I have just watched the Bryan Matthew lecture "Storage Spaces: What’s New in Windows Server 2012 R2" and I have a question regarding the 00:22:00 - 00:23:15 part of the video.
http://s22.postimg.org/z0gm8nrs1/Scale.png
http://channel9.msdn.com/Events/TechEd/NorthAmerica/2013/MDC-B218#fbid=uO1oN-588mZ
Bryan claims that CSV namespace scales linearly as the deployment scales and that Windows clustering allows to scale up to 64 nodes.
If I understand him correctly, the units (4 servers + SAS JBOD Arrays) act as cluster nodes in this setup.
My question is: what are the steps to configure such setup? How did Bryan manage to configure units to act as single cluster node? As the setup provides storage for Hyper-V computer Nodes (for VM storing?), should the Scale out file server (SOFS) be used for that purpose? If so, what about the limitation of maximum 8 nodes for SOFS?