Are storage systems and servers becoming one and the same?
In their August Storage Consumption Model , IDC’s Richard Villars talks about several intriguing trends that are changing the way storage is designed, distributed and used.
The one that caught my eye is what he calls the “serverization” of storage platforms – the addition of computing power to storage systems, especially for mass-scale clustered storage data factories.
A parallel trend is that mainstream server storage capacity is growing incredibly fast. Storage capacity (often measured in terabytes) now stands next to processing power as cornerstone specifications for most servers.
It begs the question: is the difference between a server and a storage system becoming insignificant?
I think we need a new name for systems that blur the line. Let’s call them Stervers.
Thoughts? Is this a substantial shift, or just a reinvention of the same old processing/storage partnership?
I think this is an old idea, renewed! I had an opportunity in the HPC space to play with dCache (http://www.dcache.org/). It leveraged internal storage of compute nodes, as a “cloud” of accessible storage.
I always go back to the idea that if a “Server” was so good at processing IO as a target (rather then initiator), then HDS, IBM, EMC, etc would be running disks behind “Servers” and not purpose built Storage Controllers.
In most cases, Server based solutions, require extra copies of a data set in order to protect from single points of failure (look at LeftHands network raid as an example).