- From: Mark Nottingham <mnot@mnot.net>
- Date: Tue, 21 Jul 2009 11:05:23 +1000
- To: Henrik Nordstrom <henrik@henriknordstrom.net>
- Cc: Adrien de Croy <adrien@qbik.com>, HTTP Working Group <ietf-http-wg@w3.org>
OK, I think we had rough consensus before, and it's been echoed here, that the limit needs to be raised but not completely removed. However, I think we need to highlight this with the larger IETF to make sure people understand why we're raising it. I'll put it on the agenda for next week for discussion in Stockholm. Cheers, On 21/07/2009, at 2:59 AM, Henrik Nordstrom wrote: > mån 2009-07-20 klockan 18:39 +1200 skrev Adrien de Croy: > >> I would have thought slow-startup algorithms would also work >> against the >> advantage of opening too many connections. > > Only in cases where the data requested on the connection is small. And > additionally some (not a small number) web servers are tuned to send > their data quite aggressively effectively having slowstart disabled > (when viewed from normal connection rates) until there is packet loss. > >> Also, download managers >> generally do multiple simultaneous range requests. The more parts >> you >> request, the more request/response overhead reduces your >> throughput, so >> there's an incentive not to go over the top there as well. > > Depends on the "chunk" size. > > Regards > Henrik > -- Mark Nottingham http://www.mnot.net/
Received on Tuesday, 21 July 2009 01:06:03 UTC