Startup Aims to Upend Enterprise Storage with MLC Flash-Based Systems

By Michael Feldman

August 14, 2012

Silicon Valley startup Skyera has unveiled a solid state storage system that the company believes will be a game changer for enterprise storage. The product, known as Skyhawk, will use consumer-grade multi-level cell (MLC) flash memory as the basis for a bulk storage solution at a price point of less than $3 per gigabyte. As such, it is designed to compete head-to-head against hard disk-based storage, while offering the superior performance, density and energy efficiency of flash memory.

Although solid state enterprise storage is growing rapidly, it still represents just a small fraction of the $32 billion storage market, mainly because it can’t compete on a cost-capacity basis with spinning disks. The current flash-based solutions today tend to be used as tier 0 storage caches or for data-bound applications where the higher IOPS warrants more expensive capacity. Skyera wants to change that dynamic by employing the cheapest flash in the industry — consumer MLC NAND — and go after the heart of the market. “We are going to challenge the status quo of enterprise storage,” says Skyera’s marketing VP, Tony Barbagallo.

The driving force and behind Skyera is CEO Radoslav Danilak, whose resume includes the founding of SandForce, a flash controller startup that was subsequently sold to LSI for $370 million. He also did a stint at NVIDIA as a chip architect on the Tesla GPU products. Danilak, along with SandForce alum Rod Mullendore, founded Skyera in 2010, with the mission to bring next-generation flash-based systems to the enterprise.

Also on the team are industry veterans Ken Takeuchi (NAND flash designer at Toshiba), Frankie Roohparvar (flash development exec at Micron), Dave Martin (CEO Hitachi Data Systems), Alessandro Fin (product development PNY, SMART Modular Technologies), Roy D Cruz (networking and storage architect Cisco, Brocade, Andiamo), and Dave Ferretti (sales exec at Zetta, StoredIQ, EVault).

That diversity of talent (chip development, storage system design, and networking expertise) was brought together to build Skyhawk. In a nutshell, the new offering wraps a series of “life amplification” technologies around consumer-grade MLC flash so that it behaves more like reliable enterprise-grade SLC (single-level cell) flash. . The company is claiming a Skyhawk setup will be able to deliver a complete SAN system at under $3 per gigabyte and under a $1 per gigabyte with compression and deduplication. That would provide price parity with the HDD-based bulk storage solutions.

Building a general-purpose flash-based system as inexpensive as one using high-capacity SATA drives has never been done before. The most critical challenge for flash in the datacenter is its limited lifetime, especially in regard to writing data. SLC technology can support about 100,000 writes for a given bit before a failure can be expected. That’s is about 50 times as many writes as can be coaxed from consumer MLC. Unfortunately, SLC costs two to three times as much, which means customers pay a significant premium for the extra reliability.

Enterprise MLC (eMLC) is basically a compromise between the SLC and MLC and has been adopted by a number of SSD and flash storage vendors. These companies bring eMLC up to SLC-level robustness by layering on extra flash controller smarts that optimizes write behavior (ECC, wear leveling, caching, etc.). But neither SLC or eMLC can provide the basis for a cost-competitive solution for hard disk storage.

So Skyera went one step further and chose vanilla MLC for their solution. But to give MLC-based storage the 5-year lifespan required for enterprise duty, a number of technologies had to be included to compensate for its natural lack of robustness. “There is no single magic bullet to do that,” Danilak told HPCwire.

First, Skyera built an industrial-strength flash controller that employs adaptive ECC algorithms (patent-pending) to correct for errant bits and optimizes write behavior at the chip level to significantly reduce oxide wear on the NAND devices. Further, they invented a proprietary RAID technology to protect the data in such a way that minimizes extra writes. Finally, the engineers added in-line compression and deduplication to further reduce the write (and read) load on the flash chips. According to Danilak, in aggregate they were able to increase the lifetime of the underlying MLC flash a 100-fold, which allowed them to reach their 5-year usage goal.

Performance is rated at up to a million IOPS per node, which is 25 times better than that of a spinning disk. In general MLC performance is inferior to SLC, but Danilak maintains the difference is less than usually thought, and since Skyera does write optimization, error correction, and compression/dedupe all in hardware, they don’t lose as much performance as a solution that relies on a software assist. In any case, even cheap flash is going to be a lot faster than a spinning disk.

And because Skyera is using the latest 19/20nm MLC technology, their solution is extra dense. A Skyhawk box is able to house up to 44 TB of usable storage (48 TB actual) into a half-depth 1U form factor. That’s probably the densest flash-based storage enclosure on the market and more than 100 times as compact as an HDD system of similar capacity, even using the latest 3TB SATA drives.

