Posts Tagged ‘NAND’

Samsung Takes Over SSD Caching Software Developer. – X

Sunday, December 16th, 2012

Samsung Electronics pronounced Friday that it acquired secretly hold Nvelo, a colonize of solid-state drives caching software. The merger of Nvelo will capacitate Samsung to extend a ability to yield SSD associated storage solutions to business and also rise and sell several hybrid storage solutions, essentially for businesses.

“We demeanour brazen to accelerating storage creation in tighten team-work with Samsung storage experts as we assistance to broach entirely integrated SSD solutions to a market.” pronounced Jiurong Cheng, boss and arch executive officer of Nvelo.

By cleverly handling and optimizing a use of SSDs in a storage subsystem, SSD caching has turn an appealing resolution for both consumer and craving computing platforms. The timely formation of Nvelo’s flagship storage record into Samsung’s SSD technologies will give Samsung business entrance to an ever-evolving and some-more diversified portfolio of NAND storage solutions suitable for a extended operation of computing platforms.

“We are gratified with this transaction as a employees of Nvelo share a prophesy to take SSD storage into a next-generation of opening and reliability,” pronounced Young-Hyun Jun, executive clamp boss of peep product and record during Samsung Electronics.

Since Samsung has sole a tough expostulate business to Seagate and now does not sell HDDs itself, it is expected that a association will try to rise multi-layer enterprise-oriented storage sub-systems. Given that Samsung does not have knowledge in this business, it is expected that a association will co-develop such solutions with other marketplace players with a clever brand-name in enterprise-class storage.

The merger involves all record and personnel. Further sum of a agreement were not disclosed.

Tags: Samsung, SSD, HDD, HHD, NAND, Flash, Nvelo

Article source:

Samsung debuts dual new information core prepared SSDs

Wednesday, October 31st, 2012

Samsung has launched dual new solid-state drives (SSDs) — directed precisely during a craving storage and server market.

The dual drives, a SM843 and a SM1625, are directed during companies on a surveillance for drives that can hoop quick information writing, such as those doing vast SQL databases, virtualization, or media streaming.

The SM843 uses Samsung’s possess 20-nanometer MLC NAND and can means 11,000 input/output operations per second (IOPS), with latency cut by as many as 80 percent, according to Samsung.

The expostulate can strike adult to 520 MB/s review speeds, and adult to 420 MB/s write speeds. This expostulate is ideally matched to datacenter applications, while Samsung claims that it can write adult to 800TB during a life cycle, a large 1,600 percent alleviation over a prior era SSD.

The SM1625 also uses Samsung’s 20-nanometer MLC NAND, and facilities adult to 41,000 postulated pointless write IOPS, and incidentally reads adult to 101,000 IOPS when regulating both ports. The expostulate can strike writes speeds of adult to 740 MB/s, and reads speeds of 848 MB/s when regulating both ports.

The SM1625 also facilities super capacitors to strengthen information in-flight during a energy detriment situation.

The SM1625 ships in 100GB, 200GB, 400GB and 800GB capacities.

Both drives are accessible immediately, and pricing of both a SM843 and a SM1625 depends on quantities purchases, and a channels by that they are purchases.

“Samsung’s new information centre and craving SSDs broach intensely high opening with low energy expenditure in providing a many fit storage solutions for information centre applications,” pronounced Myungho Kim, Samsung clamp boss of memory marketing.

“Samsung will aggressively furnish a new choice of SSDs commencement this month to accelerate SSDs’ pierce into not usually a server though also a storage marketplace”.

The proclamation follows Intel’s launch of a SSD 335, a first SSD to underline 20-nanometer NAND peep memory.

Image source: Samsung.

Article source:

The tech defining next-gen computing

Saturday, October 13th, 2012

Honda UNI-CUB: what is it and because would we wish one?

  • Whatever happened to wireless charging?
  • Intel to use London to examine a city of a future
  • The post PC age. The tenure competence have been coined by Steve Jobs and caused Steve Ballmer some formidable moments in interviews, nonetheless we’re not disturbed about corporate point-scoring – we’re vehement about a possibilities over a desktop.

    Once, personal computing simply meant carrying entrance to a appurtenance we could call your own, and we could do so nonetheless wanting to spend a fortune. Today, it has turn many some-more personal.

    That competence sound like boardroom rhetoric, nonetheless if we cruise how computing has altered given Windows ruled all, it’s undeniable. Today, people don’t usually rest on a PC related to a hollow in a wall for internet access. Instead we count on gadgets like smartphones, tablets and laptops.

