[geeks] Fwd: [IP] Interesting speculation on the tech behind gmail

[Sandwich Maker]

"From: Brian Dunbar <brian.dunbar at plexus.com>
"
"This seemed 'Geeks List' worthy.
"
"Begin forwarded message:
"
"> From: Dave Farber <dave at farber.net>
">
">
"> From: Suresh Ramasubramanian <suresh at hserus.net>
">
"> http://blog.topix.net/archives/000016.html
">
"> April 04, 2004
"> The Secret Source of Google's Power
[]
"> The most obvious challenge is the storage. You can't lose people's 
"> email, and you don't want to ever be down, so data has to be 
"> replicated. RAID is no good; when a disk fails, a human needs to 
"> replace the bad disk, or there is risk of data loss if more disks 
"> fail. One imagines the old ENIAC technician running up and down the 
"> isles of Google's data center with a shopping cart full of spare disk 
"> drives instead of vacuum tubes. RAID also requires more expensive 
"> hardware -- at least the hot swap drive trays. And RAID doesn't handle 
"> high availability at the server level anyway.

this is so glibly wrong.

sun's disksuite s/w - which is ten-plus years old now - implements
raid 0+1 as a stripe of mirrors.  you can lose up to half your disks
without impact as long as no two failures are on the same mirror.

and with any raid level you can define a hot-spare pool, with afaik no
limit to the number of disks you put in it.  raid5 for example is thus
protected against multi-disk failures.

okay, this is only a single datapoint, and at least some of google's
argument still stands.

"> Google has 100,000 servers.
[]
"> Google is a company that has built a single very large, custom 
"> computer. It's running their own cluster operating system. They make 
"> their big computer even bigger and faster each month, while lowering 
"> the cost of CPU cycles. It's looking more like a general purpose 
"> platform than a cluster optimized for a single application.

doesn't this put them somewhere close to the top of the supercomputer
list?  i mean, 100k 1GHz cpus, even if they are p3s...  and 50tb ram?
________________________________________________________________________
Andrew Hay                                  the genius nature
internet rambler                            is to see what all have seen
adh at an.bradford.ma.us                       and think what none thought