Become a fan of Slashdot on Facebook

 


Forgot your password?
Close
typodupeerror
×
BSD Operating Systems

1.6 Million IP Connections on FreeBSD 74

Posted by michael from the your-call-is-important-to-us-please-hold dept.
An anonymous reader writes "FreeBSD developer Terry Lambert, in a recent posting to the 'freebsd-hackers' mailing list, mentioned that he'd tuned a FreeBSD 4.4 box with 4GB of RAM to achieve 1,603,127 simultaneous IP connections, and goes on to say: 'As far as I know, I hold the single machine connection record for an x86 box.' This is an impressive achievement any way you look at it (though it begs the question of whether or not the box had any resources left to actually do anything with those connections...), and it speaks well of both FreeBSD's capabilities and Terry's skills and knowledge. I'm curious, though, if anyone has approached, matched, or exceeded that number elsewhere?"
This discussion has been archived. No new comments can be posted.

1.6 Million IP Connections on FreeBSD More

1.6 Million IP Connections on FreeBSD

Comments Filter:

  • Alternative Headline: (Score:5, Funny)

    by leviramsey ( 248057 ) on Thursday January 30, 2003 @08:59AM (#5188518) Journal

    *BSD handles 1.6 million connections without dying.

  • sure (Score:5, Funny)

    by pizza_milkshake ( 580452 ) on Thursday January 30, 2003 @09:31AM (#5188784)
    my webhosting box does twice that during peak hours, but then i'm hosting free porn so it doesn't count ;)

  • Prove it... (Score:5, Funny)

    by aridhol ( 112307 ) <ka_lac@hotmail.com> on Thursday January 30, 2003 @10:45AM (#5189301) Homepage Journal
    Post the address of that box here. We'll give it a real stress test.
  • That's a lot of connections to the intellectual property of others. Perhaps he should get a lawyer.

    "What kind of artist are you?"

    "I'm a 'Prior Artist'."

  • More Interesting (Score:5, Insightful)

    by mnmn ( 145599 ) on Thursday January 30, 2003 @11:23AM (#5189542) Homepage

    What has been tested is simply the number of concurrent connections. More practical would be simple retrieving of say 1kb data from a database and printing it out on a very simple HTML, and checking the maximum number of THESE connections. In effect trying to really httpblast DDoS style the FreeBSD with sheer number of connections. The box will have to be massive with 4GB RAM at least (we're testing OS here not hardware) and the connection maybe (multiple?) gigabit ethernet. The result would theoretically be lower than 1.6 million but we need to show FreeBSD can scale in practical tests like these. Results from a test like that will have the power to change vendors' minds from trying to run IIS and MS SQL for a high volume site.

    • Re:More Interesting (Score:3, Interesting)

      by hfastedge ( 542013 )
      theres no point in the mysql test on top of the ip test, it then becomes even moreso a pure kernel test (eg scheduler/vm....), plenty of which have been done.

      But i do think that serving out 1k of html would make the test a bit more solid.

      Id be interested, given this narrow field thats being tested on how linux would hold up under the same tweaking, and what tweaking exactly that would require as compared to the freebsd (yes its probably trivial, but im just curious).
    • The test you propose is not a more interesting test. It is simply a different test.

      The original test is designed to stress the theoretical maximum number of connections, which exercises the network stack, in particular the pcb hashing mechanisms and multiple IP address handling.

      The test you propose is a real world scalability test, which has a much different purpose.
  • What are IP "Connections"? AFAIK, the transport layers can have connections, but not the network layer. IP has datagrams. I did RTFA, but there's not a whole lot of context in that message, and I was too lazy to go read the whole thread. Are they just talking about TCP connections, or what?

  • FreeBSD is dying (Score:5, Funny)

    by mcgroarty ( 633843 ) <{moc.liamg} {ta} {ytraorgcm.nairb}> on Thursday January 30, 2003 @02:13PM (#5190454) Homepage
    ...to take on heavy server loads.
  • What did happen to the *BSD Box when reached the 1.6 millon of connections?, did he dyed?, did it just stoped accepting connections due low resources? why he wasn't able to pass the 1.6millon of IP connections when we can count as much 4294967296 differen IPs for a network?
    • Maybe that's just how many incoming connections there were? It doesn't state that this was the possible limit, just that it's a high number he reached.
  • I would like to claim that I have the world record for a one minute load average on a FreeBSD machine.

    http://gomerbud.com/daver/computing/top.asc [gomerbud.com]

    Any contenders?
    • We had a quad-Xeon FreeBSD machine sit on a load of over 900 (can't remember the exact number, but it was over 900) for about 16 hours because of some runaway scripts.

      And it was still responsive enough so that the only report from the end users (a student lab of about 400 machines) was "the network seems a little slow today..."

      • It's easy:
        Take a busy machine that's mounting, say, email boxes via NFS.
        Turn off the NFS server (or just wait for your P.O.S. EMC to crash

        Wait a moment or three as it becomes unresponsive.

        Type "uptime" and wait 3 minutes for it to return with a load average of 2085.

        Now, actual CPU usage is the key. But getting a really solid stack is what FreeBSD excels at.

      • We had a similar situation on our 4-CPU 100-user Sun. An email the next day explained that one of the CPUs was being hot swapped.
    • Just use the shell-script in my .sig :-)

  • 64-bit version (Score:3, Insightful)

    by yancey ( 136972 ) on Thursday January 30, 2003 @04:48PM (#5191560)
    I wonder if the 64-bit version of FreeBSD would be able to improve upon this, since it can access more memory.
  • The given link for Terry's message seems to be broken.

    Take this one [freebsd.org]
  • Well, then, FreeBSD certainly has come a long way since late 2000, when it couldn't handle more than a thousand or so TCP connections: (Security Advisory [bindview.com])

    Personally, I would be very interested in seeing how well the machine in this record-setting example handles an attack of the type mentioned in the above referenced article.

Slashdot Top Deals

Top Ten Things Overheard At The ANSI C Draft Committee Meetings: (10) Sorry, but that's too useful.

Close