NEC Revamps HYDRAstor for Speed, Scale

By Dennis Barker

September 9, 2008

NEC is ratcheting up its grid storage platform with new nodes designed to put more data in less space faster.

Relatively speaking, the new HYDRAstor HS8-2000, officially announced today, beats the heck out of the HS8-1000, introduced about a year ago. NEC says the second generation boosts performance by about 300 percent and improves disk capacity by 400 percent, while also consuming less power. The system is optimized for backup and archiving purposes, but nearline storage is where NEC is taking it next.

Two types of nodes comprise a HYDRAstor grid: storage nodes and accelerator nodes. NEC’s two-headed approach enables organizations to scale performance and capacity independently in order to meet their particular demands. “The nodes are the building blocks, and you can mix and match them to construct the storage system you need,” says Karen Dutch, vice president of advanced storage products for NEC America. “If you have to make adjustments to meet demand, you can add new pieces. Over the last 10 months we’ve been shipping, we’ve had many customers add capacity and performance with nodes and did it without disruption. No provisioning, no group creation. And any data that existed in the grid was automatically load balanced.”

“The HYDRAstor implementation is full of promise as an approach to storage,” says Mark Peters, an analyst with Enterprise Strategy Group who covers storage and datacenter issues. “It is scalable, self-tuning, self-healing and self-managing. Since both the processing side and the capacity side can be scaled independently, it allows for a very flexible single pool of storage. The NEC engineers were basically allowed to start from scratch — allowing them to really consider how things should be done and what should be included, rather than worry about how things have been.”

The new storage nodes, made up of 1TB SATA disk drives, have capacity of 12TB (or “effective capacity” of 157.5TB with a 20:1 deduplication rate). Accelerator nodes, based on 3GHz Xeon quad-core processors, improve data throughput with performance of 300MB per second, NEC says. A starter kit, for example, would bundle one accelerator node and two storage nodes, for raw capacity of 24TB and backup speed of about 16,500MB per second.

The nodes themselves are standard NEC servers, but the “real value-add is the software smarts” that takes HYDRAstor “beyond clustered storage systems,” Dutch says. “It’s a distributed system, DynamicStor, that gets loaded on all the nodes at set-up and creates a self-managing, self-healing, self-expanding infrastructure. The nodes are self-aware. When you add a node, capacity is automatically discovered. Our software provides the intelligence to manage it all, and from one central location.”

NEC describes its storage platform as “self-evolving,” meaning owners of the older version can plug the new nodes into their current grid and move up to a higher level of performance without having to replace any hardware or software or do any data migration or provisioning.

Distributed Reduction and Resiliency

NEC says the key to its efficient data storage — how it gets more data on a platter than competing approaches — is its brand of data deduplication (which it calls DataRedux). Deduplication, simply put, reduces storage needs by not storing redundant data. Instead, the system saves only one instance of the data; redundant material is replaced with a pointer to the saved copy. The classic example is the e-mail joke attachment that multiplies around the office; deduping would maintain just one copy of that attachment and ditch the rest. HYDRAstor uses distributed technology to “dedupe globally,” across the entire grid, which can cut storage capacity needs by as much as 95 percent, Dutch says.

“Competitor solutions deduplicate data separately within each instance of their solution,” she explains. “However, once their solution reaches its capacity or performance limitations, a new instance of their solution must be deployed. With no communication or deduplication capability across multiple instances of those solutions, the customer is virtually guaranteed to have multiple copies of some data across different instances of the solution, reducing the overall efficiency and deduplication ratio.”

HYDRAstor, however, “deduplicates data globally across the entire grid, ensuring that only unique data is stored once across the entire datacenter. Regardless which accelerator node or which filesystem the data comes from, it gets deduplicated across the entire common pool of storage on the backend to ensure only unique data is stored,” she says. “As HYDRAstor scales to larger grid configurations, so does the deduplication engine through automatic re-distribution of the deduplication processing responsibilities across all the storage nodes in the grid.”

With all the deduplication, though, data integrity is more important than usual. One lost file could be disastrous in case of rebuild or recovery. NEC says it addresses that with its protection scheme called Distributed Resilient Data (DRD), a patent-pending technology that Dutch says can provide a higher level of protection than RAID 5. “Our default level of 3 provides 300 percent more resiliency than RAID 5,” she says. “At that level, the system can withstand three node failures.” But if it’s RAID 5 you want, you can dial that in, as well, via the management software’s GUI. NEC also says data rebuild operations are quicker with DRD.

