At Tue, 8 Nov 2011 22:53:41 -0000, Matthew Law wrote: > > > I was just reviewing the sheepdog docs and looking at the architecture > overview diagram here: Thanks a lot! > > https://github.com/collie/sheepdog/wiki/Sheepdog-Design > > made me wonder if all I/O to the gateway is through one node in the > cluster, making that node a bottleneck in a large cluster? Different VMs can use different gateways, so if there are many VMs, the gateway cannot be a bottleneck. Our goal is not to provide the combined bandwidth of many disks to one VM, but rather to achieve scalability; Sheepdog provides enough bandwidth to each VM even if many VMs use Sheepdog at the same time. Thanks, Kazutaka |