points during the practicum, and this information is used to determine the final score. Participants will be provided with their score within two weeks of completing the practicum. Upon attaining a successful score, participants are awarded the Oracle9i Database Administrator Certified Master credential. Participants who do not achieve a passing score are required to retake the entire practicum at a later date.

Practicum Objectives and Areas of Focus

Review the Oracle9i Database Administrator Certified Master practicum objectives to gain insight into what to expect before completing the practicum. The objectives are located in the Test Content Checklist at the end of this guide.

Software Quality Assurance Certifications

2008-08-05T03:34:00.000-07:00

The functionality of Software Quality Assurance is to assure that the standards, processes, and procedures are appropriate for the project and are correctly implemented.

This is a process, which flows from software requirement Analysis, software design, coding, source code control, product testing, code reviews, change management, configuration management, and release management.

SQA comes in many forms, from the developer who tests the own code, and SQA testers work with automated testing tools and manual testing. Many vendors and groups have recognized Quality Assurance and Testing as an essential part of the development and maintenance process and have developed certifications to standardize and demonstrate knowledge of the QA process and testing tools.

SQA Certifications standardized by the Quality Assurance Institute (QAI) includes following levels of Certifications.

Certified Software Quality Analyst (CSQA),

Certified Software Tester (CSTE)

Certified Manager of Software Testing (CMST)

Certified Manager of Software Quality (CMSQ)

Certified Software Project Manager (CSPM)

Acquiring the above levels of certifications indicates a professional level of competence in the principles and practices of quality assurance in the IT profession and gains recognition as software quality profession, achieve more rapid career advancement, and gain greater acceptance in the role as advisor to management.

Discussing about new trends in software testing can be forced on testing tools and new methodologies in testing as the companies are trying hard to keep up a good demand by delivering reliable, bug-free, and quality software.

Continuous Integration (CI) is one such software testing trend that is quickly being adopted into the testing arena. It is a practice where the team members integrate their work with rest of the development team on a frequent basis. The development team and the testing team commit their changes to a central versioning system.

With CI the integration between the developed code and testing can be done as soon as the changes by developers get committed. This results in many integrations per day, as the development team will not have to wait for long to get feedbacks from the team of testers. CI would result in removing the code duplication, authenticity of dependencies, and coding standards.

Using the correct testing tool at the right time in a project can significantly increase the efficiency of testing by automating processes, increasing communication, promoting best practices and re-use of tests and test data.

The leading functional automated testing tools include:

Mecury: QTP and WinRunner

QuickTest Professional™ (QTP) and Mercury WinRunner® deliver a complete solution for functional test, GUI test, and regression test automation with support for practically every software application and environment.

Compuware: TestPartner and QARun

When testing complex applications based on Microsoft, Java and web-based technologies, TestPartner helps reassure that testing efforts are complete. For other application types, QARun gives developers and testers the automation capabilities needed to create and execute test scripts, verify tests and analyze test results.

IBM Rational Robot

IBM Rational Robot is a general purpose test automation tool for QA teams who want to perform functional testing of client/server applications.

Odin Technology - Axe

Axe is a new class of business process-oriented tools that allow non-technical users to automate testing. It provides a means to rapidly deploy automated testing systems that can be used by staff without specialist automation skills and minimal training. This reduces the cost of introducing and maintaining test automation by a factor of four. Axe can translate scripts to run with any of the functional testing tools mentioned above.

These are the leading automated testing tools for functional and regression testing. As well as functional regression testing tools there are also automated performance testing tools