Resultados 1 a 4 de 4
  1. #1
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010

    [EN] Intel Optane

    Disregarding 5 to 1 price difference, 3D XPoint-based SSDs are superior to NAND flash-based SSDs in every aspect.

    Alan Chen

    Intel Corp. has made waves in the server SSD market with the reveal of its Optane SSD DC P4800X Series this March.

    Based the company's in-house 3D XPoint memory, the DC P4800X Series is scheduled for shipments in the second half of 2017.

    "3D XPoint greatly outperforms NAND flash and will help Intel reinforce its leading position in the high-end segment of the SSD market that has become increasingly competitive," said Chen. "In the future, the product roadmap of 3D XPoint could include client-grade SSDs and storage-class memory (SCM) devices. Intel naturally wants to expand the adoption of this technology in other areas as to create flexibility in pricing and consumption of in-house production capacity."

    DRAMeXchange's data show that Intel was the top brand in the global server SSD market for 2016 with more than 30% of the market share. Rival Samsung closely followed Intel with a roughly similar market share. So far, Intel's leadership in the server SSD market comes from leveraging the company's dominance in the CPU market. However, Samsung and later entrants to this field are gaining experience and have adopted aggressive pricing to erode Intel's market shares for mid-range and low-end server SSDs.

    Chen added: "Intel lags behind Samsung and Toshiba in the production capacity for NAND flash. Putting forward a new premium SSD line that uses a proprietary memory technology is a way for Intel to maintain high profit margin for its server business with limited NAND flash capacity. Furthermore, the introduction of 3D XPoint will help Intel reduce the risk of being overly reliant on the high-end server SSD market. Intel later on will likely use 3D XPoint for client-grade SSDs and SCM devices. By introducing this memory technology to other application markets, Intel can have more flexibility in terms of pricing and capacity consumption."

    Comparatively speaking, the price per gigabyte of a 375GB SSD from the DC P4800X Series is about four times as much as that of a NAND flash-based SSD with similar specifications and PCIe interface. Whether 3D XPoint can penetrate into the mid-range and low-end product segments will be determined mainly by pricing. This new type of memory chips have to become more attractive cost-wise for OEMs in the client-grade SSD and SCM device markets.

    From the design perspective, SSDs from the DC P4800X Series use 128Gb 3D XPoint chips that are stacked together. The interface is PCIe and the controller chip is developed in house.

    In terms of capacity options, Intel will first release 375GB products and then roll out 750GB and 1.5TB models.

    According to the listed specifications, SSDs of this series is capable of reaching 2,400MB/s for sequential reading, 2,200MB/s for sequential writing and over 500,000 in IO/s. Furthermore, they have a typical latency of less than 10μs and score 30 on the major endurance indicator drive writes per day.

    "The DC 4800X Series is presented as a SSD solution that will fully take advantage of PCIe G3x4 while having a high level of endurance," said Chen.

    Based on the official recommended price, a 375GB SSD of the DC P4800X Series is set at around $4/GB. Meanwhile, a NAND flash-based SSD with similar specifications and PCIe interface can cost as low as $1/GB. This four-fold difference means that the DC P4800X Series is probably targeting the high-end segment of the server SSD market during the initial release period.

    Intel's new 3D XPoint-based SSDs can expand the capacity of server DRAM via the company's proprietary virtualization software known as Intel Memory Drive Technology, which is a separate product and costs extra. The bundle price for the 375GB SSD of the DC P4800X Series and the virtualization software together comes to around $5/GB. Considering the average sales price of server DRAM DIMMs (currently at about $7/GB), the bundle price is actually competitive in this respect.

    "Intel's new SSD line is between server DRAM DIMMs and NAND Flash server SSDs in terms of efficiency and pricing," noted Chen. "3D XPoint-based SSDs lack the performance capability to fully substitute server DRAM DIMMs. However, products from the DC P4800X Series can take on the role of SCM devices and work with server DRAM DIMMs to improve the operation of the entire server system."