    These unstable inclination concede us to entrance a information wherever and whenever we want. That’s a complicated hint of personal computing – your data, your way, anywhere, any time. We’re going to try a technologies that a world’s biggest tech firms are cooking adult to make personal computing even some-more evident and powerful.

    You competence also like

    Nvidia CEO Jen-Hsun Huang

    Cloud gaming: quick to play, faster to start

    We’re not meddlesome in a pie-in-the-sky musings of futurologists and veteran guessers, though. We’ve consecrated a group of moment writers to puncture around and expose technologies and products that are impending completion. We’ll demeanour during a subsequent chips Intel hopes will pull computing to newer heights while scantily sipping battery power. We’ll try a graphics technologies that Nvidia believes will let your smartphone play games with daunt and refinement formerly indifferent for desktop PCs.

    Finally, we’ll examine new storage technologies, cutting-edge screens and other innovations in mobile record that will shortly be nearing in your hand.

    The internet of things


    What is it?

    The internet used to be usually about joining computers to share information, nonetheless it ‘s grown into something many bigger. In 2008 there were some-more inclination (not usually computers and phones) connected to a internet than there were people on Earth. The internet of things is about a towering intensity of putting positively all online.

    Why is it important?

    Imagine joining all to a internet, from pacemakers to roads. Now cruise what would be probable if those objects could all promote information too. Suddenly we would have a many larger bargain of a universe around us, and a resources of new information that could be used to urge a lives. Entire cities would be networked, with comparison infrastructure being upgraded and incorporated to work alongside a new.

    How does it work? Under IPv4, a universe was regulating out of a IP addresses compulsory for inclination to bond to a internet. Some countries had even tired their supply. IPv6 – a subsequent era internet custom – will capacitate 340 trillion, trillion, trillion inclination to bond to a internet. That’s 100 inclination for any atom on Earth.

    IPv6 creates a internet of things possible, and was a outcome of technologists essay towards that goal. Cisco estimates that by 2020 there will 50 billions inclination online, all broadcasting and receiving data.

    NTV processors

    NTV processors

    What is it?

    A critical x86 processor means of regulating during adult to 10 times a potency of today’s CPUs, with appetite expenditure as low as 2mW and a ability to scale from 280mV when regulating during 3MHz, all a proceed adult to 1.2v during 1GHz.

    Why is it important?

    Improved appetite potency is critical for a expansion of a computing continuum, not usually in terms of HPC servers, nonetheless also in terms a inclination in your pocket.

    How does it work?

    Back in Sep 2011, Intel showed off a antecedent x86 CPU so low-powered that it could be powered by a solar dungeon a distance of a postage stamp. This was a initial introduction of a Claremont height and a prophesy for near-threshold voltage (NTV) processors.

    The threshold voltage is a voltage compulsory for a given transistor to switch into an on state. Current CPU designs work during several times a threshold voltage in sequence to make gains in performance. If we run a processor during tighten to a threshold voltage, we make poignant gains in potency – Intel is claiming 5-10 times.

    Another by-product of doing this tighten to a threshold voltage is that a processor can work during a far-reaching energetic operation of doing frequencies.

    Nvidia GeForce GRID

    GeForce GRID

    What is it?

    GeForce GRID is a name given to a new GPU means of powering diversion streaming servers with twice a appetite potency of a tide crop. It allows a servers to pierce from one diversion any to around 8 streams per server.

    Why is it important?

    Again, such appetite potency is going to be critical for a flourishing marketplace like diversion streaming, that relies on such high opening and power-hungry servers. It will also meant a user needs usually a H. 264 decoder (like a complicated tablet, for example) in sequence to play full PC games.

    How does it work?

    Cloud gaming services have been around for a while, nonetheless due to diseased network structure, and twice a latency and loiter of a home console experience, it has nonetheless to take off.

    One of a categorical improvements that has come from a Nvidia GeForce GRID has been a slicing of latency between a user and a behind end. One of a ways it has finished this is by changeable a constraint and encoding partial of a tide onto a GPU itself. That means that once a label has finished displaying a frame, it’s prepared to be dismissed opposite a network nonetheless carrying to go anywhere nearby a CPU.

    The latency we knowledge with a home console is around 160ms, with a initial era of cloud gaming services being roughly double that. Nvidia says that with Gaikai and GeForce GRID, latency will come down to roughly 150ms. The home console is in trouble.

    Conflict-free microprocessor

    Conflict giveaway CPU


    What is it?