For disaster recovery, NEC has an optional feature for DynamicStor called RepliGrid Asynchronous Grid-to-Grid Replication. “RepliGrid determines what data being replicated does not already exist on the disaster recovery (DR) site grid, and then compresses and sends only new unique data over the WAN to the DR site grid. This minimizes the amount of bandwidth to reduce DR costs,” Dutch says.

RepliGrid is licensed on a per accelerator node basis, so users have the flexibility to only license it for the parts of the infrastructure that need to replicate. RepliGrid supports 1-1 replication as well as N-to-1 Grid-to-Grid Replication. This enables concurrent replication of multiple grids to a single DR site grid, which deduplicates data globally from all sources (both local and remote) to ensure that only unique data is actually stored.

Datacenter-Sized Backup

With the considerable compression capabilities, it is apparent NEC’s first target is backup and archiving. Dutch says that backup, in particular, benefits from global deduplication and compression since it involves so many repetitious files. “DataRedux coupled with data compression delivers greater than 20:1 capacity space reduction for backup data,” Dutch says. “And DRD protection ensures the deduped data is safe at all times.”

“We expect more growth in backup and archive environments” she says, “and will also expand into nearline applications, such as Web 2.0, which require affordable enterprise storage that is highly scalable, reliable and delivers high performance.”

NEC has tried to make a system that plays well all around. A HYDRAstor grid will work with existing network gear and storage software. “On the front end, we provide the standard protocols, we work with NFS or CICS, we can communicate with any backup or archive software,” says Gideon Senderov, director of technical marketing.

Installation also is pretty easy, the company says. According to Dutch, customers are “unboxing and backing up archival data in about 45 minutes.”

Coming Up: Geographic Reach

Grid storage has been emerging since the earlier part of the decade, and big players like HP and IBM have systems of similar intent. NEC doesn’t seem to pitch HYDRAstor against other solutions, but says its automated management features and storage optimization set it apart. “Other storage capacity optimization solutions may deliver similar space reduction and capacity optimization,” Dutch says, “but they lack HYDRAstor’s massive scalability, data protection, automatic self-management, and ability to evolve over time. And, finally, HYDRAstor includes data protection technologies that go beyond the limits of RAID and guarantee that deduplicated data is safe.”

“NEC is smart to not try, at least initially, to compete head-on with the big boys like EMC, NetApp, HP and IBM in terms of saying HYDRAstor is an ‘XYZ killer,'” analyst Peters says. “However, if it can get more traction with the product, the basic architecture could be extended in many ways, and be an approach that morphs into a number of products or simply one very flexible product. NEC’s issue is to become a recognized and serious player in storage. With its sheer size, high R&D spend, and strong Japanese market share — not to mention excellent products — it clearly has the credibility.”

As for the next version, Dutch says “NEC will continue to optimize HYDRAstor for multiple I/O patterns … and support implementation of multiple storage tiers with different performance and availability characteristics within a single grid.” And, she says, the system will evolve “to support geographically dispersed grids that span multiple datacenters and deliver a paradigm shift from disaster recovery to data resiliency.”

Subscribe to HPCwire's Weekly Update!

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

Supercomputers Take to the Solar Winds

June 5, 2020

The whims of the solar winds – charged particles flowing from the Sun’s atmosphere – can interfere with systems that are now crucial for modern life, such as satellites and GPS services – but these winds can be d Read more…

By Oliver Peckham

HPC in O&G: Deep Sea Drilling – What Happens Now   

June 4, 2020

At the beginning of March I attended the Rice Oil & Gas HPC conference in Houston. That seems a long time ago now. It’s a great event where oil and gas specialists join with compute veterans and the discussion tell Read more…

By Rosemary Francis

NCSA Wades into Post-Blue Waters Era with Delta Supercomputer

June 3, 2020

NSF has awarded the National Center for Supercomputing Applications (NCSA) $10 million for its next supercomputer - named Delta – “which will kick-start NCSA’s next generation of supercomputers post-Blue Waters,” Read more…

By John Russell

Dell Integrates Bitfusion for vHPC, GPU ‘Pools’

June 3, 2020

Dell Technologies advanced its hardware virtualization strategy to AI workloads this week with the introduction of capabilities aimed at expanding access to GPU and HPC services via its EMC, VMware and recently acquired Read more…

By George Leopold

Supercomputers Streamline Prediction of Dangerous Arrhythmia

June 2, 2020

Heart arrhythmia can prove deadly, contributing to the hundreds of thousands of deaths from cardiac arrest in the U.S. every year. Unfortunately, many of those arrhythmia are induced as side effects from various medicati Read more…

By Staff report

AWS Solution Channel

Join AWS, Univa and Intel for This Informative Session!

