SPONSORS:






User Tag List

Thanks Thanks:  0
Likes Likes:  0
Dislikes Dislikes:  0
Results 1 to 5 of 5
  1. #1
    Member
    Join Date
    Jul 2010
    Posts
    31
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Basic performance questions for web server.

    Let's say there are around 10,000 registered users on some website.

    1. How many servers do we need to accomodate such crowd?

    2. What is a goal for the throughtput for 10,000 users?

    3. What should be a bandwidth of these web servers?

    4. What is expected latency for such web servers?

  2. #2
    Member
    Join Date
    Jul 2010
    Posts
    31
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Basic performance questions for web server.

    Also, how would you test endurance of such web application (simple "login" for instance):

    How many concurrent users doing how many login transactions?

    Thanks!

  3. #3
    Moderator
    Join Date
    Aug 2001
    Location
    NC
    Posts
    6,018
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    1 Thread(s)
    Total Downloaded
    0

    Re: Basic performance questions for web server.

    1. At least 1, but fewer than 10,000. This depends upon how many users and how each user makes use of a finite resource associated with disk, CPU, RAM or network.

    2. Time to refer to your requirements documents, your system design documents, your architecture documents and to interview your stakeholders for their expectations. Throughput can be measured many ways, business critical transactions per (insert timeframe here), bits, bytes, hits, etc.. Depending upon your interviewed stakeholder you may receive business throughput information or technical throughput information.

    3. Assume this is network bandwidth (although it could be CPU bandwidth). This is directly related to how your application makes use of the network, how it manages the caching/aging of objects on the end users PC and whether or not distributed content management (such as AKAMAI or Amazon S3 storage or ....) is used. Nothing allowed to be cached, very busy interface with lots of bytes being transferred, lots of active users lends itself to very wide pipes. Conversely, an application which allows much content to be cached with a large number of return visitors, has small amount of information which is transferred with small byte counts on state management, session and graphical objects and distributed content might be able to support your 10,000 on a T1

    4. Latency is network question. This may be something you can control (to a degree) with priortization of packets and QOS assignments by port inside of a corporate LAN, but once you hit the Internet you are at the mercy of the big bad cloud between your requestor and the target.

    5. How many concurrent? Go to the test plan. You should have a load profile present in your performance test plan. You should also have requirements which directly lead to your test design, includinjg perhaps sustained load requirements for your LOGIN transaction.

    There are models for internet facing and corporate facing applications based upon a defined credential population. There do not seem to be any reliable models yet for sites like Facebook & Twitter, social networking sites, that would identify the percentage out of your 10,000 which were active at any given time.
    James Pulley

    Replace ineffective offshore contracts, LoadRunnerByTheHour. Starting @ $19.95/hr USD.

    Put us to the test, skilled expertise is less expensive than you might imagine.

    Twitter: @LoadRunnerBTH @PerfBytes

  4. #4
    Member
    Join Date
    Feb 2001
    Location
    Raleigh, NC USA
    Posts
    84
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Basic performance questions for web server.

    [ QUOTE ]
    How many concurrent users doing how many login transactions?

    [/ QUOTE ]

    If you have some basic metrics on the usage of the site, one of these Virtual User Calculators may help you come up with a rough starting point.
    Chris Merrill
    Load Testing Software and Services
    http://www.webperformance.com/

  5. #5
    Member
    Join Date
    Jul 2010
    Posts
    31
    Post Thanks / Like
    Mentioned
    0 Post(s)
    Tagged
    0 Thread(s)
    Total Downloaded
    0

    Re: Basic performance questions for web server.

    Thanks guys!

 

 

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 02:30 AM.

Copyright BetaSoft Inc.