    There are 4 materials classed as ‘conflict minerals’ in today’s microprocessors. One of these, tantalum, is processed from coltan. This is especially found in a Democratic Republic of Congo, where boost from bootlegging financial a troops regime.

    Why is it important?

    Profits from a sale of these materials have been related to tellurian rights atrocities in a eastern segment of a DRC. Intel hopes that stealing this income will assistance stop a violence.

    How does it work?

    When a emanate initial arose, Intel expelled a matter that it was “unable to determine a start of a minerals that are used in a products.” It afterwards started mapping and tracing a supply sequence for all a 4 categorical dispute minerals: tantalum, tungsten, bullion and tin.

    Flash brazen to 2012 and Intel claims to have a supply line mapped to 90 per cent, and is operative to safeguard that in 2012 it will be means to denote that a microprocessors are accurate as conflict-free for tantalum. The idea is that, by a finish of 2013, it will be means to contend for certain that it is production a world’s initial microprocessor certified as wholly conflict-free for all 4 metals.

    System on a chip


    What is it?

    The holy grail for chip manufacturers and designers has prolonged been a full mechanism complement shrunk down to fit on a singular chip. We have been removing closer and closer, and will shortly be closer still.

    Why is it important?

    It’s all about appetite expenditure and efficiency. Instead of carrying innumerable chunks of fixed-function silicon doing opposite things and usually afterwards articulate to any other, carrying a singular chip means of doing all is distant some-more efficient.

    How does it work?

    These chips are a alloy of a normal CPU and GPU, putting both pieces of silicon in a same processor die. This is distant some-more fit than regulating dissimilar GPU and CPU silicon, nonetheless as nonetheless a opening isn’t adult there with a distant brethren.

    AMD will be a initial to be regulating a latest GPU and CPU record in a arriving Kaveri height in 2013 for a genuine opening APU.

    Intel has taken a somewhat opposite approach, radically creation a tip processors tighten to system-on-a-chip tools with a Ivy Bridge CPUs. These have an softened DX11 GPU on-die along with a fastest consumer CPU tools around. The arriving Haswell tools will embody an even larger concentration on graphics performance.

    Nvidia isn’t to be left out either, with a stirring Project Denver. Nvidia CEO Jen-Hsun Huang betrothed that it will be a critical gamer’s chip with a full Nvidia GPU as partial of a package. It’s approaching to be ARM-based, not x86, nonetheless could be another constrained SoC choice going forward.



    What is it?

    A memristor is a ‘memory resistor’ – a resistor that ‘remembers’ a formerly reason resistance.

    Why is it important?

    The memristor was theorised in 1971 by Leon Chua, nonetheless it wasn’t until 2008 that it was indeed constructed by HP Labs. The competition is now on to make memristors in an affordable way, as they could potentially reinstate NAND peep memory.

    How does it work?

    The speculation behind a memristor is candid adequate – when tide flows by it one way, a insurgency increases. Flow tide by it a other proceed and it decreases. When no tide is flowing, it retains a final insurgency that it had.

    As memristors don’t need a assign to reason their state, they are ideal for non-volatile operations and are a intensity deputy for existent NAND flash. This is finished some-more tantalizing by a fact that memristors don’t need many tide to operate, and can be run during distant faster speeds than existent NAND flash.

    There is a problem nonetheless – they’re costly to manufacture. Recent examine by UCL happened on a proceed of creation silicon-oxide memristors regulating existent manufacturer technologies, that should be significantly cheaper than a large labs.

    Memory Modem

    Memory Modem


    What is it?

    Memory Modem record takes many of a ideas used in a universe of communications and uses them to urge a trustworthiness and throughput of NAND peep memory, as used in SSDs.

    Why is it important?

    SSD prices have depressed significantly in new months. This has been driven by an contentment of manufacturers and softened production for a NAND peep used in SSDs, nonetheless nonetheless new technologies, these cost reductions will plateau.

    How does it work?

    The proclamation that Seagate is collaborating with DensBits to furnish cheaper plain state drives for a desktop and craving marketplace will hopefully change a attention as a whole to furnish even some-more affordable SSDs, that should meant they turn truly ubiquitous. While we’ve seen pricing of SSDs decrease this year, this freefall isn’t approaching to continue due to a fundamental costs of production NAND chips.

    Technologies are indispensable that maximize existent NAND chips, and this is accurately what a agreement between DensBits and Seagate should enable. DensBits Memory Modem tech will find a proceed into several of Seagate’s storage technologies, including 3 bits/cell (TLC) consumer inclination and dual bits/cell (MLC) craving offerings.

