Cisco flexes its HyperFlex hyper-converged muscles

Let battle be joined; Cisco has re-entered the storage market via the hyper-converged infrastructure appliance door, with its HyperFlex* system able to separately scale performance and capacity. It uses Springpath HALO software, as many thought it would, and places Cisco in direct competition with other HCIA vendors, …

[EKlipseMC]

Looking at the hardware this doesn't win any density crown. This is exactly what I did 6 years ago with Lefthand. Rack mounts servers with a distributed storage layer.

[Anonymous Coward]

You totally missed the point. This is not about density at all.

The correct questions are:

Did you provision Storage+Compute+Network and starting spinning your VMs in less than one hour?

Did you manage all those components in a single tool?

And then of course:

Was it cheaper than buying all those things separately?

Including the "DYI" costs part?

[Jo_seph_B]

The answer is yes. Using my UCS, Nexus and Nimble array coupled with UCS Director. Whats even better is I did it all in about 4 clicks and then sat back and got on with something else.

[EKlipseMC]

The answer with my Nutanix is YES and I did 4 nodes in 2U! Oh and I'm hypervisor agnostic!

[Anonymous Coward]

Nutella? You surely missed the cheap part....

[Anonymous Coward]

So a HX220C is an 1U single node appliance with storage......

A Nutanix or VxRail is a 2U quad node appliance with storage....

Which means a 4 node HyperFlex solution will be 4U...... Yeah, something's not adding up just quite yet.....

But knowing how Cisco operate, they'll drop their pants on price just to get market share (similar to their UCS product)

[Bcraig]

Re:

220 is 1u, 240 is 2u, two fabric interconnects are 2u. Lot of flexibility between compute vs capacity nodes and if the customer has B200M4 nodes they can reuse those. Cisco will definitely show up to compete with this price wise.

[steinweo]

What is Hyperconvergence

Good marketing, but scaling compute and storage independantly is not hyperconvergence !!!

Here is commonly accepted definition : in a hyperconvergence environment all elements of the storage, compute and network components are optimized to work together on a single commodity appliance from a single vendor. What really matters is: "a single commodity appliance".

What Cisco is referring about looks more like to me to hyperscale : "hyperscale is the ability of an architecture to scale appropriately as increased demand is added to the system. This typically involves the ability to seamlessly provision and add compute, memory, networking, and storage resources to a given node or set of nodes that make up a larger computing, distributed computing, or grid computing environment."