HP Service Manager (SM): An Introduction to Performance Testing

I’ve wanted to write something about performance testing for some time now. Upon sitting down and putting pen to paper, or rather, fingers to keys, I've realized how very vast this topic is. As such, I think the best way to explore this topic will be as a series of posts on the specific challenges we face when dealing with HP Service Manager (SM) performance testing.

 

Let’s start with an introduction to this series. I would like to give you an overview of our vision here at HP Software R&D for HP Service Manager (SM), and provide you with some insight into the passion we have to make our product better with each release.  I would also like to offer information about existing resources that can help you deal with performance challenges you may face with your Service Manager system. There are a lot of preconceptions about performance.  I hope, after reading this post, you will better understand how our main goal is to make you, our SM customers, as happy as possible. When you are not happy, we wantto help you improve the performance of your deployment of Service Manager (SM).

Pic 1.png

It’s simple, Performance is Speed!  

 

Right? Well, it’s not that easy. I won’t argue the point that the speed with which your product performs is important but to define how fast is “fast enough” is something else entirely. When dealing with performance testing you need to know what you want your system to do based on clear goals and/or requirements. Quite often users don’t have a clear picture of what they want and expect from their Service Manager deployment. Perception plays a role here as well, and sometimes fast enough is less a question of speed than a question of good design and user interface (UI). Even more importantly, performance testing differs from funtional testing. Not only can your requirements/criteria vary immensely for the same test, but your test(s) need to focus on one factor at a time and must limit the other variables that could impact your results.

 

Suppose you test your system’s response time and are happy with the results, but what happens with 10 times the number of concurrent users ? Aha! So, if it’s not all about speed, then what else can impact the performance of your system?

Pic2.png

I know! Performance is Scalability!

 

Good one! How the product will scale with a larger number of users is crucial. And, I say it again, when you test how your product scales in either a test or production environment, you want to control other independent variables so you can identify and work around each bottleneck as you encounter it.

 

However, we aren’t there just yet because good performance goes beyond even scalability. You can test your environment until you’re blue in the face and make sure it scales up to 10 thousand users and then you go into production and BAM - everything goes wrong. Why? How could this have happened?

 

Life can be unfair, and I don’t have time in this post to talk about latency, load balancing, API's and DB tuning.  You’ll have to stay tuned for the follow-up articles from the experts on my team which will focus more on specific topics surrounding performance testing.

 

Until then, I want to remind you that we have a number of readily available documents to help you get the most out of your Service Manager system.

 

Our Performance Testing team in R&D has published the following material on the self-service portal:

 

-        Deployement and sizing guides

 

-        Performance benchmark test results (included in the sizing guide for SM 9.31)

 

-        Customer Benchmark Kits

 

We also have a number of dedicated white papers available for a variety of specific topics such as “Tuning your HardWare Load Balancer” or “Tailoring best practices to your system”.

Pic 3.png

In closing, just remember that performance testing is a journey not a certified destination.  Your requirements are unique and changing, so too will be the tests required to make sure you are getting the most out of Service Manager(SM). Rest assured that within the R&D department, we do everything we can (hand in hand with the Development team and the Solution Architects team) so that each and every version of Service Manager (SM) is faster, I mean more scalable, I mean… performs better than the last!

 

Pic 4.png

 

 

Until next time,

Rémi Temmos

HPSW SPM R&D Shanghai – Performance, Build and Automation Manager for SM

Leave a Comment

We encourage you to share your comments on this post. Comments are moderated and will be reviewed
and posted as promptly as possible during regular business hours

To ensure your comment is published, be sure to follow the Community Guidelines.

Be sure to enter a unique name. You can't reuse a name that's already in use.
Be sure to enter a unique email address. You can't reuse an email address that's already in use.
Type the characters you see in the picture above.Type the words you hear.
Search
Showing results for 
Search instead for 
Do you mean 
About the Author
Being number two in the ITSM install-based software industry, we feel it's our duty to enlighten and educate you about everything ITSM. We'l...


Follow Us
The opinions expressed above are the personal opinions of the authors, not of HP. By using this site, you accept the Terms of Use and Rules of Participation