Professional Web Applications Themes

website-cluster, cost-effective hardware? - MySQL

Hi, I'm about to start a website on a shoestring, so I'm looking to get the most bang for my buck in terms of hardware. I will be running this site on a linux cluster with mysql and there will be lots of reads from the DB and not so many writes. For various reasons, I have to create my own 'datacenter' so can't use other provider so please don't suggest that :-). I'm thinking I'll start with maybe 5-6 or so machines and add to it as necessary. So has anyone done studies on what is most cost effective? ...

  1. #1

    Default website-cluster, cost-effective hardware?

    Hi,

    I'm about to start a website on a shoestring, so I'm looking to get
    the most bang for my buck in terms of hardware. I will be running this
    site on a linux cluster with mysql and there will be lots of reads
    from the DB and not so many writes.
    For various reasons, I have to create my own 'datacenter' so can't use
    other provider so please don't suggest that :-). I'm thinking I'll
    start with maybe 5-6 or so machines and add to it as necessary.

    So has anyone done studies on what is most cost effective? f.ex. on
    the processors a Xeon 1.6 quad-core costs about 10 times more than a
    AMD Sempron 3600+.
    But the Xeon does have 2MB L2 cache compared with the AMDs 256kb and
    I'm not sure exactly how much faster it is (is it 10 times faster? I
    think not).
    How important is the L2 cache in DB applications?

    Also, the bottleneck with probably be the disk, so would it not be
    better with lots of cheap-cpu machines compared to a few high-end-cpu
    machines? Since there will be mostly reads from the DB I would think
    it better to spread it on many smaller machines. So would 10 small
    machines not be better than 5 larger? (assuming the larger ones have
    f.ex. dual or quadcore cpu's and the smaller have semprons for
    example). There would be more disks in the low-end-cpu machines since
    there are more machines, so wouldn't that be better for the
    bottleneck?
    What about motherboard? is there a lot to gain from a high-end board
    compared to a cheap one? is it worth the extra costs for this
    application?

    Finally, I have no idea if I should start with 5 machines or 20? or 2?
    does anyone have some real-world examples? Like f.ex. what did myspace
    start with? or facebook? I know I will have to run some tests and
    experiment, but it would be nice to have some idea of what other DB
    sites started with.

    My gut-feeling is to go with cheap cpu's (like AMD sempron 3600+),
    lots of fast RAM, cheap motherboards, very good gigabit NICs and
    switch, fast high rpm high cache SATA disks (500gb or so) (one in each
    server), cheap tower cabinets. And then start with maybe 5 or 6 of
    them or so. My only worry with that setup is the small L2 cache. I
    wouldn't go with redundant PSUs or disks (raid) or NICs because there
    is redundancy in the cluster itself anyway. What do you think of that
    setup?

    Thanks.
    Tobias

    nsa.usa@gmail.com Guest

  2. #2

    Default Re: website-cluster, cost-effective hardware?

     
    People get $80 (more???) an hour to design networking hardware suited
    to specific tasks. Pay someone before you bet thousands of $$ on your
    guess that "the bottleneck with probably be the disk."

    You do know the answer to *every* question you asked was either "it
    depends" or "it isn't relevant what Fed Ex did in 1987" right?

    Oh, and you left out load balancers, switches, routers, monitoring
    software, etc, etc, etc.


    christopher@dailycrossword.com Guest

  3. #3

    Default Re: website-cluster, cost-effective hardware?

    Hi,

    Thanks for your reply.
    The main question was about if anyone has done tests to see whats
    cost effective., mainly on the CPU side.
    And I explained it's a DB app, etc etc. I do realize that the DB app
    can be CPU intensive or not, but there should still exist some info
    (that others have made about this) about both. F.ex. I did come across
    a site that showed benchmark tests comparing Xeon to Opteron in a
    website-DB-app environment with 500 simultaneous users doing CC
    transactions. Ok, not exactly the same as what I'm doing, but it
    showed Opteron was slightly better and also cheaper (cost/trans was
    lower). So this is what I'm looking for regarding info on CPU's. But
    has anyone compared Xeon to low-end Semprons in DB environment? Then
    there are all the other questions.
    So the answer is not 'it depends', the answer is looking at various
    examples of what others have done and then coming to a reasonably
    qualified conclusion.
    What do you think your 80$-expert does? He looks at what others have
    done, and makes a reasonable qualified quess as well as calculating
    loads. Maybe there are books on the subject? websites dedicated to
    this sort of thing? whatever? I'm looking for resources.
    Allthough it may not be relevant what fedex did in 87, it might very
    well be relevant what myspace did in 2006!
    And a good example is Friendster that becamed bogged down due to
    mainly too much cpu overhead I beleive. I mentioned myspace and
    facebook, because what their needs are seems similar to mine, not that
    I'm making the same type of site but the DB queries and frequency of
    updates etc seem somewhat similar. i.e. they store stuff, that is then
    looked at by many users.
    Where are their bottlenecks and what did they do about it?

