Home > Articles > Programming

Like this article? We recommend

Complex Performance-Testing Scenarios

Now that you know what a performance-degradation curve is, the next thing you need to understand is the modeling of performance scenarios. We'll build on the work of others again and review a modeling tool and practice that's already available. In his article on the user community modeling language (UCML™), Scott Barber shows a method to visually depict complex workloads and performance scenarios. When applied to performance testing, UCML can "serve to represent the workload distributions, operational profiles, pivot tables, matrixes, and Markov chains that performance testers often employ to determine what activities are to be included in a test and with what frequency they'll occur." I've used UCML diagrams to help me plan my performance testing, to document the tests I executed, and to help elicit performance requirements.

Figure 3 shows a sample UCML diagram for an online bookstore. This example has four types of users: new and existing users of the site, web site administrators, and vendors (people who sell their books on the web site). Each user starts at the home page; from there, their paths vary depending on what functions they need to perform. Some functions are unique to certain roles, and others are shared between roles. The model shows us all the potential paths through the application, as well as expected percentages of users for any given role or for any given path through the model. The legend in Figure 4 may be helpful as you review the diagram.

Figure 3

Figure 3 Sample UCML diagram for an online bookstore.

Figure 4

Figure 4 UCML 1.1 symbols.

Models like this allow us to create realistic (or reasonably realistic) performance test scenarios. Using a model like this, we can create a series of performance tests to measure application performance in terms of a specific user type or an aggregate of all user types, overall response given a specific load, effects of load on any given type of user, etc. The power behind a modeling approach like this is that it's intuitive to developers, users, managers, and testers alike. That means faster communication, clearer requirements, and better tests.

The Problem

We've looked at two commonly used tools for performance testers: performance-degradation curves and performance-usage models. Like any popular tool, these tools are often misused. I've seen performance-degradation curves (and/or very similar tools) used to justify more performance testing than is probably necessary. Many times, a performance tester will receive a requirement stating that a web page should load within X time under Y load. Using a performance degradation curve, the performance tester can find the knee in performance; if it's below the specified threshold, the cycle of application fix/modification and repeat testing can go on indefinitely—often with the only goal being rightward movement of the knee in performance along the performance-degradation curve.

While this is certainly a part of performance testing, when taken out of the context of the expected usage of the application, these requirements can be misleading. For which users should the page load in X seconds? Which users are logged into the system to create a load of Y? Is this requirement for the month of December, when the bookstore's web site is under the highest load; or is it during July, when there is very low traffic; or is it based on average values? Simply moving the knee in performance is not enough. We need to better understand the context in which that knee exists. Perhaps that knee is acceptable for certain circumstances and unacceptable in others.

Instead of focusing on what's wrong with the application performance, if we focus on what's right we may be able to better determine where problems exist—if we even have problems. By focusing on performance plateaus and stress areas (where the application is performing well) instead of the knee in performance, we're able to better understand the larger picture of how the application operates. Let's look at an example.

Developing Scenarios and Degradation Curves

Let's look at how we might determine whether our web site fulfills a performance requirement. Suppose we have a requirement that the home page should load within six seconds under a 150-user load. Using our user community model above, we can create complex workloads to execute against the site. We come up with the performance-degradation curve shown in Figure 5.

Figure 5

Figure 5 Home page basic response-time degradation curve.

We can see in Figure 5 that we've fulfilled our basic requirement. The results for the test represented in our usage model show that we have a response time around 5.33 seconds for the web site's home page. This is good news. Unfortunately, this is where most performance testers stop. While it's meaningful information, it's not enough information. We still know very little about how the home page loads with a load of 150 users. We only know how it loads with those 150 users ("those users" referring to the user percentages and user types defined in the model).

If we've taken the time to develop a model like the one in Figure 5, and then developed our performance scripts to represent that model, most likely it will be very easy for us to modify those scripts to reflect workload changes in the model. For example, in many tools, if I swap my New User and Member percentages, it's a simple variable change to reflect that in my scenario. I can then run the test again and create the response-time degradation curve for that new scenario. It might look something like Figure 6.

Figure 6

Figure 6 Home page response-time degradation curve with a greater percentage of new users.

What we find with this scenario is that by increasing the number of new users, we're increasing the number of users who need to create an account before they can make a purchase. This has an overall effect of raising the home page response time by about 20% because of the increased load on the server, due to users remaining in session longer and the server having to process more transactions for each user.

