Advertisement
If you have a new account but are having problems posting or verifying your account, please email us on hello@boards.ie for help. Thanks :)
Hello all! Please ensure that you are posting a new thread or question in the appropriate forum. The Feedback forum is overwhelmed with questions that are having to be moved elsewhere. If you need help to verify your account contact hello@boards.ie

Single-core VM Server acceptable for performance testing?

Options
  • 29-08-2013 2:12pm
    #1
    DublinWriter
    Registered Users Posts: 9,555 ✭✭✭


    I have a client speed testing an asp.net app using a virtualised 2003 Windows server.

    They are encountering issues with performance. They configured the server to be single-core.

    While I realise this is a very general question, I can't believe anyone is deploying single-core servers in this day and age. Should they even be testing with a single-core virtual server?


Comments

  • Options
    #2
    CreepingDeath
    Closed Accounts Posts: 8,016 ✭✭✭CreepingDeath


    DublinWriter wrote: »
    Should they even be testing with a single-core virtual server?

    Depends... if their production environment is also single core, then it's a valid test environment.

    But two cores doesn't mean twice the speed, it doesn't always scale linearly eg. if the disk, memory or network is the bottleneck.

    Or if all their performance testing is relative to previous tests, they may not want to alter the environment so they can
    compare previous test runs against the new ones.


  • Options
    #3
    ChRoMe
    Registered Users Posts: 2,021 ✭✭✭ChRoMe


    CreepingDeath wrote: »

    But two cores doesn't mean twice the speed, it doesn't always scale linearly eg. if the disk, memory or network is the bottleneck.

    Indeed single core may actually be acceptable depending on the use cases


  • Options
    #4
    jimmycrackcorm
    Registered Users Posts: 14,329 ✭✭✭✭jimmycrackcorm


    I think it's bizarre to set single core in a Vmware instance (unless it is for a very low usage/demand process). Vmware has the advantage of utilising CPU and memory over-commitment vs standard servers. When you consider that with a number of machines over a period of time will balance out processor usage given typical peaks and troughs for CPU use, it make sense (and I've found in practice) to spec vms well. For critical VM's I'd set minimum resource thresholds.

    In any case a single core doesn't provide much capacity outside O/S functionality. Memory also is a very significant factor. We had the same performance issues described with single core 1GB ram server 2003 at work. For windows server I'd spec a minimum virtual server with dual core and 4gb of RAM.


Advertisement