    Memory Modem record enables cells to enclose some-more bits, and lets drives use smaller routine nodes. In other words, a SSDs can use cheaper NAND flash, nonetheless work during rival speeds.

    GPU supercomputer

    GPU supercomputers

    What is it?

    Nvidia’s latest recover competence demeanour like usually another graphics card, nonetheless a dictated focus of Tesla K20 is distant private from powering a latest games. This is a graphics processor section designed to assistance Nvidia retrieve a covering of a world’s many absolute computer.

    Why is it important

    Nvidia’s GPUs are already used by Oak Ridge National Laboratory in Tennessee, nonetheless this ascent should see a supercomputer leapfrog into initial place as a fastest in a world.

    How does it work?

    In sequence to explain a series one mark in a supercomputer gymnasium of celebrity (, Nvidia indispensable to redesign a GPUs significantly. Not usually did it need to furnish a some-more absolute GPU, it indispensable to make a some-more fit one during a same time.

    It has finished this with a latest GPU, a Tesla K20 (codename KG110) – negligence a doing frequency, nonetheless also cramming some-more CUDA cores into any chip. 19,000 of these new GPUs will be used to ascent a existent Titan supercomputer, that is housed in a US Department of Energy ‘s biggest scholarship laboratory.

    Nvidia claims that this ascent will proportion to an eightfold alleviation in performance, means of doing 25,000 trillion floating indicate operations per second. It should also meant that Titan is twice as quick as a world’s second fastest supercomputer, Fujistu’s K Computer in Japan.



    What is it?

    The deputy for a existent DDR3 memory customary found in a machines.

    Why is it important?

    As a systems get faster, a volume of information we change around increases, that puts a bucket on a complement buses and memory architecture. We can’t simply hang with DDR3 and wish for a best.

    How does it work?

    The idea for DDR4 isn’t a million miles divided from a strange aim for DDR3 – to urge opening over a prior record while shortening appetite requirements. This latter indicate is achieved by environment a appetite requirement for DDR4 during 1.2V, as against to DDR3’s 1.5V.

    There are DDR3 memory modules accessible during reduction than 1.5V, nonetheless DDR4 is approaching to follow suit, and low voltage memory regulating during 1.05V is approaching to seem too.

    When it comes to augmenting a opening of DDR4, a selection doesn’t see a extensive change in a prefetch buffers, instead relying on an boost in a doing magnitude to hoard speed improvements.

    The initial DDR4 modules will have an doing magnitude of 1,600MHz, nonetheless given that DDR3 already lays claims to such speeds, faster modules should be accessible too. The top extent of a selection defines a 3,200MHz frequency, nonetheless it competence be a while before modules regulating this quick are released.

    John Carmack’s practical existence


    What is it?

    A practical existence headset that Doom and Quake creator John Carmack pieced together in his gangling time. It uses an ‘affordable’ optics kit, a 6-inch shade and a program covering to expostulate a display, giving we a sense of old-school practical reality.

    Why is it important?

    Despite a earnest gash during a mainstream in a late 80s, practical existence was singular by a hardware of a time. When a likes of graphics supremo John Carmack start removing vehement though, it’s value holding a demeanour to see where it could lead.

    How does it work?

    Virtual reality, if we don’t remember it a initial time around, works in a identical proceed to stereoscopic 3D. Each eye sees a somewhat opposite image, that we afterwards appreciate as a 3D space. When it comes to VR, instead of polarising glasses, we use a helmet containing dual displays. The transformation of your conduct is available as an input, and is eliminated to a practical world.

    VR resurfaced when John Carmack demonstrated his pet plan on a theme behind sealed doors during this year’s E3, finish with a operative headset and game. The headset comes in pack form, nonetheless a optics are good adequate to make it a constrained experience.

    Carmack’s id Software is looking re-release Doom III after this year, and a diversion will support VR helmets, including a program covering indispensable to make those optics work when trustworthy to a customary screen.

    Article source:

    Micron sees Elpida understanding shutting notwithstanding challenge

    Friday, October 12th, 2012

    Fri Oct 12, 2012 11:30am EDT

    SAN FRANCISCO (Reuters) – Micron Technology’s arch executive pronounced he expects to finish a merger of unsuccessful Japanese memory chipmaker Elpida Memory Inc in a initial half of subsequent year, notwithstanding antithesis from a organisation of bondholders.

