[lkml]   [1999]   [Mar]   [4]   [last100]   RSS Feed
Views: [wrap][no wrap]   [headers]  [forward] 
Messages in this thread
SubjectRe: OSDI paper - IO-Lite: A Unified I/O Buffering and Caching System

> > Sure, there's still complexity to be debated -- but don't dismiss the
> > paper out of hand before you've read the details... it does address a lot
> > of issues.
> The checksum one is a seperate issue. Let me explain that since its a bit
> of a casual thing to say about someones research work without justification.
> Right now we compute the checksum in kernel space.
> If the checksum is wrong the other end will drop the frames
> o Therefore the checksum is not trust critical data.
> This has an important implication, you can take the whole clever page
> management model being discussed and replace it with a single change
> send_checksumis(fd, buf, len, flags)
> since you know that the application lying is not a security
> violation.

There's one thing I do not understand: you have different packet sizes
to each hosts (potentialy), you can send partial packets somethimes,
etc. Checksums of what are you going to cache? Are you going to split
file into 256b chunks, compute checksums of that and if someone
requests 1025 bytes packet, take 4 checksums, combine them and add
single byte?

The best software in life is free (not shareware)! Pavel
GCM d? s-: !g p?:+ au- a--@ w+ v- C++@ UL+++ L++ N++ E++ W--- M- Y- R+

To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to
Please read the FAQ at

 \ /
  Last update: 2005-03-22 13:50    [W:0.050 / U:4.504 seconds]
©2003-2018 Jasper Spaans|hosted at Digital Ocean and TransIP|Read the blog|Advertise on this site