Hi,
The GFS is a file system. The info from google DBAs covers
very different set of issues.
The gfs paper: labs.google.com/papers/gfs-sosp2003.pdf
thanks,
Jacek
[EMAIL PROTECTED] wrote:
Hi,
There is a paper written about the Google File System that describes
most of this. Not things like how many DBAs they have, but it does
describe
how they do failover and fault recovery. You can probably google it to
find it.... ;)
Michelle Miller
On Wed, 22 Nov 2006, Ani Thakar wrote:
On Wed, 22 Nov 2006, Jacek Becla wrote:
- I'm planning to meet with google DBAs over lunch to talk more
if it was lunch today i guess it's too late, but it wd be really
informative to find out (among other things):
- how they do the fault tolerance,
- how much redundancy they have,
- what kind of RAID they use,
- what is their disk failure rate,
- how exactly they do the failovers, and
- how many hardware (sysadmin) people they have to do the failovers.
my guess is that their budget is a lot bigger than ours ;-), so we wont
be able to quite emulate their model.
few other notes picked from discussions with user community
- mentioned by several people independently:
- very bad experience with expensive fancy hardware:
- high expectations, breaks more often then expected,
cheap hardware catches up quickly (the fancy hardware
no longer fancy after 2 years)
- much better in long term to keep it simple:
cheap hardware, build in redundancy, fault tolerance...
maybe so, but there's a big step between "cheap" hardware and "fancy"
hardware. we shd aim for something in between, i.e., more reliable
hardware vendors without going for the high-end fancy stuff, esp. when
it comes to disks.
ani
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
_______________________________________________
LSST-data mailing list
[email protected]
http://www.lsstmail.org/mailman/listinfo/lsst-data