    Moreover, SSDs using 3D XPoint have several advantages over NAND Flash-based counterparts. The random reading and random writing speeds of the former are faster and more consistent. DC P4800X products also vastly outperform conventional NAND flash-based SSDs on the latency and endurance indicators.

    Disregarding the price difference, 3D XPoint-based SSDs are superior to NAND flash-based SSDs in every aspect.

  2. #2
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010

    Violin Memory Acquired by Quantum Partners


    Violin Memory, Inc. announced the completion of the restructuring and sale process it initiated in December, and its acquisition by Quantum Partners LP (Soros), a private investment fund managed by Soros Fund Management LLC (SFM).

    Violin retains its proprietary technology, customer base and innovative team of professionals.

    Ebrahim Abbasi, the newly appointed CEO and president of Violin, stated: "Soros' acquisition of Violin recognizes the value of Violin's world-class customers which include Fortune 500 companies, as well as the tradition of innovation and dedication demonstrated by its team. As the pioneer in the all-flash array market, Violin is the market leader in high performance, low latency data services for private, public and hybrid cloud environments. I am honored to lead the new company into the future and focus on product innovation and customer excellence."

    Eric Carey, the CIO, Valley Health System, an US healthcare provider, also stated: "Throughout the reorganization process, Violin continued its enterprise customer support enabling us with a seamless operation of Violin's flash storage platform deployed in our mission critical Meditech application. This commitment to customer success is what we have come to experience with Violin."

    "We are excited to welcome Violin to our portfolio of businesses and continue our long relationship with the company. We are committed to support Violin in its tradition of customer excellence and product innovation," said Nicholas Esayan, principal, SFM Private Equity.

    Violin will continue to offer customers a suite of professional and support services that help customers transition to all-flash in their data centers. The company is committed to enabling customers to meet their business and technical objectives with Violin's high performance and low latency all-flash arrays. Customers can rely upon Violin's technical experts and best practices to optimize deployments, migrate data, and access support resources 24x7x365 that provide customers peace of mind in running mission critical operations.

    Our Comments

    Sad story than that of Violin Memory even if it was a true pioneer in all-flash array market (for tier 1), currently one the - fastest - if not the fastest - storage business. But then almost every storage company in the world has become a competitor by adding their own all-flash arrays to portfolio.

    Founded in 2005, Violin is headquartered in Santa Clara, CA, and got $186 million in financial funding including:

    - $10 million in 2010
    - $35 million and then $40 million in 2011
    - $80 million in 2012

    Among investors there are Samsung, Toshiba, Jupiter Networks, SAP and Imation

    It acquired assets (NAS and web caches) of Gear6 in 2010 and GridIron Systems in SAN application accelerator appliance in 2012.

    The first signs of the difficulties began when the start-up falls on $162 million IPO in 2013 by more than 17% on opening day. Then the stock never performed the way it expected. Ultimately, the stock went essentially worthless a year ago.

    Two months after the IPO, CEO Don Basile was fired.

    In 2014, the firm reduces workforce by 103 or 21%.

    During an interview published in in 2014, Kevin De Nuccio, president and CEO, told us: "Profitability is achievable at end of next year." It never happens.

    Also at the question: "You don't have de-dupe or compression a must-have features considering high price of SSDs. Why not?", he answered: "I think there is a place for de-duplication. In the space where we penetrated the market, flash has been adopted early, it's mostly about performance in databases. You don't do de-duplication in databases because it takes to much power away."

    Another error was to try to design and manufacture its own SSDs. About no all-flash player did that because it's a complete different activity. Per comparison, did you see any HDD subsystem company manufacturing disk drives? Violin Memory finally sold is PCIe product line to SK for $23 million in cash in 2014.

    The same year, it was obliged to offer $105.0 million of convertible senior notes in private placement to qualified institutional buyers.

    Being in financial trouble, unhappy stockholder Clinton Group with Imation threatens to change board of directors the following year and later to sell the company.

    In 2016, Violin was obliged to submit a plan to restore compliance with New York Stock Exchange continued listing requirement and proposed reverse stock split ratio of 1:4.

    After being suspended by NYSE and under Chapter XI, common stock began trading on OTCQX.

