Nimble Upgrades Adaptive Flash System, Adds REST APIs

Reliable service-level agreements in hybrid storage weren't available until recently, mainly because guaranteeing anything in IT hardware is a risky business.

Flash storage continues to replace spinning disk drives at a steady pace as prices diminish and the longevity of the media itself improves.

Hybrid solid-state and disk-drive storage provider Nimble Storage, taking advantage of this trend as data centers get refreshed, has announced new enterprise-grade capabilities to its Adaptive Flash platform that include service-level agreement guarantees of uptime and availability.

In years past, one of the doubting points of flash was its durability, and the data points were often justified. However, with consistent improvement in load-balancing and wear-leveling across the entire media as read and writes are made, the service use of flash has been increased greatly in only the last few years. NAND flash media disks are now regularly turning in five or more years of use, compared with two to three years previously.

Advancements to Nimble's Adaptive Flash platform include the ability to deliver an all-flash service level for applications with stringent requirements around responsiveness, an auto-flash service level for mainstream applications and a disk-only service level for applications requiring low-cost capacity, all within a single storage platform. Nimble and others have never been able to offer these previously.

Additional capabilities announced July 15 included software-based encryption and a REST API-based extensibility framework. All of the new features complement existing enterprise functionality, including granular virtual machine-level monitoring/reporting and integrated data protection.

This type of guarantee wasn't available until only recently, mainly because guaranteeing anything regarding IT hardware is a risky business. But recent major advancements in solid-state storage hardware, software and connectivity—as noted above—have changed that picture.
Nimble's unified platform contrasts with conventional best-of-breed storage systems that use various point products for specific types of use cases, creating difficult-to-control silos. Nimble claims to dynamically cater to the needs of enterprise workloads across performance, capacity and data protection to supply full storage consolidation.

New or upgraded features in the Adaptive Flash platform include the following:

--Software-based encryption, Nimble SmartSecure: Using this, storage administrators can enable data encryption in consolidation and multi-tenant storage environments at no additional cost. This new software feature enables both encryption and shredding on a per-application or per-array level, the company said.

SmartSecure does not require custom drivers or dedicated firmware to implement. Once data is compressed and encrypted by the company's flash-optimized file system, it can be securely replicated to another site with the target site understanding the encryption key, streamlining administration and preserving storage capacity, Nimble said.

SmartSecure encryption also enables service providers with multi-tenant environments to isolate and encrypt a particular client's data.

--REST APIs: These enable storage administrators to build customized reporting and workflows around storage, simplifying integration into existing customer frameworks. With the latest software release, Nimble now enables administrators to scale non-disruptively to more than 3.5PB of capacity and 160TBs of flash per cluster to optimize business-critical workloads at scale.

--Nimble InfoSight: The company's cloud-connected management and support engine now provides granular visibility into virtual machine, network and logical unit number (LUN) latencies enabling administrators to identify and resolve resource contention issues and maintain service-level agreements.

--InfoSight-VMVision: This new per-VM monitoring feature intuitively flags conditions that may cause performance issues showing throughput, IOPs (input/output operations per second) and latency levels across all layers: host, network and storage. Armed with actionable intelligence, administrators can proactively identify and address rogue or "noisy neighbor" VMs that consume more than their fair share of infrastructure resources.

For more information, go here.

Chris Preimesberger

Chris J. Preimesberger

Chris J. Preimesberger is Editor-in-Chief of eWEEK and responsible for all the publication's coverage. In his 13 years and more than 4,000 articles at eWEEK, he has distinguished himself in reporting...