| To: | Leonid Grossman <leonid.grossman@xxxxxxxx> |
|---|---|
| Subject: | Re: The ultimate TOE design |
| From: | Nivedita Singhvi <niv@xxxxxxxxxx> |
| Date: | Thu, 16 Sep 2004 09:18:49 -0700 |
| Cc: | "'Andi Kleen'" <ak@xxxxxxx>, "'David S. Miller'" <davem@xxxxxxxxxxxxx>, "'John Heffner'" <jheffner@xxxxxxx>, netdev@xxxxxxxxxxx |
| In-reply-to: | <200409161310.i8GDAnqG005123@xxxxxxxxxxxxxxxxx> |
| References: | <200409161310.i8GDAnqG005123@xxxxxxxxxxxxxxxxx> |
| Sender: | netdev-bounce@xxxxxxxxxxx |
| User-agent: | Mozilla/5.0 (Windows; U; Windows NT 5.0; en-US; rv:1.7.1) Gecko/20040707 |
Leonid Grossman wrote: We can dream about benefits of huge MTUs, but the reality is that moving beyond 9k MTU is years away. Reasons - mainly infrastructure, plus MTU above ~10k may loose checksum protection (granted, this depends whether the errors are simple or complex, and also this may not be a showstopper for some people). Even 9k MTU is very far from being universally accepted, eight years after our Alteon spec went out :-). One other factor is TCP congestion control, and congestion windows we obey. Most of the time, you just can't send that much. thanks, Nivedita |
| <Prev in Thread] | Current Thread | [Next in Thread> |
|---|---|---|
| ||
| Previous by Date: | Re: Kernel connector - userspace <-> kernelspace "linker"., Evgeniy Polyakov |
|---|---|
| Next by Date: | [PATCH][resend] Update e1000 to use module_param(), Roland Dreier |
| Previous by Thread: | RE: The ultimate TOE design, Leonid Grossman |
| Next by Thread: | RE: The ultimate TOE design, Leonid Grossman |
| Indexes: | [Date] [Thread] [Top] [All Lists] |