    For 2FQ17, it recorded ugly financial results with worst revenue ($7.5 million) and continuing huge net loss since the 15 former quarters.

    Following bankruptcy auction, it finally finds a buyer, Soros Fund Management (behind VM Bidco), in a deal valued at a mere $14.5 million, and retains its IP, customer base and professional team.

    Ebrahim Abbasi, has just been appointed CEO and president. He was COO of the company since 2014 and replaces Kevin De Nuccio having the same positions for three years and three months.


    The acquisition by a private investment fund means that:

    - no storage company wanted to buy Violin
    - it's better for Violin to find an acquirer than to perish - price was very low
    - there will be a lot of restructurations before Quantum Partners will try to sell the acquired firm

  3. #3
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010

    Toshiba NAND Chip Business for Sale, but What About HDDs and SSDs?

    Could Western Digital acquire total package?

    According to different sources, potential bidders are Apple with Foxconn, Broadcom, Google, Hon Hai Precision Industry Co, Innovation Network Corp. with KKR & Co., Micron, SK Hynix, Tsinghua Group and Western Digital.

    Jean-Jacques Maleval

    142-year old conglomerate Toshiba is in big financial trouble with operating loss of ¥573 billion for global revenue of ¥3,847 billion revenue down 167% Y/Y for the nine months ended last December. Main reason was the ugly situation of subsidiary Westinghouse Electric company (nuclear power unit) in bankruptcy.

    To get money to survive, the company has decided to sell a jewel, its NAND chip business, and is aiming to complete the operation by March 2018 to raise much-needed cash.

    There are only half a dozen of NAND chips manufacturers in the world as this business involves enormous investments to build fab:

    • Intel/Micron in Lehi, UT, Manassas, VA and Singapore (IMS)
    • Samsung in Austin, TX, China and Korea
    • Toshiba/WD in Japan
    • SK Hynix in China and Korea
    • Powerchip in Taiwan

    DRAMeXchange estimates that revenue ranking of branded NAND flash suppliers in 4Q16 was $12 billion with the following ranking:

    1. Samsung (37% market share)
    2. Toshiba (18%)
    3. Western Digital/SanDisk (18%)
    4. Micron (11%)
    5. SK Hynix (10%)
    6. Intel (7%)

    According to different sources, potential bidders are Apple with Foxconn, Broadcom, Google, Hon Hai Precision Industry Co, Innovation Network Corp. with KKR & Co., Micron, SK Hynix, Tsinghua Group and Western Digital.

    Apple is probably the biggest client of these chips in the world for its smartphones and computers as the company decided few years ago to stop to offer HDDs and to integrate SSDs only. If it produces its own chips via Toshiba's acquisition - with the help of Chinese iPhone subcontractor manufacturer Foxconn - and not from outsourced firms like Samsung, it will get them at better prices

    For SK Hynix or Micron, getting Toshiba business will help them to reach the second position in the market.

    Broadcom in Singapore is an avid acquirer of companies (Brocade, Cirrus Logic storage portfolio), Gadzook, LSI, NetLogic, RAIDcore, Siliquent Technologies, Sunext Design these past years) and could enter in flash chip manufacturing. It may consider joining with private equity investor Silver Lake on a bid of $23 million.

    We will be surprise to see Chinese firm Tsinghua Unigroup being able to realize such a big acquisition.

    Google has the money to do it but we don't see flash chips being really complementary to its business.

    Hon Hai is a Taiwan-based main assembler for Apple and is considering a bid with Sharp for about $27 billion.

    State-backed turnaround fund Innovation Network Corp. of Japan is also considering investing in the memory chip company.

    The last one is number one HDD maker Western Digital, already in flash chips following its acquisition of SanDisk. WD could get Toshiba activity for the better price ($16 billion?) - probably with the help of some Japanese banks to reassure the Japanese government - because its has a strong agreement with Toshiba concerning IP and their common chip plant in Japan. WD recently said that Toshiba breaching contract and wanted exclusive chip talks and looks to operate a veto over this flash deal. All other potential acquirers will be embarrassed by this partnership. But the company may not be able to dish out more cash so soon after SanDisk's acquisition.