To live up to its enterprise-ready credentials, Skyhawk incorporates a software stack expected of typical SAN systems, including snapshots, clones, storage QoS, multi-path support, consistency groups, performance monitoring, LUN management, thin provisioning, and dynamic resizing.

Skyera also integrates internal networking to relieve the communication bottleneck caused by the high bandwidth flash. A Skyhawk box is equipped with 40 Gigabit Ethernet and three 10GbE ports that can hook the storage directly to servers or to intermediate Ethernet switches.

For its initial debut, Skyhawk will come in three configurations 12TB ($48,000), 22 TB ($77,000) and 44TB ($131,000). If the customer chooses to buy the extra capacity enabled by the compression/dedupe feature, disk capacity can be more than doubled. For example, the 44TB configuration becomes a 100TB box. The compression and deduplication processing is actually always turned on since it’s used to extend the lifetime of the MLC flash, but Skyera will charge a 20 to 30 percent premium over the base price if the customer wants to access the extra capacity.

As long as the application needs those extra terabytes, paying that premium is a no-brainer, since it can drive the cost per gigabyte down below a $1. But some customers might get a little tweaked that they’re essentially paying for the same feature twice.

Although Skyhawk has the potential to become a breakthrough bulk storage product, a handful of other vendors are offering MLC-based solutions, including STEC, SMART Storage Systems and Pure Storage. Like Skyera, they are using a variety of error correction and write optimization schemes to improve the usable lifetime of the flash storage. From Danilak’s perspective, he thinks Skyera has the edge until his competitors “figure out how to get write amplification technology like we have.”

General availability for Skyhawk is planned for the first quarter of 2013, but for select customers, the company has an early access program that will make systems available in Q3 2012. And if you’re really eager to see one in action, Skyera will be demonstrating Skyhawk next week at the Flash Memory Summit in Santa Clara, California.

Subscribe to HPCwire's Weekly Update!

Be the most informed person in the room! Stay ahead of the tech trends with industry updates delivered to you every week!

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion XL — were added to the benchmark suite as MLPerf continues Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing power it brings to artificial intelligence.  Nvidia's DGX Read more…

Call for Participation in Workshop on Potential NSF CISE Quantum Initiative

March 26, 2024

Editor’s Note: Next month there will be a workshop to discuss what a quantum initiative led by NSF’s Computer, Information Science and Engineering (CISE) directorate could entail. The details are posted below in a Ca Read more…

Waseda U. Researchers Reports New Quantum Algorithm for Speeding Optimization

March 25, 2024

Optimization problems cover a wide range of applications and are often cited as good candidates for quantum computing. However, the execution time for constrained combinatorial optimization applications on quantum device Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at the network layer threatens to make bigger and brawnier pro Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HBM3E memory as well as the the ability to train 1 trillion pa Read more…

MLPerf Inference 4.0 Results Showcase GenAI; Nvidia Still Dominates

March 28, 2024

There were no startling surprises in the latest MLPerf Inference benchmark (4.0) results released yesterday. Two new workloads — Llama 2 and Stable Diffusion Read more…

Q&A with Nvidia’s Chief of DGX Systems on the DGX-GB200 Rack-scale System

March 27, 2024

Pictures of Nvidia's new flagship mega-server, the DGX GB200, on the GTC show floor got favorable reactions on social media for the sheer amount of computing po Read more…

NVLink: Faster Interconnects and Switches to Help Relieve Data Bottlenecks

March 25, 2024

Nvidia’s new Blackwell architecture may have stolen the show this week at the GPU Technology Conference in San Jose, California. But an emerging bottleneck at Read more…

Who is David Blackwell?

March 22, 2024

During GTC24, co-founder and president of NVIDIA Jensen Huang unveiled the Blackwell GPU. This GPU itself is heavily optimized for AI work, boasting 192GB of HB Read more…

Nvidia Looks to Accelerate GenAI Adoption with NIM

March 19, 2024

Today at the GPU Technology Conference, Nvidia launched a new offering aimed at helping customers quickly deploy their generative AI applications in a secure, s Read more…

The Generative AI Future Is Now, Nvidia’s Huang Says

March 19, 2024

We are in the early days of a transformative shift in how business gets done thanks to the advent of generative AI, according to Nvidia CEO and cofounder Jensen Read more…

Nvidia’s New Blackwell GPU Can Train AI Models with Trillions of Parameters

March 18, 2024

Nvidia's latest and fastest GPU, codenamed Blackwell, is here and will underpin the company's AI plans this year. The chip offers performance improvements from Read more…

Nvidia Showcases Quantum Cloud, Expanding Quantum Portfolio at GTC24

March 18, 2024

Nvidia’s barrage of quantum news at GTC24 this week includes new products, signature collaborations, and a new Nvidia Quantum Cloud for quantum developers. Wh Read more…

Alibaba Shuts Down its Quantum Computing Effort

November 30, 2023

In case you missed it, China’s e-commerce giant Alibaba has shut down its quantum computing research effort. It’s not entirely clear what drove the change. Read more…

Nvidia H100: Are 550,000 GPUs Enough for This Year?

