[Beowulf] Redundant Array of Independent Memory - fork(Re: Checkpointing using flash)

Andrew Holway andrew.holway at gmail.com
Sun Sep 23 07:02:20 PDT 2012


2012/9/23 Alan Louis Scheinine <alscheinine at tuffmail.us>:
> Jim Lux wrote that one giant [distributed] memory has scalability
> problems from physical distance reasons.  Yes indeed.  Simply to clarify,
> I was refering to a specific niche in parameter space (physical and
> programmatic) associated with programs using file I/O.  That is to say,
> there is a realm for which the idea of Andrew Holway is appropriate.

I was not thinking of one giant distributed memory but more of a way
of making small sub clusters fault tolerant.

I made a sketch :) http://bit.ly/TlkHpH

There are systems that have memory mirroring and other fancy tech to
make single machines fault tolerant. Its just these systems are
somewhat expensive at the moment due to low demand and they are not on
our radar 'cos we're not banks.

http://en.wikipedia.org/wiki/Redundant_array_of_independent_memory


More information about the Beowulf mailing list