Path: utzoo!attcan!uunet!wuarchive!sdd.hp.com!think.com!paperboy!osf.org!tmt From: tmt@osf.org (Tom Talpey) Newsgroups: comp.protocols.nfs Subject: Re: Summary: NFS over links slower than 10 Mbit/sec Message-ID: <18848@paperboy.OSF.ORG> Date: 11 Feb 91 15:23:31 GMT References: <1991Feb5.031603.2969@amd.com> <7371@exodus.Eng.Sun.COM> Sender: news@OSF.ORG Reply-To: tmt@osf.org (Tom Talpey) Organization: Open Software Foundation, Cambridge, Massachusetts Lines: 22 In article , jim@cs.strath.ac.uk (Jim Reid) writes: |> In article <7645@exodus.Eng.Sun.COM> brent@terra.Eng.Sun.COM (Brent Callaghan) writes: |> In article , meissner@osf.org (Michael Meissner) writes: |> > Always set UDP checksums, even over ethernet. |> When NFS first shipped on Suns, we had just one or two MIPS |> of CPU to play with. UDP checksumming was a significant overhead. |> In the early days of Sun and NFS, most vendors were using the 4.2 |> BSD TCP/IP code. This didn't compute UDP checksums properly The 4.2 code botched UDP checksums on big-endian machines. Suns, being of that ilk, had obvious reasons to preserve interoperability. However, the "performance" benefit was another reason. UDP checksums should generally be enabled these days, but remember that doing so affects both outgoing and incoming checksum computation on all UDP packets, not just NFS. The best solution is to turn off checksums on the broken systems and enable everything else, but sometimes it's not feasible. Tom Talpey tmt@osf.org Brought to you by Super Global Mega Corp .com