August 17, 2023

The GPU Squeeze continues to place a premium on Nvidia H100 GPUs. In a recent Financial Times article, Nvidia reports that it expects to ship 550,000 of its lat Read more…

Shutterstock 1285747942

AMD’s Horsepower-packed MI300X GPU Beats Nvidia’s Upcoming H200

December 7, 2023

AMD and Nvidia are locked in an AI performance battle – much like the gaming GPU performance clash the companies have waged for decades. AMD has claimed it Read more…

DoD Takes a Long View of Quantum Computing

December 19, 2023

Given the large sums tied to expensive weapon systems – think $100-million-plus per F-35 fighter – it’s easy to forget the U.S. Department of Defense is a Read more…

Synopsys Eats Ansys: Does HPC Get Indigestion?

February 8, 2024

Recently, it was announced that Synopsys is buying HPC tool developer Ansys. Started in Pittsburgh, Pa., in 1970 as Swanson Analysis Systems, Inc. (SASI) by John Swanson (and eventually renamed), Ansys serves the CAE (Computer Aided Engineering)/multiphysics engineering simulation market. Read more…

Choosing the Right GPU for LLM Inference and Training

December 11, 2023

Accelerating the training and inference processes of deep learning models is crucial for unleashing their true potential and NVIDIA GPUs have emerged as a game- Read more…

Intel’s Server and PC Chip Development Will Blur After 2025

January 15, 2024

Intel's dealing with much more than chip rivals breathing down its neck; it is simultaneously integrating a bevy of new technologies such as chiplets, artificia Read more…

Baidu Exits Quantum, Closely Following Alibaba’s Earlier Move

January 5, 2024

Reuters reported this week that Baidu, China’s giant e-commerce and services provider, is exiting the quantum computing development arena. Reuters reported � Read more…

Leading Solution Providers

Contributors

Comparing NVIDIA A100 and NVIDIA L40S: Which GPU is Ideal for AI and Graphics-Intensive Workloads?

October 30, 2023

With long lead times for the NVIDIA H100 and A100 GPUs, many organizations are looking at the new NVIDIA L40S GPU, which it’s a new GPU optimized for AI and g Read more…

Shutterstock 1179408610

Google Addresses the Mysteries of Its Hypercomputer 

December 28, 2023

When Google launched its Hypercomputer earlier this month (December 2023), the first reaction was, "Say what?" It turns out that the Hypercomputer is Google's t Read more…

AMD MI3000A

How AMD May Get Across the CUDA Moat

October 5, 2023

When discussing GenAI, the term "GPU" almost always enters the conversation and the topic often moves toward performance and access. Interestingly, the word "GPU" is assumed to mean "Nvidia" products. (As an aside, the popular Nvidia hardware used in GenAI are not technically... Read more…

Shutterstock 1606064203

Meta’s Zuckerberg Puts Its AI Future in the Hands of 600,000 GPUs

January 25, 2024

In under two minutes, Meta's CEO, Mark Zuckerberg, laid out the company's AI plans, which included a plan to build an artificial intelligence system with the eq Read more…

Google Introduces ‘Hypercomputer’ to Its AI Infrastructure

December 11, 2023

Google ran out of monikers to describe its new AI system released on December 7. Supercomputer perhaps wasn't an apt description, so it settled on Hypercomputer Read more…

China Is All In on a RISC-V Future

January 8, 2024

The state of RISC-V in China was discussed in a recent report released by the Jamestown Foundation, a Washington, D.C.-based think tank. The report, entitled "E Read more…

Intel Won’t Have a Xeon Max Chip with New Emerald Rapids CPU

December 14, 2023

As expected, Intel officially announced its 5th generation Xeon server chips codenamed Emerald Rapids at an event in New York City, where the focus was really o Read more…

IBM Quantum Summit: Two New QPUs, Upgraded Qiskit, 10-year Roadmap and More

December 4, 2023

IBM kicks off its annual Quantum Summit today and will announce a broad range of advances including its much-anticipated 1121-qubit Condor QPU, a smaller 133-qu Read more…

  • arrow
  • Click Here for More Headlines
  • arrow
HPCwire