    Micron, that is losing income due to a exploding PC industry, wants to emanate incomparable economies of scale and charity in early Jul to buy Elpida for about $750 million in money and to compensate creditors a sum of $1.75 billion in annual installments by 2019.

    However, a organisation of Elpida bondholders contend Micron is charity too small for a chipmaker.

    Low prices and usually rising investment costs to exercise new technologies have been pushing converging in a rarely rival and cyclical memory chip industry.

    Micron’s squeeze of Elpida, an Apple (AAPL.O) retailer that creates chips for smartphones, tablets and computers, would lift a U.S. association into second place behind marketplace personality Samsung Electronics (005930.KS) in a tellurian marketplace for DRAM chips.

    A cabinet set adult by a Japanese justice to inspect Elpida’s offer to be taken over by Micron, along with a offer by a bondholders, is approaching to make a recommendation by Oct 29, after than progressing estimated.

    “We still design to tighten this in a initial half of 2013. Exactly when that is, we can’t contend with a lot of pointing since we consider unequivocally a pivotal domino is — when we get regulatory capitulation from all a endangered countries,” Micron Chief Executive Mark Durcan pronounced during an analysts eventuality in Boise, Idaho, where Micron is based.

    Slow economies in Europe and a United States and an oversupply of NAND chips used by Apple and other manufacturers of mobile inclination have beaten prices and contributed to waste for Micron, that is a usually vital U.S. memory chipmaker.

    Consumers’ flourishing welfare for smartphones and tablets instead of laptops has been spiteful PC manufacturers, that traditionally have been vital buyers of DRAM chips. Two weeks ago, Micron posted lower-than-expected formula for a Aug entertain since of debility in a PC market.

    Micron creates both NAND chips and DRAM chips. The association pronounced during a eventuality on Friday that new DRAM pricing has been somewhat reduce than expected, while new NAND pricing has been somewhat aloft than expected.

    To equivalent a disappearing PC memory business, Micron, Samsung and SK Hynix (000660.KS) are focusing some-more of their resources on building improved — and some-more essential — memory chips for mobile devices.

    In a pointer of a flourishing concentration on mobile, Micron recently hired Michael Rayfield, before a comparison executive during Nvidia (NVDA.O) who was heavily concerned in mobile processors directed during tablets and smartphones.

    The Elpida bondholder group, that says it binds about $1.2 billion in Elpida bonds, has submitted a devise to a Tokyo justice that values a chipmaker during some-more than 300 billion yen ($3.8 billion), good above Micron’s offer.

    The bondholders did not put brazen an equity investor, nonetheless they charity to lend Elpida 30 billion yen to assistance a chipmaker restructure.

    “When we consider about who has a wherewithal to yield a suggestive lapse to a cumulative creditors on a go-forward basement it’s a Micron offer,” Durcan said. “The other offers out there … unequivocally don’t give most declaration that there will be anything for anybody.”

    Shares of Micron were down 1.04 percent during $5.68. ($1 = 78.34 Japanese yen)

    (Reporting By Noel Randewich; Editing by Theodore d’Afflisio)

    Article source:

    Micron Technology Given “Overweight” Rating during Piper Jaffray (MU)

    Tuesday, October 2nd, 2012

    Piper Jaffray validated their overweight rating on shares of Micron Technology (NYSE: MU) in a investigate note expelled on Friday morning. They now have a $12.00 cost aim on a stock.

    “We design NAND ASPs to strengthen given augmenting brew of SSDs. Overall sum margins were adult 40bps q/ q notwithstanding a 10% q/q income decline. … While a augury for NAND stays robust, we do not see DRAM recuperating anytime before C4Q12, as pivotal drivers such as Windows 8 and a successive PC modernise cycle benefit traction. During a quarter, Micron generated $450m in money upsurge from operations with money in a change piece rising to $2.9bn. The batch is trade during 0.8x book value, that we trust is rarely ignored given a supply rationalisation holding place in both DRAM and NAND, as good as a Elpida acquisition.,” a firm’s researcher wrote.

    Micron Technology traded adult 0.43% on Friday, attack $5.795. Micron Technology has a 52-week low of $3.97 and a 52-week high of $9.16. The company’s marketplace top is $5.871 billion.

    A series of other firms have also recently commented on MU. Analysts during Wells Fargo Co. reiterated an outperform rating on shares of Micron Technology in a investigate note to investors on Friday. Separately, analysts during JPMorgan Chase reiterated a neutral rating on shares of Micron Technology in a investigate note to investors on Friday. They now have a $7.50 cost aim on a stock, down formerly from $8.50. Finally, analysts during Jefferies Group cut their cost aim on shares of Micron Technology from $11.00 to $10.00 in a investigate note to investors on Friday. They now have a buy rating on a stock.

