Slashdot Mirror


Sendmail On IBM Mainframes Running GNU/Linux

raffe writes: "Cnet reports that Sendmail has released a version of its e-mail server software that can run on Linux-powered IBM mainframe computers. In one benchmark test, IBM found that it was possible to house 2 million e-mail accounts on a single server, with 10 percent of the users accessing their mail at any given moment" For some reason though, IBM zSeries machines aren't listed at pricewatch ;)

4 of 132 comments (clear)

  1. The mainframe's not dead... by Jay+Maynard · · Score: 5, Interesting
    This is the kind of thing that mainframes do well: information processing with little or no actual computation. Their I/O abilities really make the difference here.


    Hopefully, this kind of result will show the skeptics that there's a real purpose for the big boxes.

    --
    Disinfect the GNU General Public Virus!
  2. Re:Shock the Monkey by Anonymous Coward · · Score: 0, Interesting

    If you let him sit in his shit and piss and don't feed him for about two weeks, he'll get ebola. Ask, if you keep feeding him Cokes, he'll puke all over the place.

  3. What did they use to generate 400k users? by Anonymous Coward · · Score: 1, Interesting

    Did they use Specmail2001 (from http://www.spec.org) to generate the load on their MTA? If so, I don't think that 400k is that special.

    I've generated very close to that number (minus 100k) with a dual 1Ghz cpu w/1Gb ram system, and a Netapp F820 for storage with the Syntegra Intrastore product. I even had the close to the same CPU utilization they had. This is user space CPU of course. The system CPU time is higher due to the NFS overhead.

    Once I get around a linux 2.4.x virtual memory problems, I'll publish the results to the spec.org site for everyone to see.

    http://intrastore.us.syntegra.com

    1. Re:What did they use to generate 400k users? by valdis · · Score: 2, Interesting
      You did that with ONE system? Interesting - the Mirapoint results for 400K users needed:
      • 2 POP server machines
      • 5 SMTP router machines
      • 10 message store machines
      • 1 benchmark manager machine
      • 1 mail sink machine
      • 5 load generator machines

      Dealing with 300K outbound postings is no biggie - I've been able to deal with that level on an old IBM RS6000-F30 (166mz 604). You don't need really big iron for outbound mail until you have more than 500K or so RCPT TO's on one piece of mail. It's mostly a matter of good queue management, and Sendmail 8.12 has new queue management code that makes it even easier (I should know, I tested it). The only real magic is not getting logjammed due to DNS waits and unreachable destinations.

      On the other hand, having 40K people doing POP accesses while you're dumping mail into their mailboxes is trickier. Some of the more obvious issues:

      • The obvious popd solution leaves you 40K processes running at once. This could be bad.
      • Locking issues get interesting.
      • Even with a journaling filesystem, you can get killed on the I/O. Remember that writing to a file means you also need to do stuff with the inode....