WIKIPEDIA defines Composability as: "...a highly composable system [that] provides components that can be selected and assembled in various combinations to satisfy specific user requirements."
HPE Corporate Website states: "In composable infrastructure, compute, storage, and networking resources are abstracted from their physical locations, and can be managed by software, through a web-based interface. Composable infrastructure makes data center resources as readily available as cloud services, and is the foundation for private and hybrid cloud solutions." HPE clarifies that Composable infrastructure is "particularly useful when you need to keep running some applications on bare metal rather than on virtualized infrastructure." So the question is this: Does HPE Synergy satisfy their own definition of Composable Infrastructure when it comes to Dynamic/Scalable GPUs?
Here are 12 Key Questions you can use when starting your evaluation of competing Composable platforms:
1- Does it scale GPU's across servers and racks?
2- Can it dynamically allocate (bare-metal) GPUs?
3- Does it limit GPU performance w/ Ethernet?
4- Can it dynamically orchestrate GPU-centric work-loads?
5- Can it leverage legacy, bare metal HW & SW?
6- Does it allow Mission Critical applications to remain on Legacy Equipment?
7- Does it lock me into an OEM vendor?8- Is it (truly) SW agnostic?
9- Does it enable bare metal orchestration - across racks?
10- Does it rely on a hypervisor to allocate resources?
11- Does it rely on Ethernet? Or is it PCIe based?
12- Does it limit my NVME M.2 deployment by relying on Ethernet?
Today, LIQID's CEO, Sumit Puri discusses ONE MORE key difference between LIQID Composable vs Synergy - Dynamic GPUs.
#composableIT #HPEsynergy #dynamicGPU