Date: Mon, 08 Mar 2004 15:32:37 -0800 From: Jeffrey Hsu <hsu@freebsd.org> To: freebsd-net@freebsd.org Cc: freebsd-current@freebsd.org Subject: Who wants SACK? (Re: was My planned work on networking stack) Message-ID: <200403082325.i28NPAa3010399@mtaw6.prodigy.net> In-Reply-To: Message from Luigi Rizzo <rizzo@icir.org> of "Mon, 08 Mar 2004 10:56:42 PST." <20040308105641.A47564@xorpc.icir.org>
next in thread | previous in thread | raw e-mail | index | archive | help
>> I know that our organization would love to see SACK. Much of the >> high-performance network development that used to be on FreeBSD has >> moved to Linux simply because SACK is essential. You can't run >> trans-oceanic TCP streams of gigabit or more throughput without it. > > Whenever i hear these comments, i am very annoyed at one thing > (which in a smaller scale repeats all over the place): > people are more than happy to spend big money for things like > routers or bandwidth or any kind of "commercial" stuff, but when > it comes to open source it must be free or nothing. > > I hope it is clear to everyone that an investment in the 50K$ > range would provide a professional-grade implementation of SACK > for FreeBSD, and this money is in the noise for any organization > that uses trans-oceanic gigabit links. > The fact that nobody seems to care about funding such a work > either means that whatever is available already fits their What Luigi says is absolutely correct. It doesn't take a lot to get this done. I've talked to a number of companies about implementing SACK for them and while there was interest, no one wanted to fund it all themselves, potentially for the benefit of their competitors. I know of two that went and did it themselves for FreeBSD --- one of which did it wrong and saw zero benefit from SACK and another that did it right, but are keeping it proprietary as an edge. Given that Linux and Windows already have it, these and the multiple past efforts collectively seem like an unnecessary duplication of work. Perhaps if we could pool enough interest, we can raise enough to put this issue to rest once and for all. Jeffrey
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?200403082325.i28NPAa3010399>