[tahoe-dev] starting to use volunteergrid1 -- survivability and performance measurement

[Zooko O'Whielacronx]

Following-up to my own post:

On Fri, May 27, 2011 at 5:19 PM, Zooko O'Whielacronx <zooko at zooko.com> wrote:
> Share ID                Nickname                                                Node ID
> --------                --------                                                -------
> seq1-7ahx-sh0   xxxxxxxxxxx                                     m2sbabta5mc22ijtfkcidmpwbuxbp3bf
> seq1-7ahx-sh1   xxxxxxxxxxxx                                    bwxr6cbkutq2s5slg7dhh5ukejn6567i
> seq1-7ahx-sh2   xxxxx                                           63hhqm64talajpn5zu343ui3bhi4tfcc
> seq1-7ahx-sh3   xxxxxxxxxxxxxxxxxxxxxxxxxxxxx   xlnzwmko5inqjs4krxk35yj2sqjtb6ma
> seq1-7ahx-sh4   xxxxxxxxxxxxxxx                         x2l47knnwwvfxonturznfh3n5hh7g7a2
> seq1-7ahx-sh5   xxxxxxxxxxxxxxxxxxxxxxxx                iutxbkwuy5ekowafh7r4nrxl3v446irx
> seq1-7ahx-sh6   xxxxxxxxxxxx                                    hpdlcilo7cjag7c4ffzehavh2wdv2zrp
> seq1-7ahx-sh7   xxxxxxxxxxx                                     ysxpoywqf37dpaxxuonoqc3bod24ic6w
> seq1-7ahx-sh8   xxxxxxxxxxxxxxx                         gtnnoahwi5ukb4ztjs3un7mh7joopb33
> seq1-7ahx-sh9   xxxxxxxxxxx                                     m2sbabta5mc22ijtfkcidmpwbuxbp3bf

Since I edited the nicknames out it isn't as clear, but the directory
is now split into 10 separate shares, any 3 of which are sufficient to
completely recover it, and each share it stored on a different server.
There are 10 different servers holding one share each of the
directory. I believe each of the ten of them to be in a different
network location. Two of them share an owner/operator.

That's why I say that this grid has high survivability.

Regards,

Zooko