This variance is a good example of the ways in which you may want to vary your workload models while testing. This variant works on the assumption that the model initially developed was for the application while it was in production, after it had been in production for some time. That model neglected the fact that when the application is initially released, there will be no existing members. All users will be new users. If we had gone to production based on the first test alone, we would have lost the business of the unfortunate new users who couldn't access the home page because our server was too busy creating accounts for other users.

This is an example of including business scenarios in your testing. Following are some other examples:

  • Seasonal sales variances. Fewer users search during the winter because they know what they're going to purchase. More users search in the summer because they're looking for a couple of new books to read over a break, but have no real preferences in mind when they go to the site.
  • New product releases. We decide to add CD sales to the web site. Not only will it shift how many administrators are on the site adding new product; it may also increase searching on the web site for the first few months in which the new products are available, and then over time the usage pattern simply matches that of books.
  • Advertising campaigns or a partnership with another vendor. Suppose we partner with Discover Card to offer a 10% discount to all Discover Card users on the site. While this may or may not affect load, it will most likely have a short-term effect on the usage pattern of existing members as they update their account information to switch their primary method of payment to Discover.

Many more examples are possible, and this is just for an eCommerce web site. None of the examples above necessarily affect load. That is, regardless of the scenario, we're still attempting to prove our 150-user requirement. Notice that I didn't include purchasing a Super Bowl commercial. The resulting 500,000-user load generated from that scenario is outside the requirement's scope.

The intent is to learn new information about the application so we can determine whether a) we have the correct requirement; b) a requirement has been fulfilled; and c) we need to perform other tests based on the new information uncovered with this testing. Through this type of testing, we can both confirm that the application really does what we think it does and learn information that may help us identify possible requirements omissions.

Correlating Performance Plateaus and Stress Areas for Each Scenario

So what do you do with all of these degradation curves once you've generated them? Let's look at correlating that data into some more useful format. Conceptually, we want to overlay all of this information so we can look at performance plateaus and stress areas side by side (see Figure 7).

Figure 7

Figure 7 Conceptual overlay of response-time degradation curves.

Remember that we're not interested in what's broken—well, we are, but that's not what we're focusing on right now. We're focusing on what works within the system. We want to know what functionality appears to be stable, regardless of how we (realistically) vary the user scenarios for any given user load. To do that, we need to look at performance plateaus (the best performance you can expect without further performance tuning) and stress areas (the max recommended user load at which the application "degrades gracefully").

By looking at these areas instead of the knee in performance, we shift our focus from "performance tuning for the sake of performance tuning" to one of "good enough" performance, so we can concentrate on improving quality in those areas that really need improvement. The basic idea works like this: Because we can't possibly test everything, we need to continuously compare the present quality of the product against the cost and value of further improvement. "Good enough" testing is the process of developing a sufficient assessment of the current quality of the application, and then making both wise and timely decisions concerning the cost and value of further improvement. (James Bach has given a much better and more detailed definition than my feeble attempt to paraphrase his work; see the References section at the end of this article.)

In terms of performance testing, this means assessing what works, not only what doesn't. By knowing our plateaus and stress areas (those areas where our application works), we gain an accurate assessment of the quality of the product. If we never do this and we focus only on the areas that are broken, it's much more difficult to know the value of more testing, because we don't have a full picture of where the software is right now. How can we predict how much better the application can be with 40 more hours of testing, if we don't know how good it is now?

Back to our example. We've correlated the data and come up with a chart similar to the one in Figure 8. (Note that there are several acceptable ways to correlate this data; this is just one that works well for this specific example.)

Figure 8

Figure 8 Correlated data for each scenario.

By analyzing this data, we can see that for three out of our five scenarios we didn't meet the requirement. That may be okay. It's not up to us to decide, we just report that information so project stakeholders can make informed decisions about the application's quality. What's more interesting is that by lining up all the scenario data we can see that four out of the five scenarios have the same basic performance plateaus and stress areas. The only exception is scenario 5. In this scenario, we enter our stress area right away at 25 users. A little more research may show that in this scenario our "vendor" users (6% of the total load based on the performance model above) run a specific type of report. This may be a problem that someone thinks we should fix, or it may be something we decide to go to production with because we think few vendors will run that specific report.