    Nobody spoke about another potential acquirer: Seagate. This company fails to acquire SanDisk and could regain credibility with the acquisition of Toshiba chip business. But il will be more than the price paid by WD ($16 billion), something between $18 billion and $28 billion, and Seagate seems reluctant to spend so much money, being more inclined to protect its shareholders than to assure the long-term future of the company.

    But beyond this huge acquisition remain a big questions: what will be the destiny of Toshiba HDD and SSD businesses?

    Up to now, nothing was revealed by the Japanese firm. So here is our own opinion.

    Toshiba is not a leader in HDD or SSD but with recent significant growth in these two fields.

    Toshiba shipped 24 million HDDs in 1Q17 corresponding to a 24% worldwide market share, according to Trendfocus. For IDC, in this $25 billion segment in 2016, Toshiba's unit market share increased by 9% in 4Q16, reaching a total market share of 24%.

    For SSDs, IDC estimated that company's SSD revenue grew 114% year over year in 2016, getting the fourth place in total market share. Trendfocus calculates that the Japanese manufacturer got a 7.6% market share in units shipped and 8.2% in revenue in 4Q16 corresponding to the number four ranking in both categories.

    Toshiba SSD activity is worthless without its NAND chip manufacturing. An acquirer could only benefit of a portfolio of clients and a good controller technology following the acquisition of OCZ Technology in 2013 for $35 million.

    Which company could buy Toshiba's HDDs? Remember that Toshiba bought Fujitsu HDD in 2009 for 345 million. We don't see anyone entering into the decreasing HDD market and trying to compete with much bigger competitors Seagate and WD. So these two firms are the only one that could get their hand on this business. But why they will pay even one dollar for that? If Toshiba stops HDD manufacturing, all of its customers will go directly to the two other ones, being obliged to get their disk drives somewhere. Furthermore, Seagate and WD risk antitrust reviews by several countries, in particular the Chinese government that asks for a lot of conditions to accept the former mergers between WD and HGST on one side, Seagate and Samsung HDD on the other side. With only two HDD makers remaining on the market, OEMs will not be happy at all and prices of drives will have a good chance to increase with such a tiny competition.

    WD is in the best position to get Toshiba NAND chip business and why not to imagine the US firm looking for a complete package with the addition of Toshiba HDD and SSD businesses?

  4. #4
    WHT-BR Top Member
    Data de Ingresso
    Dec 2010

    Using Optane Memory Like A Hyperscaler

    Timothy Prickett Morgan
    March 28, 2017

    The ramp for Intel’s Optane 3D XPoint memory, which sits between DDR4 main memory and flash or disk storage, or beside main memory, in the storage hierarchy, is going to shake up the server market. And maybe not in the ways that Intel and its partner, Micron Technology, anticipate.

    Last week, Intel unveiled its first Optane 3D XPoint solid state cards and drives, which are now being previewed by selected hyperscalers and which will be rolling out in various capacities and form factors in the coming quarters. As we anticipated, and as Intel previewed last fall, the company is also rolling out Optane memory sticks in the M.2 form factor, which splits the difference between a traditional DRAM memory module and a non-volatile memory card that plugs into a more traditional PCI-Express x8 peripheral slot.

    The M.2 2280 form factor, which measures 22 millimeters by 80 millimeters, has room for a PCI-Express x2 slot and two fat Optane memory chips plus the electronics to make it look like a regular PCI-Express card. At the moment, the consumer-grade Optane M.2 cards that Intel has just announced come in 16 GB and 32 GB capacities and while we have only tangential performance information on these devices, we think that hyperscalers that have been using M.2 devices to boot operating systems and other software on their servers will be interested in using Optane devices as fast caches and extended memory where latency and total cost of ownership factors outweigh those of high capacity. The initial Optane 3D XPoint card comes with 375 GB of capacity and costs $1,520 at list price, plus another $431 on top of that if you want to pool the Optane memory with the DDR4 memory in a system using the Memory Drive Technology memory hypervisor. That works out to about $4 per GB for the raw Optane SSD, and that is about one-tenth the cost per GB of relatively skinny DDR4 memory sticks that are about one-tenth the capacity; the price/performance spread is even larger with fatter DRAM memory modules, which can cost nearly two or three times as much per unit of capacity. As we explained in our detailed analysis of Optane memory for servers, this Optane memory can be used to get more memory capacity for the dollar without sacrificing performance, or to act as a fast caching layer sitting in front of more capacious and less expensive flash or disk drive storage. Or both.

