On Tue, 20 Jan 2009, Stas Oskin wrote: > So basically the data stored on 2 nodes, but their coordination requires a > shared block device? Because they do seem to speak about specific storage > nodes - but I might not understand this correctly. AFAIK, on Lustre the data is stored on the shared block device, if one wants to be able to fail over the service to another node. But the shared block device may be a distributed mirrored block device (like DRBD) which mirrors each data block as it is written to its peer node. In such a configuration the data is actually stored on both nodes in the failover pair. My guess is that this is not a common configuration for production use. Regards, Jerker Nyberg.
