NetApp will launch “enterprise-class hyper-converged infrastructure” products later this year, based on SolidFire storage and unnamed compute hardware with the VMware hypervisor on-board and full control possible from vSphere.
By submitting your personal information, you agree that TechTarget and its partners may contact you regarding relevant content, products and special offers.
NetApp HCI products will be available in the last quarter of 2017.
The company would not give speeds and feeds this week, but said small (6x 480GB flash drives, 16 CPU cores), medium (6x 960GB, 24 cores) and large (6x 1.9TB, 36 cores) nodes would be available.
NetApp UK chief technologist Grant Caley, would not say where the compute element of NetApp HCI comes from, except that it is “black box, custom-built for NetApp by the vendor”.
The HCI product will come with two compute slots and four storage slots in a 2U starting configuration, scalable in 0.5U increments, said Caley, with customers able to scale either compute or storage.
Upper scaling limits will be based on those of the SolidFire storage and the VMware hypervisor, he said.
Read more on hyper-converged
- Hyper-converged infrastructure is well and truly mainstream. 451 Research finds hyper-converged all set to become the number one platform for core datacentre workloads.
- Forrester rates 12 hyper-converged infrastructure providers according to products, company strategy and market presence, with Nutanix, Simplivity and Pivot3 coming out on top.
There are plans for other hypervisor environments to be supported in future, but no hints were given about those this week.
Caley said the aim was to improve on what he called “first-generation HCI”.
“The challenge with first-generation HCI is that customers found architectural constraints, such as having to scale compute if they only needed more storage and so paying more in licence costs.
“Also, first generation was targeted at predictable workloads. This product will allow HCI to be used with enterprise mixed workloads with guaranteed performance for those workloads.”