* [9fans] Google's GFS
@ 2003-09-30 3:06 Skip Tavakkolian
2003-09-30 3:37 ` mirtchov
0 siblings, 1 reply; 6+ messages in thread
From: Skip Tavakkolian @ 2003-09-30 3:06 UTC (permalink / raw)
To: 9fans
Interesting paper. I'm curious why the authors didn't cite
ken fs, or venti/fossil. Surely they'ld have heard of them.
http://www.cs.rochester.edu/sosp2003/papers/p125-ghemawat.pdf
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Google's GFS
2003-09-30 3:06 [9fans] Google's GFS Skip Tavakkolian
@ 2003-09-30 3:37 ` mirtchov
2003-09-30 4:08 ` Skip Tavakkolian
2003-09-30 4:19 ` Russ Cox
0 siblings, 2 replies; 6+ messages in thread
From: mirtchov @ 2003-09-30 3:37 UTC (permalink / raw)
To: 9fans
> Interesting paper. I'm curious why the authors didn't cite
> ken fs, or venti/fossil. Surely they'ld have heard of them.
>
> http://www.cs.rochester.edu/sosp2003/papers/p125-ghemawat.pdf
ten years have passed since Ken's file server -- that's plenty enough
to start reinventing it. i doubt they've heard of fossil -- google
hasn't discovered plan9 yet.
;)
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Google's GFS
2003-09-30 3:37 ` mirtchov
@ 2003-09-30 4:08 ` Skip Tavakkolian
2003-09-30 4:19 ` Russ Cox
1 sibling, 0 replies; 6+ messages in thread
From: Skip Tavakkolian @ 2003-09-30 4:08 UTC (permalink / raw)
To: 9fans
> ten years have passed since Ken's file server -- that's plenty enough
> to start reinventing it. i doubt they've heard of fossil -- google
> hasn't discovered plan9 yet.
Hmm... Rob Pike was one of the reviewers of the paper, according to them.
(i am become boyd, annoyer of the worlds)
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Google's GFS
2003-09-30 3:37 ` mirtchov
2003-09-30 4:08 ` Skip Tavakkolian
@ 2003-09-30 4:19 ` Russ Cox
2003-09-30 4:25 ` okamoto
2003-09-30 6:27 ` Skip Tavakkolian
1 sibling, 2 replies; 6+ messages in thread
From: Russ Cox @ 2003-09-30 4:19 UTC (permalink / raw)
To: 9fans
> Interesting paper. I'm curious why the authors didn't cite
> ken fs, or venti/fossil.
Why would they?
Maybe I'm missing something fundamental, but I really don't think
that either Plan 9 file server is anything like Google's file system.
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Google's GFS
2003-09-30 4:19 ` Russ Cox
@ 2003-09-30 4:25 ` okamoto
2003-09-30 6:27 ` Skip Tavakkolian
1 sibling, 0 replies; 6+ messages in thread
From: okamoto @ 2003-09-30 4:25 UTC (permalink / raw)
To: 9fans
>> Interesting paper. I'm curious why the authors didn't cite
>> ken fs, or venti/fossil.
>
> Why would they?
>
> Maybe I'm missing something fundamental, but I really don't think
> that either Plan 9 file server is anything like Google's file system.
Yeah! those two have quite diffrent goal.
Kenji
^ permalink raw reply [flat|nested] 6+ messages in thread
* Re: [9fans] Google's GFS
2003-09-30 4:19 ` Russ Cox
2003-09-30 4:25 ` okamoto
@ 2003-09-30 6:27 ` Skip Tavakkolian
1 sibling, 0 replies; 6+ messages in thread
From: Skip Tavakkolian @ 2003-09-30 6:27 UTC (permalink / raw)
To: 9fans
> Maybe I'm missing something fundamental, but I really don't think
> that either Plan 9 file server is anything like Google's file system.
I find it hard to believe that a comparative analysis wouldn't have included
at least one of the Plan9 fs'es. To my (novice) thinking some concepts sounded
similar: append-only writes, namespace management, snapshots, two tiered
architecture.
^ permalink raw reply [flat|nested] 6+ messages in thread
end of thread, other threads:[~2003-09-30 6:27 UTC | newest]
Thread overview: 6+ messages (download: mbox.gz / follow: Atom feed)
-- links below jump to the message on this page --
2003-09-30 3:06 [9fans] Google's GFS Skip Tavakkolian
2003-09-30 3:37 ` mirtchov
2003-09-30 4:08 ` Skip Tavakkolian
2003-09-30 4:19 ` Russ Cox
2003-09-30 4:25 ` okamoto
2003-09-30 6:27 ` Skip Tavakkolian
This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).