    Article source:

    Resistive Ram cache to make Flash fly, contend boffins

    Monday, June 25th, 2012

    SSDs interest to typical mechanism users since of their speed and silence. Data centre folk conclude those qualities too, though also like a SSD’s really low energy consumption. Energy is no tiny cost for a information centre, where there can be tens of thousands of drives all slurping electricity during once.

    But it’s not a giveaway ride. SSDs aren’t cheap, and they have longevity issues. Write information on them too many times and you’ll usually be means to review them after that. The fact that vast areas of memory rather than singular pieces need to be created during a time – and erased first, a routine that clears an even bigger space – means they are not befitting to applications that perceptibly write and re-write data.

    Solving these problems is something of a Holy Grail in a storage business. Many options exists – or have, during least, been due – though they’re all many years from replacing indiscriminate a NAND Flash chips used in today’s SSDs.

    One Japanese group, from Chuo University, Tokyo, has suggested a half-way house: regulating one of a new kinds of non-volatile memory, Resistive Ram (RRam) as a aegis between a outward universe and a horde drive’s NAND chips. This “hybrid” SSD uses Flash for bulk storage, and RRam for speed.

    RRam uses a underline of certain dielectric materials, that routinely vaunt a high resistance, to casually emanate physical, permanent paths, called “filaments”, down that stream can flow. All it takes is a focus of amply high though not unreal voltage.

    The pathways revoke a material’s resistance, though it’s probable to mangle them, returning a dielectric to a high insurgency state. Re-apply a voltage, new paths form, and a element re-enters low-resistance mode.

    It’s this ability to say possibly of dual states – high and low insurgency – that allows a dielectric to store binary information. Hook it adult to a determining transistor and we have a serviceable non-volatile memory cell.

    And it’s one that can switch most some-more fast than a Flash equivalents can: 10ns compared to 100,000ns. The upshot: a write speed is much, most aloft – closer, in fact, to flighty memory like Ram.

    Coupling it with Flash puts a RRam in a purpose of a cache. Controller algorithms store frequently review information in a RRam, from where a information can be collected some-more quickly.

    Next page: Cache in hand

    Article source:

    OWC Adds Mercury Electra MAX 3G 960GB SandForce Driven SSD …

    Saturday, June 23rd, 2012


    The Mac enthusiasts and plain state evangelists over at Other World Computing (OWC) have announced a new 2.5″ SSD with 1024GB of peep onboard. Dubbed a Mercury Electra MAX 3G 960GB SSD, it’s one of a initial 1TB 2.5″ drives to hit North America.

    While a ability to have 1TB of plain state storage in a 2.5″ form factor is certain to be a large offered point, a MAX also has a flattering reasonable cost to foot – $1,299.99. That’s $1.27 per GB of peep on board, or $1.45 per user-available GB that means a MAX is already at half a price of some currently accessible 2.5″ 1TB SSDs.

    Interestingly, a new OWC is dual SATA II SandForce 2181 processors in a RAID configuration. Each controller is interconnected with 512GB of “Tier 1″ Asynchronous NAND flash. Tying it all together is Silicon Image’s SteelVine RAID controller, that handles information striping duties between a dual SF controllers. OWC rates opening with both compressible and incompressible results, and the 3Gbps MAX pushes both to a 3Gbps limit. Sequential reads are listed during adult to 254MB/s with compressible information and 252MB/s with pointless data. Writes are pronounced to achieve 250MB/s repeating and 223MB/s with random data. After formatting and SandForce’s share of accessible flash, 894GB is left for users.

    Two controllers in an inner RAID pattern used to be distant some-more common, though such designs have depressed out of preference in a past few years. OWC’s examination with a MAX could be a pointer that there are some-more than a few Mac users looking for 1TB of plain state storage in a 2.5″ drive. While 1TB drives substantially won’t go mainstream until smaller NAND lithographies double a ability of any NAND die, the assertive pricing of a new Electra means some potential customers won’t have to wait until then.

    OWC’s Mercury MAX 3G is already on sale during OWC’s storefront.

    OWC Mercury Electra Prices during Amazon!

    Article source:

    Super Talent Introduces 2133 MHz Over-Clocked Quad-Channel DDR3

    Friday, March 2nd, 2012

    super talent quadra dimm 01 300x119 Super Talent Introduces 2133 MHz Over Clocked Quad Channel DDR3Super Talent Technology, a heading manufacturer of NAND peep solutions, currently announces a 2133 MHz Quadra kit, a fastest over-clocked quad-channel DDR3 pack to date.