Let's look at another example. Figure 9 shows response-time degradation curves for the time needed to execute a search for our predicted usage patterns in each yearly quarter.

Figure 9

Figure 9 Search response-time degradation curves for each yearly quarter.

Q1, Q2, and Q3 are all fairly similar, as usage patterns remain about the same. That is, any given 100 users will do about the same thing regardless of the quarter, as long as it's not Q4. According to our usage models, users are doing much more searching in Q4. That is, for any given 100 users, the probability rises of those users executing several more searches than they did in the previous quarters. By correlating our data, we find that this has an effect on search response time for that quarter.

If a requirement says we need a year-round search response time of less than three seconds at a 150-user load, by using this information we can actually figure out what those users are doing differently. We can go back to the model, make changes, rerun the tests, re-correlate the data, and by using both tools attempt to determine where the problem is. Or we can tune our servers and make application changes and then rerun and re-correlate the tests. If we don't correlate the plateaus and stress areas, all we know is that if we make changes the response time may decrease. If we don't correlate this data, we don't know if it went down for all quarters, down for Q4 and up for the others, or if we're sliding the entire curve(s) left or right. Correlating the data gives us the bigger picture of the effect(s) of our changes on the entire system.

InformIT Promotional Mailings & Special Offers

I would like to receive exclusive offers and hear about products from InformIT and its family of brands. I can unsubscribe at any time.

Overview


Pearson Education, Inc., 221 River Street, Hoboken, New Jersey 07030, (Pearson) presents this site to provide information about products and services that can be purchased through this site.

This privacy notice provides an overview of our commitment to privacy and describes how we collect, protect, use and share personal information collected through this site. Please note that other Pearson websites and online products and services have their own separate privacy policies.

Collection and Use of Information


To conduct business and deliver products and services, Pearson collects and uses personal information in several ways in connection with this site, including:

Questions and Inquiries

For inquiries and questions, we collect the inquiry or question, together with name, contact details (email address, phone number and mailing address) and any other additional information voluntarily submitted to us through a Contact Us form or an email. We use this information to address the inquiry and respond to the question.

Online Store

For orders and purchases placed through our online store on this site, we collect order details, name, institution name and address (if applicable), email address, phone number, shipping and billing addresses, credit/debit card information, shipping options and any instructions. We use this information to complete transactions, fulfill orders, communicate with individuals placing orders or visiting the online store, and for related purposes.

Surveys

Pearson may offer opportunities to provide feedback or participate in surveys, including surveys evaluating Pearson products, services or sites. Participation is voluntary. Pearson collects information requested in the survey questions and uses the information to evaluate, support, maintain and improve products, services or sites, develop new products and services, conduct educational research and for other purposes specified in the survey.

Contests and Drawings

Occasionally, we may sponsor a contest or drawing. Participation is optional. Pearson collects name, contact information and other information specified on the entry form for the contest or drawing to conduct the contest or drawing. Pearson may collect additional personal information from the winners of a contest or drawing in order to award the prize and for tax reporting purposes, as required by law.

Newsletters

If you have elected to receive email newsletters or promotional mailings and special offers but want to unsubscribe, simply email information@informit.com.

Service Announcements

On rare occasions it is necessary to send out a strictly service related announcement. For instance, if our service is temporarily suspended for maintenance we might send users an email. Generally, users may not opt-out of these communications, though they can deactivate their account information. However, these communications are not promotional in nature.

Customer Service

We communicate with users on a regular basis to provide requested services and in regard to issues relating to their account we reply via email or phone in accordance with the users' wishes when a user submits their information through our Contact Us form.

Other Collection and Use of Information


Application and System Logs

Pearson automatically collects log data to help ensure the delivery, availability and security of this site. Log data may include technical information about how a user or visitor connected to this site, such as browser type, type of computer/device, operating system, internet service provider and IP address. We use this information for support purposes and to monitor the health of the site, identify problems, improve service, detect unauthorized access and fraudulent activity, prevent and respond to security incidents and appropriately scale computing resources.