Event Date: June 18, 2020

More enterprises than ever are turning to HPC cloud computing. Whether you’re just getting started, or more mature in your use of cloud, this HPC Cloud webinar is an excellent opportunity to gain valuable insights and knowledge to help accelerate your HPC cloud projects. Read more…

Indiana University to Deploy Jetstream 2 Cloud with AMD, Nvidia Technology

June 2, 2020

Indiana University has been awarded a $10 million NSF grant to build ‘Jetstream 2,’ a cloud computing system that will provide 8 aggregate petaflops of computing capability in support of data analysis and AI workload Read more…

By Tiffany Trader

NCSA Wades into Post-Blue Waters Era with Delta Supercomputer

June 3, 2020

NSF has awarded the National Center for Supercomputing Applications (NCSA) $10 million for its next supercomputer - named Delta – “which will kick-start NCS Read more…

By John Russell

Indiana University to Deploy Jetstream 2 Cloud with AMD, Nvidia Technology

June 2, 2020

Indiana University has been awarded a $10 million NSF grant to build ‘Jetstream 2,’ a cloud computing system that will provide 8 aggregate petaflops of comp Read more…

By Tiffany Trader

10nm, 7nm, 5nm…. Should the Chip Nanometer Metric Be Replaced?

June 1, 2020

The biggest cool factor in server chips is the nanometer. AMD beating Intel to a CPU built on a 7nm process node* – with 5nm and 3nm on the way – has been i Read more…

By Doug Black

COVID-19 HPC Consortium Expands to Europe, Reports on Research Projects

May 28, 2020

The COVID-19 HPC Consortium, a public-private effort delivering free access to HPC processing for scientists pursuing coronavirus research – some utilizing AI Read more…

By Doug Black

$100B Plan Submitted for Massive Remake and Expansion of NSF

May 27, 2020

Legislation to reshape, expand - and rename - the National Science Foundation has been submitted in both the U.S. House and Senate. The proposal, which seems to Read more…

By John Russell

IBM Boosts Deep Learning Accuracy on Memristive Chips

May 27, 2020

IBM researchers have taken another step towards making in-memory computing based on phase change (PCM) memory devices a reality. Papers in Nature and Frontiers Read more…

By John Russell

Hats Over Hearts: Remembering Rich Brueckner

May 26, 2020

HPCwire and all of the Tabor Communications family are saddened by last week’s passing of Rich Brueckner. He was the ever-optimistic man in the Red Hat presiding over the InsideHPC media portfolio for the past decade and a constant presence at HPC’s most important events. Read more…

Nvidia Q1 Earnings Top Expectations, Datacenter Revenue Breaks $1B

May 22, 2020

Nvidia’s seemingly endless roll continued in the first quarter with the company announcing blockbuster earnings that exceeded Wall Street expectations. Nvidia Read more…

By Doug Black

Supercomputer Modeling Tests How COVID-19 Spreads in Grocery Stores

April 8, 2020

In the COVID-19 era, many people are treating simple activities like getting gas or groceries with caution as they try to heed social distancing mandates and protect their own health. Still, significant uncertainty surrounds the relative risk of different activities, and conflicting information is prevalent. A team of Finnish researchers set out to address some of these uncertainties by... Read more…

By Oliver Peckham

[email protected] Turns Its Massive Crowdsourced Computer Network Against COVID-19

March 16, 2020

For gamers, fighting against a global crisis is usually pure fantasy – but now, it’s looking more like a reality. As supercomputers around the world spin up Read more…

By Oliver Peckham

[email protected] Rallies a Legion of Computers Against the Coronavirus

March 24, 2020

Last week, we highlighted [email protected], a massive, crowdsourced computer network that has turned its resources against the coronavirus pandemic sweeping the globe – but [email protected] isn’t the only game in town. The internet is buzzing with crowdsourced computing... Read more…

By Oliver Peckham

Global Supercomputing Is Mobilizing Against COVID-19

March 12, 2020

Tech has been taking some heavy losses from the coronavirus pandemic. Global supply chains have been disrupted, virtually every major tech conference taking place over the next few months has been canceled... Read more…