    I do realize there are lots of variables, and also the rest of the
    network need to be considered. I can (and will have to) sit down and
    do calculations on everything and find out if f.ex gigabit ethernet
    between nodes is necessary or not, how many cpu cycles does the app
    use etc. etc. but it's still nice to see some examples as there is no
    reason to re-invent the wheel...

    Cheers,
    Tobias

    On Jul 26, 7:10 am, com wrote: 
    >
    > People get $80 (more???) an hour to design networking hardware suited
    > to specific tasks. Pay someone before you bet thousands of $$ on your
    > guess that "the bottleneck with probably be the disk."
    >
    > You do know the answer to *every* question you asked was either "it
    > depends" or "it isn't relevant what Fed Ex did in 1987" right?
    >
    > Oh, and you left out load balancers, switches, routers, monitoring
    > software, etc, etc, etc.[/ref]


    nsa.usa@gmail.com Guest

  4. #4

    Default Re: website-cluster, cost-effective hardware?

    com wrote:
     

    There are lies, damned lies and benchmarks.

    Benchmarks prove nothing. In most cases one decides what result is
    desired and then choses the benchmark. BTSTMT.

    Also "cost effectiveness" is a broad field. I.e. Sun claims the T2000
    to deliver the best performance per power consumtion. Since you have
    to pay for power (and airconditioning) this might be worth to consider.

    As for simple "off the shelf" PC hardware: it is widely acknowledged
    that AMD and Intel deliver virtually the same performance with AMD
    being cheaper. But as I said above: both AMD and Intel have benchmarks
    that "prove" the superiority of their respective platform.
     

    Apples and pears. Semprons are usually compared to Celerons (and win
    hands down). High end Xeons are compared to high end Opterons.

    There is another (more or less) obvious truth about power-for-the-buck:
    the more performance you need in a single box, the more costly it
    becomes. There is no linear correlation between price and performance.
    The sweet spot is somewhere in the mid segment. This is the exact
    reason why Google & Co run on thousands of cheap boxes.

    And the good news is: MySQL very well supports this style of scaling
    (by tion mostly). It's called scaleout.
     

    Indeed. And many of them are just *unknown* now. There is no answer
    except "it depends" for now.


    XL
    Axel Guest

  5. #5

    Default Re: website-cluster, cost-effective hardware?

    In article <homelinux.org>,
    Axel Schwenke <de> wrote:
     
    >
    > There are lies, damned lies and benchmarks.
    >
    > Benchmarks prove nothing. In most cases one decides what result is
    > desired and then choses the benchmark. BTSTMT.
    >
    > Also "cost effectiveness" is a broad field. I.e. Sun claims the T2000
    > to deliver the best performance per power consumtion. Since you have
    > to pay for power (and airconditioning) this might be worth to consider.
    >
    > As for simple "off the shelf" PC hardware: it is widely acknowledged
    > that AMD and Intel deliver virtually the same performance with AMD
    > being cheaper. But as I said above: both AMD and Intel have benchmarks
    > that "prove" the superiority of their respective platform.

    >
    > Apples and pears. Semprons are usually compared to Celerons (and win
    > hands down). High end Xeons are compared to high end Opterons.
    >
    > There is another (more or less) obvious truth about power-for-the-buck:
    > the more performance you need in a single box, the more costly it
    > becomes. There is no linear correlation between price and performance.
    > The sweet spot is somewhere in the mid segment. This is the exact
    > reason why Google & Co run on thousands of cheap boxes.
    >
    > And the good news is: MySQL very well supports this style of scaling
    > (by tion mostly). It's called scaleout.

    >
    > Indeed. And many of them are just *unknown* now. There is no answer
    > except "it depends" for now.
    >
    >
    > XL[/ref]

    Or the other answer "Hire a consultant who's done this before"

    --
    DeeDee, don't press that button! DeeDee! NO! Dee...



    Michael Guest

Similar Threads

  1. Is there a cost effective printer?
    By pisosse@adobeforums.com in forum Adobe Photoshop Mac CS, CS2 & CS3
    Replies: 14
    Last Post: April 7th, 01:20 AM

Bookmarks

Posting Permissions

  • You may not post new threads
  • You may not post replies
  • You may not post attachments
  • You may not edit your posts
  •  

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60 61 62 63 64 65 66 67 68 69 70 71 72 73 74 75 76 77 78 79 80 81 82 83 84 85 86 87 88 89 90 91 92 93 94 95 96 97 98 99 100 101 102 103 104 105 106 107 108 109 110 111 112 113 114 115 116 117 118 119 120 121 122 123 124 125 126 127 128 129 130 131 132 133 134 135 136 137 138 139