Web Analytics

Pearson may use third party web trend analytical services, including Google Analytics, to collect visitor information, such as IP addresses, browser types, referring pages, pages visited and time spent on a particular site. While these analytical services collect and report information on an anonymous basis, they may use cookies to gather web trend information. The information gathered may enable Pearson (but not the third party web trend services) to link information with application and system log data. Pearson uses this information for system administration and to identify problems, improve service, detect unauthorized access and fraudulent activity, prevent and respond to security incidents, appropriately scale computing resources and otherwise support and deliver this site and its services.

Cookies and Related Technologies

This site uses cookies and similar technologies to personalize content, measure traffic patterns, control security, track use and access of information on this site, and provide interest-based messages and advertising. Users can manage and block the use of cookies through their browser. Disabling or blocking certain cookies may limit the functionality of this site.

Do Not Track

This site currently does not respond to Do Not Track signals.

Security


Pearson uses appropriate physical, administrative and technical security measures to protect personal information from unauthorized access, use and disclosure.

Children


This site is not directed to children under the age of 13.

Marketing


Pearson may send or direct marketing communications to users, provided that

  • Pearson will not use personal information collected or processed as a K-12 school service provider for the purpose of directed or targeted advertising.
  • Such marketing is consistent with applicable law and Pearson's legal obligations.
  • Pearson will not knowingly direct or send marketing communications to an individual who has expressed a preference not to receive marketing.
  • Where required by applicable law, express or implied consent to marketing exists and has not been withdrawn.

Pearson may provide personal information to a third party service provider on a restricted basis to provide marketing solely on behalf of Pearson or an affiliate or customer for whom Pearson is a service provider. Marketing preferences may be changed at any time.

Correcting/Updating Personal Information


If a user's personally identifiable information changes (such as your postal address or email address), we provide a way to correct or update that user's personal data provided to us. This can be done on the Account page. If a user no longer desires our service and desires to delete his or her account, please contact us at customer-service@informit.com and we will process the deletion of a user's account.

Choice/Opt-out


Users can always make an informed choice as to whether they should proceed with certain services offered by InformIT. If you choose to remove yourself from our mailing list(s) simply visit the following page and uncheck any communication you no longer want to receive: www.informit.com/u.aspx.

Sale of Personal Information


Pearson does not rent or sell personal information in exchange for any payment of money.

While Pearson does not sell personal information, as defined in Nevada law, Nevada residents may email a request for no sale of their personal information to NevadaDesignatedRequest@pearson.com.

Supplemental Privacy Statement for California Residents


California residents should read our Supplemental privacy statement for California residents in conjunction with this Privacy Notice. The Supplemental privacy statement for California residents explains Pearson's commitment to comply with California law and applies to personal information of California residents collected in connection with this site and the Services.

Sharing and Disclosure


Pearson may disclose personal information, as follows:

  • As required by law.
  • With the consent of the individual (or their parent, if the individual is a minor)
  • In response to a subpoena, court order or legal process, to the extent permitted or required by law
  • To protect the security and safety of individuals, data, assets and systems, consistent with applicable law
  • In connection the sale, joint venture or other transfer of some or all of its company or assets, subject to the provisions of this Privacy Notice
  • To investigate or address actual or suspected fraud or other illegal activities
  • To exercise its legal rights, including enforcement of the Terms of Use for this site or another contract
  • To affiliated Pearson companies and other companies and organizations who perform work for Pearson and are obligated to protect the privacy of personal information consistent with this Privacy Notice
  • To a school, organization, company or government agency, where Pearson collects or processes the personal information in a school setting or on behalf of such organization, company or government agency.

Links


This web site contains links to other sites. Please be aware that we are not responsible for the privacy practices of such other sites. We encourage our users to be aware when they leave our site and to read the privacy statements of each and every web site that collects Personal Information. This privacy statement applies solely to information collected by this web site.

Requests and Contact


Please contact us about this Privacy Notice or if you have any requests or questions relating to the privacy of your personal information.

Changes to this Privacy Notice


We may revise this Privacy Notice through an updated posting. We will identify the effective date of the revision in the posting. Often, updates are made to provide greater clarity or to comply with changes in regulatory requirements. If the updates involve material changes to the collection, protection, use or disclosure of Personal Information, Pearson will provide notice of the change through a conspicuous notice on this site or other appropriate way. Continued use of the site after the effective date of a posted revision evidences acceptance. Please contact us if you have questions or concerns about the Privacy Notice or any objection to any revisions.

Last Update: November 17, 2020