    The Quadra array was designed to give PC enthusiasts a ultimate computing experience. These new 2133 MHz procedure capacitate a business to take full advantage of a new 6-Core processor formed motherboards by regulating during speeds adult to 2133 MHz.  With these new 2133 MHz Quadra Series modules support, a new Sandy Bridge E formed motherboards can now support a full 16 GB of DDR3 DRAM during over-clocked speeds of 2133 MHz.

    Super Talent has grown a specifically designed feverishness penetrate that enables this resolution well work with a i7-3960X 6-Core processors.  Validated regulating a X79 chipset, a 2133 MHz overclocking DDR3 DIMMs is now accessible possibly away or as a pack of 4 modules.

    Article source:

    OCZ Technology and Marvell to Debut Next Generation PCI Express

    Saturday, January 7th, 2012

    OCZ Technology Group, Inc. (Nasdaq:OCZ), a heading provider of high-performance solid-state drives (SSDs) for computing inclination and systems, and Marvell (Nasdaq: MRVL), a worldwide personality in integrated silicon solutions, currently denounced a industry’s fastest and many versatile PCI Express (PCIe) storage system, a Z-Drive R5. The increasing performance, reliability, and continuation of a Z-Drive R5 is designed to take PCIe-based plain state storage to a subsequent turn for craving environments.
    The Z-Drive R5 facilities a jointly grown “Kilimanjaro” OCZ and Marvell local PCIe to NAND peep controller platform, permitting for totally scalable opening and excess while expelling a need for a apart storage controller, so shortening a cost to muster high opening plain state storage systems in a information center.

    The local PCIe Z-Drive R5 offers an intelligent, hardware-meets-software managed resolution designed to dramatically accelerate database, enterprise, and virtualization applications. Combining high-performance pattern with OCZ managed software, a Z-Drive R5 provides second-to-none performance, flexibility, durability, and extended trustworthiness features, permitting datacenters to rest on a PCIe-based SSD as their primary Tier 0/1 storage resolution and a viable end-to-end SAN replacement.

    The Z-Drive R5 is ideal for perfectionist computing environments and features:

    Incredible bandwidth capabilities and limit transactional performance
    High capacities adult to 12TB
    Ideal for all craving information forms with both compressible and non-compressible files as good as vast information sets
    Complete storage subsystem government with OCZ Virtualized Controller Architecture™ 3.0 program functions
    Compatible with VMware ESXi and ESX, Linux, Windows Server 2008, and OS X to support a far-reaching operation of systems and servers
    Complete energy destroy insurance choice for limit information integrity
    Full tallness and half tallness sizes, ideal for space compelled 1U servers and multi-node rackmount servers
    MLC, eMLC , and SLC NAND Flash options
    “We are ceaselessly pulling a pouch to broach innovative PCIe SSDs to residence hurdles that a clients face when estimate and handling ever flourishing amounts of formidable information sets,” pronounced Ryan Petersen, CEO of OCZ Technology. “The Z-Drive R5, leverages a absolute Kilimanjaro tallness to broach higher performance, larger capacity, and an extended hardware and program feature-set that is unmatched in a industry. This best-of-breed subsequent era local resolution enables clients to discharge I/O bottlenecks for even a many information complete server and craving storage applications.”

    “Marvell is vehement to work with OCZ on this local PCIe to NAND peep controller platform, formed on a 88NV9145 silicon,” pronounced Alan Armstrong, clamp boss of Marketing for a Storage Business Group during Marvell Semiconductor, Inc. “We trust a PCIe SSD marketplace will fast change to a local PCIe to NAND architecture, and a Kilimanjaro tallness represents OCZ and Marvell’s clever partnership in bringing this rarely scalable pattern to market.”

    The Z-Drive R5 will entrance during Storage Visions 2012 in a OCZ counter (No. 13) and will also be showcased during a 2012 Consumer Electronics Show (CES) in a Marvell counter (No. 30542) located in a South Hall 3 on a top level.