By Oliver Peckham

Supercomputer Simulations Reveal the Fate of the Neanderthals

May 25, 2020

For hundreds of thousands of years, neanderthals roamed the planet, eventually (almost 50,000 years ago) giving way to homo sapiens, which quickly became the do Read more…

By Oliver Peckham

DoE Expands on Role of COVID-19 Supercomputing Consortium

March 25, 2020

After announcing the launch of the COVID-19 High Performance Computing Consortium on Sunday, the Department of Energy yesterday provided more details on its sco Read more…

By John Russell

Steve Scott Lays Out HPE-Cray Blended Product Roadmap

March 11, 2020

Last week, the day before the El Capitan processor disclosures were made at HPE's new headquarters in San Jose, Steve Scott (CTO for HPC & AI at HPE, and former Cray CTO) was on-hand at the Rice Oil & Gas HPC conference in Houston. He was there to discuss the HPE-Cray transition and blended roadmap, as well as his favorite topic, Cray's eighth-gen networking technology, Slingshot. Read more…

By Tiffany Trader

Honeywell’s Big Bet on Trapped Ion Quantum Computing

April 7, 2020

Honeywell doesn’t spring to mind when thinking of quantum computing pioneers, but a decade ago the high-tech conglomerate better known for its control systems waded deliberately into the then calmer quantum computing (QC) waters. Fast forward to March when Honeywell announced plans to introduce an ion trap-based quantum computer whose ‘performance’ would... Read more…

By John Russell

Leading Solution Providers

SC 2019 Virtual Booth Video Tour

AMD
AMD
ASROCK RACK
ASROCK RACK
AWS
AWS
CEJN
CJEN
CRAY
CRAY
DDN
DDN
DELL EMC
DELL EMC
IBM
IBM
MELLANOX
MELLANOX
ONE STOP SYSTEMS
ONE STOP SYSTEMS
PANASAS
PANASAS
SIX NINES IT
SIX NINES IT
VERNE GLOBAL
VERNE GLOBAL
WEKAIO
WEKAIO

Contributors

Tech Conferences Are Being Canceled Due to Coronavirus

March 3, 2020

Several conferences scheduled to take place in the coming weeks, including Nvidia’s GPU Technology Conference (GTC) and the Strata Data + AI conference, have Read more…

By Alex Woodie

Exascale Watch: El Capitan Will Use AMD CPUs & GPUs to Reach 2 Exaflops

March 4, 2020

HPE and its collaborators reported today that El Capitan, the forthcoming exascale supercomputer to be sited at Lawrence Livermore National Laboratory and serve Read more…

By John Russell

‘Billion Molecules Against COVID-19’ Challenge to Launch with Massive Supercomputing Support

April 22, 2020

Around the world, supercomputing centers have spun up and opened their doors for COVID-19 research in what may be the most unified supercomputing effort in hist Read more…

By Oliver Peckham

Cray to Provide NOAA with Two AMD-Powered Supercomputers

February 24, 2020

The United States’ National Oceanic and Atmospheric Administration (NOAA) last week announced plans for a major refresh of its operational weather forecasting supercomputers, part of a 10-year, $505.2 million program, which will secure two HPE-Cray systems for NOAA’s National Weather Service to be fielded later this year and put into production in early 2022. Read more…

By Tiffany Trader

15 Slides on Programming Aurora and Exascale Systems

May 7, 2020

Sometime in 2021, Aurora, the first planned U.S. exascale system, is scheduled to be fired up at Argonne National Laboratory. Cray (now HPE) and Intel are the k Read more…

By John Russell

Australian Researchers Break All-Time Internet Speed Record

May 26, 2020

If you’ve been stuck at home for the last few months, you’ve probably become more attuned to the quality (or lack thereof) of your internet connection. Even Read more…

By Oliver Peckham

Summit Supercomputer is Already Making its Mark on Science

September 20, 2018

Summit, now the fastest supercomputer in the world, is quickly making its mark in science – five of the six finalists just announced for the prestigious 2018 Read more…

By John Russell

Nvidia’s Ampere A100 GPU: Up to 2.5X the HPC, 20X the AI

May 14, 2020

Nvidia's first Ampere-based graphics card, the A100 GPU, packs a whopping 54 billion transistors on 826mm2 of silicon, making it the world's largest seven-nanom Read more…

By Tiffany Trader

  • arrow
  • Click Here for More Headlines
  • arrow
Do NOT follow this link or you will be banned from the site!
Share This