What are you looking for ?
Infinidat
Articles_top

Scale Computing: Integration of Flash-Enabled Automated Storage Tiering Into HC3 Platform

Optimizes hyper-convergence by intelligently moving data based on workload priority and usage patterns.

Scale Computing, Inc. announced the integration of flash-enabled automated storage tiering into its HC3 platform.

SCALE_hc3_2This update to company’s converged HC3 system adds hybrid storage including SSD and spinning disk with HyperCore Enhanced Automated Tiering (HEAT). Firm’s HEAT technology uses a combination of built-in intelligence, data access patterns, and workload priority to automatically optimise data across disparate storage tiers within the cluster.

Hyper-convergence is nothing if not about simplicity and cost. But it is also about performance, especially in the SMB to mid-size enterprises where most, if not all workloads will simultaneously run on a single cluster of nodes,” said Arun Taneja, founder and consulting analyst, Taneja Group. “Introducing flash into a HDD based system is easy; the question is how do you do it so that it maintains low cost and simplicity while boosting performance. This is what Scale has done in these new models. The only decision the IT admin and the business user need to make is to determine the importance of the application and its priority. After that flash is invisible to them. The only thing visible is better application performance. This is how it should be.

Click to enlarge.
SCALE_hc3-TABL1

HC3 platform brings storage, servers, virtualisation, and HA together in a single system. With no virtualisation software to license and no external storage to buy, HC3 solutions lower out-of-pocket costs and simplify the infrastructure needed to keep applications optimised and running.

This update to the HC3 HyperCore storage architecture combines firm’s HEAT technology with SSD-hybrid nodes that add a new tier of flash storage to new or existing HC3 clusters. HEAT technology combines intelligent automation with simple, granular tuning parameters to further define flash storage utilisation on a per virtual disk basis for optimal performance.

Through an easy-to-use slide bar, users can optionally tune flash priority allocation to more effectively utilise SSD storage where needed from no flash at all for a virtual disk, to virtually all flash by ‘turning it to 11.’ Every workload is different and even a small amount of flash prioritisation tuning, combined with the automated, intelligent I/O mapping, can have a big impact on the overall performance of flash storage in the HC3 cluster.

Unlike other storage systems that use flash storage only for disk caching, company’s HC3 virtualisation platform adds flash capacity and performance to the total storage pool. Customers will immediately and automatically take advantage of the flash I/O benefits without any special knowledge about flash storage.

Like any organisation, we have applications that need maximum performance, applications where performance isn’t a priority, and still others where higher performance would be helpful but not mission critical,” said Mike O’Neil, director, IT, Hydradyne LLC. “But unlike some organisations, we weren’t in a position to dedicate the resources needed to support these differing workloads. With Scale, we will have an architecture in place that immediately and automatically allows VMs to take advantage of flash storage without us even thinking about storage or virtualisation configuration.

HyperCore architecture simplifies VM storage management without VSAs (Virtual Storage Appliances), SAN protocols and file system overhead. VMs have direct access to virtual disks, allowing all storage operations to occur as efficiently as possible. HyperCore applies logic to stripe data across multiple physical storage devices in the cluster to aggregate capacity and performance. The HyperCore backplane network lets any node and any VM access any disk and is performance optimised to scale as nodes are added.

With this release, we change the economics and maximise the value of flash storage for all customer segments, from the SMB to the enterprise,” said Jeff Ready, CEO, Scale Computing. “Many vendors use a flash write-cache as a way to mask otherwise sluggish performance. Instead, we have built an architecture that intelligently adjusts to the changing workloads in the datacentre, to maximise the performance value of flash storage in every environment.

The company is deploying its HEAT technology across the HC3 product line and is introducing a flash storage tier as part of its HC2150 and HC4150 appliances. Available in 4 or 8-drive units, Scale’s latest offerings include one 400 or 800GB SSD with three NL-SAS HDD in 1-6TB capacities and memory up to 256GB, or two 400 or 800GB SSD with six NL-SAS HDD in 1-2TB capacities and up to 512GB memory respectively. Network connectivity for either system is achieved through two 10GbE SFP+ ports per node.

These products can be used to form new clusters, or they can be added to existing HC3 clusters. Existing workloads on those clusters will automatically utilise the new storage tier when the new nodes are added.

Articles_bottom
AIC
ATTO
OPEN-E