    About OCZ Technology Group, Inc.
    Founded in 2002, San Jose, CA-based OCZ Technology Group, Inc. (OCZ), is a personality in a design, manufacturing, and placement of high opening and arguable Solid-State Drives (SSDs) and reward mechanism components. OCZ has built on a imagination in high-speed memory to turn a personality in a craving and consumer SSD markets, a record that competes with normal rotating captivating tough hoop drives (HDDs). SSDs are faster, some-more reliable, beget reduction feverishness and use significantly reduction energy than a HDDs used in a infancy of computers today. In further to SSD technology, OCZ also offers high opening components for computing inclination and systems, including enterprise-class energy government products as good as leading-edge mechanism gaming solutions. For some-more information, greatfully visit:

    About Marvell
    Marvell (NASDAQ: MRVL) is a universe personality in a growth of storage, communications and consumer silicon solutions. Marvell’s different product portfolio includes switching, transceiver, communications controller, wireless and storage solutions that energy a whole communications infrastructure, including enterprise, metro, home and storage networking. As used in this release, a tenure “Marvell” refers to Marvell Technology Group Ltd. and a subsidiaries. For some-more information, revisit

    Article source:

    Apple buys SSD builder Anobit

    Wednesday, December 21st, 2011

    Apple has purchased solid-state expostulate (SSD) builder Anobit, an Israeli start-up that creates storage products for information centre use out of consumer-grade NAND peep memory.

    According to a Israeli journal, Apple paid about £320 million to squeeze a private company.

    According to published reports, Anobit’s executives collected employees during a Herzliya domicile to surprise them of a execution of a Apple deal.

    Apple already uses Anobit’s peep chip record in a iPhone, iPad and MacBook Air laptops. According to reports, Apple would use Anobit’s record to boost a peep memory in a devices, as a cost to furnish it would be significantly reduce than purchasing it from a third-party provider.

    Israeli Prime Minister Benjamin Netanyahu tweeted about a deal stating, “Welcome to Israel, Apple Inc. on your initial merger here. I’m certain that you’ll advantage from a fruit of a Israeli knowledge.”

    Neither Anobit, nor Apple would immediately endorse a acquisition.

    Anobit has constructed dual generations of a Genesis SSD technology. The egghead skill that sets it detached from other SSD manufacturers is a controller, that uses firmware it calls Memory Signal Processing (MSP).

    The controller record extends a continuation of customary consumer-grade multi-level dungeon peep from about 3,000 write/erase cycles to some-more than 50,000 cycles – creation MLC record suitable for high-duty cycle applications such as relational databases.

    “You’re possibly regulating a some-more modernized controller with consumer class NAND or your leveraging enterprise-class NAND. Anobit’s proceed is to use a cheapest NAND they can find and afterwards use their some-more modernized controller technology,” pronounced Jeff Janukowicz, a investigate executive during IDC.

    Anobit is not alone among SSD processor makers who can extend MLC NAND’s reliability. For example, Sandforce creates a processor that uses information application and RAID design to get around a stipulations of MLC.

    Sandforce uses 24-bit/512-byte blunder improvement formula in a peep controllers to extend a life of MLC-based SSDs. “However, a elemental emanate is that a vigilance peculiarity is declining, and Anobit’s record helps to get a ‘cleaner’ signal,” pronounced Gregory Wong, an researcher with Forward Insights.

    As of September, Anobit pronounced it had sole 20 million MSP controllers to systems makers, that embody consumer-grade products as good as single-level dungeon (SLC) SSDs sole by other vendors.

    If Anobit’s numbers are accurate, a Genesis 2 SSD has staggeringly high performance. The expostulate generates pointless read-write rates of 70,000 I/Os and 40,000 I/Os per second, respectively.

    The Genesis 2 SSD boasts a limit consecutive read/write rate of 540MBps and 510MBps, respectively. The new SSD’s prototype had a postulated consecutive read/write rate of 220MBps and 180MBps, respectively.

    Anobit has shrunk a electronics from a progressing 40+ nanometer (nm) routine to 25nm, thereby permitting it to double a ability of a expostulate while also timorous it from a 3.5-in. form cause to 2.5-in. The Genesis 2 ranges in ability from 100GB to 200GB, 400GB and 800GB.

    As electronics shrinks in peep memory products, blunder rates boost since a electrons used to emanate pieces of information trickle between a thinner dungeon walls.

    In sequence to repair information errors, manufacturers embody error-correction formula in their NAND flash. The aloft a bit-error rates, a some-more ECC is required. Simply put, Anobit’s processor is means to continue to review information for a longer duration of time compared with standard NAND peep with hard-coded ECC, that takes adult peep capacity.

    Wong pronounced Anobit is a initial association to commercialize a signal-processing technology, that uses program in a controller to boost a signal-to-noise ratio, creation it probable to continue reading information even as electrical division rises.

    Article source: