Re: data on what makes web access slow [Fwd: end user impact of T CP speed (was: Re: fast tcp && win2000)]

From: Nicholas Roberts (nicholas@synarchy.net)
Date: Fri, Jan 14 2000

  • Next message: Shamrok888@aol.com: "more info"

    Message-ID: <000f01bf5e92$21bcf110$a25519cb@synarchy.net>
    From: "Nicholas Roberts" <nicholas@synarchy.net>
    To: "Neff, Robert" <Robert.Neff@usmint.treas.gov>, <www-wca@w3.org>
    Date: Sat, 15 Jan 2000 00:20:26 +1100
    Subject: Re: data on what makes web access slow [Fwd: end user impact of T 	CP speed   (was: Re: fast tcp && win2000)]
    
    might is right, bullshit walks, money talks, charge it. enjoy the
    experience.
    
    
    
    
    ----- Original Message -----
    From: "Neff, Robert" <Robert.Neff@usmint.treas.gov>
    To: <www-wca@w3.org>
    Sent: Friday, January 14, 2000 10:35 PM
    Subject: RE: data on what makes web access slow [Fwd: end user impact of T
    CP speed (was: Re: fast tcp && win2000)]
    
    
    > Does anyone have any stats on what constituents "running with the big
    boys"
    > in hi-end e-commerce.  We are trying to get better data so we can improve
    > and tune capacity planning.  We just got slammed on Tuesday. Within 20
    > minutes we were sustaining 20,000 concurrent users with peaks up to 48,000
    > concurrent users. This was non stop throughout the day!
    >
    > Does anyone have any comparative data?  Old or recent will do, but prefer
    > recent...any suggestions as to other sources?
    >
    > Previous concurrent usage was 1300 so I rearchitected the site for 13,000
    > (times 10).  Say I was a little off?  We have another product drop in
    > approximately 15 days and are working to re-architect the site to plan for
    > minimum 250,000 concurrent users and trying to get up to 500,000.
    >
    > Just moved the couch closer to the office....
    >
    > Robert Neff
    > Chief of Web Operations
    > United States Mint
    >
    >