How to size a Scale-out File Server
hi,
we looking implement 2-node 2012 r2 scale-out file server cluster (using sas jbod enclosure) primary purpose of storing vhd files accessed 4-node 2012 r2 hyper-v cluster using 10 gigabit ethernet (no rdma). our environment can characterised having large number of idle vms experience sporadic, low intensity use (this *not* vdi environment). have 2 questions.
1) extent ram consideration sofs servers? can't find documentation suggest there benefits gained having more ram in sofs servers don't know if should go 8/16/32/64+ gb ram in each of nodes.
2) need keep costs down, don't think rdma / smb-direct nics going within our reach. should have 2 * dual-port 10 gbps nics in both sofs & hyper-v boxes?
unless vms read-intensive , you're going deploy csv cache memory requirement serving idle vms can pretty low. ram cheap these days going less 16gb per node not sound reasonable. sample, see:
windows server 2012 file server tip: enable csv caching on scale-out file server clusters
http://blogs.technet.com/b/josebda/archive/2012/11/14/windows-server-2012-file-server-tip-enable-csv-caching-on-scale-out-file-server-clusters.aspx
http://blogs.msdn.com/b/clustering/archive/2013/07/19/10286676.aspx
hyper-v on smb: scale-out file server , storage spaces
http://www.thomasmaurer.ch/2013/08/hyper-v-over-smb-scale-out-file-server-and-storage-spaces/
hope helped :)
Windows Server > High Availability (Clustering)
Comments
Post a Comment