SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 6 of 6
  1. #1
    Junior Member
    Join Date
    Feb 2002
    Location
    Pleasanton, CA, USA
    Posts
    2
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Proxy Server Performance Testing

    The company that I work for is in the process of developing a proxy server and is interested in performance/load testing.

    There are 3 major aspects that the company would like to test:

    Concurrent Users - HTTP
    Maximum Throughput - HTTP
    Concurrent Users - POP3
    Maximum Throughput - POP3
    Concurrent Users - SMTP
    Maximum Users - SMTP

    The question that I have is, how would you "define" a user. Is there an industry standard or any data I can start with, as to how often a user browses a web site, and the amound of data the web pages contain?

    Again, Is there an industry standard, or any data I can start with, as to how often a user checks their POP3 account. How many e-mails the user downloads, and the size of e-mail.

    The same goes for SMTP.


    If anyone can give me some information or point me in the right direction, I would appreciate it. Also, if you have any additional questions, feel free to ask.


    Thank you,

    Sean

  2. #2
    Senior Member
    Join Date
    Jul 2002
    Location
    Palm Bay, FL USA
    Posts
    2,346
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Proxy Server Performance Testing

    Wow, I think this is the first time that someone has asked for industry standards in a case where I think they would actually apply! You definately should get an award for that.

    There are two places that I think you should start looking. Neilsen (http://www.nielsen-netratings.com/) and Keynote (http://www.keynote.com/). Those are mostly web usage and not mail usage though.

    Doing a Yahoo search for "email usage statistics" I found a ton of tools to collect statistics, some statistics from isolated organizations and surprising number of email usage policy documents. <shrug> Sorting through those might result in something. I'm betting that one of those tool vendors has some white papers or quotable statistics somewhere. I, of course, will not vouch for their reliability.

    That is where I would start at least. Good luck, and let us know if you find anything really good!

    ------------------
    Scott Barber, Sr. Performance Engineer
    sbarber@noblestar.com
    http://www.noblestar.com
    http://www.perftestplus.com
    Scott Barber
    Chief Technologist, PerfTestPlus
    Executive Director, Association for Software Testing
    Co-Author, Performance Testing Guidance for Web Applications
    sbarber@perftestplus.com

    If you can see it in your mind...
    you will find it in your life.

  3. #3
    Junior Member
    Join Date
    Feb 2002
    Location
    Pleasanton, CA, USA
    Posts
    2
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Proxy Server Performance Testing

    Scott,


    Thank you for the information. I can't see myself being the only person in need of this type of data...or is it that the majority of people who use the QAForums are performing load testing against web servers and not some sort of device between the user and the web server.

    I hope I do find some usefull information and if I do, I will do my best to pass the valueable information to the forum.


    Thanks again,


    Sean

  4. #4
    Moderator
    Join Date
    Aug 2000
    Location
    Vancouver, BC, Canada
    Posts
    1,189
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Proxy Server Performance Testing

    Sean,

    The only thing I have ever done with Proxy server testing is outbound testing.
    Outbound testing in the case meant that we expected an increase of 100% of our internal users going to the Internet through the proxy server and the systems people were concerned about the handling capability of the proxy server.

    Since we knew the actual amount of users, the new amount and the actual current usage (logs), we had no problem coming up with a set of tests that would simulate that load for double the amount of users.

    As driver for our test, we used the actual logfiles and extracted each individual get and replayed that. We skipped all the HTML post statements because it was complicated to deal with that and they only represented a tiny percentage of our total test. We also skipped all SSL traffic because they were often related to electronic banking, purchasing and such. Since we were mostly looking for an order of magnitude, we figured we could get away with this.

    Interestingly enough we found that the proxy server was fine, but that our Internet pipe did not have enough bandwidth.

    We did not do any POP3 traffic, streaming media etc. (Basically because it is not allowed through our proxy servers)

    Potential issues with proxy server testing:
    * Difficult (to simulate) variants of traffic
    * Caching, proxy servers also often act as accellerators and rely on caching. We found that out when we started with a cold server and compared that to a warm server.
    * Proxy server authentication. Most MS proxy products (Proxy server and ISA) deploy NTLM or NTLM/Kerberos. Some Load test tools have difficulties dealing with that efficiently.

    Anyway sofar this little braindump, I hope this helps you a little bit.

    ------------------
    Roland
    Roland Stens

  5. #5
    Senior Member
    Join Date
    Jul 2002
    Location
    Palm Bay, FL USA
    Posts
    2,346
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Proxy Server Performance Testing

    Actually, it's that most folks doing performance testing are doing so against applications (or in situations) where user patterns are fairly predictable. By virtue of the fact that you are building something that could be deployed "in front of" anything, our methods of determining usage patterns (I like to call them user community models) don't really apply. For instance Roland mentioned using logs from existing applications - that is the ideal approach, but you really don't have the option of calling all of your potential clients and say "send me all of your log files from you current proxy server for the last month so I can do an accurate performance test."

    You might be able to find out what kind of benchmarks your competitors use and replicate that distribution.

    ------------------
    Scott Barber, Sr. Performance Engineer
    sbarber@noblestar.com
    http://www.noblestar.com
    http://www.perftestplus.com
    Scott Barber
    Chief Technologist, PerfTestPlus
    Executive Director, Association for Software Testing
    Co-Author, Performance Testing Guidance for Web Applications
    sbarber@perftestplus.com

    If you can see it in your mind...
    you will find it in your life.

  6. #6
    Moderator
    Join Date
    Aug 2000
    Location
    Vancouver, BC, Canada
    Posts
    1,189
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Proxy Server Performance Testing

    Maybe this white paper about testing proxy caching is of use.
    I have not read it myself, but I am just tossing it out. http://www.itpapers.com/cgi/PSummary...=4218&scid=151

    ------------------
    Roland
    Roland Stens

 

 

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  
Search Engine Optimisation provided by DragonByte SEO v2.0.36 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Resources saved on this page: MySQL 10.71%
vBulletin Optimisation provided by vB Optimise v2.6.4 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
User Alert System provided by Advanced User Tagging v3.2.8 (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
vBNominate (Lite) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Feedback Buttons provided by Advanced Post Thanks / Like (Pro) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
Username Changing provided by Username Change (Free) - vBulletin Mods & Addons Copyright © 2016 DragonByte Technologies Ltd.
BetaSoft Inc.
Digital Point modules: Sphinx-based search
All times are GMT -8. The time now is 04:28 PM.

Copyright BetaSoft Inc.