    But if you only need modest memory expansion, or just lower cost memory, or a smidgen of caching, and you can find a motherboard or a system with one or two M.2 slots, then these commercial grade M.2 slots could be very useful indeed – particularly if they are more widely available than the full sized Optane cards and drives.

    The M.2 memory sticks certainly are a lot less expensive. The 16 GB version of the M.2 2280 card costs $44, or $2.75 per GB, and the 32 GB unit costs $77, or $2.41 per GB. This is considerably less expensive than DRAM in equivalent capacities, and we think it could take off for more than booting system software considering that the M.2 devices provide the equivalent of a PCI-Express 3.0 x2 slot for connectivity into the compute complex. Four of these are as good as one Optane card in terms of connectivity back to the CPU and out to the network.

    For now, Intel is pairing the Optane M.2 sticks with its “Kaby Lake” 7th generation of Core processors for PCs, gamer systems, and workstations, and it has created systems software it calls Rapid Storage Technology that automatically pairs an Optane M.2 device with a flash or disk drive to create a unified tiered storage pool. This software keeps the hottest data on the Optane drives and over time it figures out what application files also need to be moved to Optane drives to be accelerated. There is no good reason at all that Rapid Storage Technology should not be allowed on servers, and should not be allowed to span as many M.2 devices paired to flash or disk storage devices as customers want. But thus far Intel has not said anything about this. You can bet hyperscalers and cloud builders are already playing with the M.2 devices, and more than a few storage array makers are probably trying to figure out how to make a shared storage device with lots and lots of M.2 cards that will ramp in the commercial space and drive the Optane cost down farther and faster than it will with its full-on PCI cards and drive form factor SSDs using 3D XPoint memory.

    The price/performance arguments that Intel is making for high-end PCs work just as well for servers.

    When it comes to PC and application responsiveness, Intel is telling customers that a PC equipped with 4 GB of DDR4 main memory plus 16 GB of Optane M.2 memory and a 1 TB disk drive will deliver better responsiveness on the SYSmark SE 2014 benchmark test than simply boosting the main memory to 8 GB and not using any Optane memory in from of that 1 TB drive.

    It is intriguing to think of what a server might look like next year when Optane DIMMs, which will plug into DDR4 memory slots, plus Optane M.2 cards sprinkled in and maybe a mix of Optane SSDs alongside flash drives and disk drives can all be put into the same machines. What will companies test, much less choose? Will they just go with the server parts as Intel proscribes or will they monkey around with consumer grade M.2 cards and get into to support them in server chipsets. On PCs, the older 100 series chipsets for Core processors cannot support the M.2 Optane devices, so it is not just a matter of having an M.2 slot. Intel is making a system play, and wants to use Optane to help encourage customers to upgrade to Kaby Lake Core chips in their PCs. It may push the Optane U.2 and PCI-Express cards in servers and deflect interest away from the less capacious and even less costly M.2 devices with “Skylake” Xeon processors coming this summer or so.

    It might not be long before people start making storage arrays – or storage servers if they have compute – based on M.2 devices, given the economics and the performance. One Stop Systems showed off an M.2 array with 224 flash memory sticks in 61 carriers back in the summer of 2015, and that experimental device had a capacity of 262 TB. A mix of Optane 3D XPoint and NAND flash using M.2 memory sticks could provide the right mix of performance and capacity at a very attractive price.

Permissões de Postagem

  • Você não pode iniciar novos tópicos
  • Você não pode enviar respostas
  • Você não pode enviar anexos
  • Você não pode editar suas mensagens