Data Recovery Wiki » About Flash http://www.hddwiki.org Data Recovery Training,Data Recovery Tools,Data Recovery Tutorial Wed, 01 Sep 2010 07:35:07 +0000 en hourly 1 http://wordpress.org/?v=3.0 Micron to make even teenier NAND dies http://www.hddwiki.org/index.php/2010/03/03/micron-to-make-even-teenier-nand-dies/ http://www.hddwiki.org/index.php/2010/03/03/micron-to-make-even-teenier-nand-dies/#comments Wed, 03 Mar 2010 05:14:32 +0000 admin http://www.hddwiki.org/index.php/2010/03/03/micron-to-make-even-teenier-nand-dies/ Not content with shrinking its NAND flash process to 25nm, Micron is heading below that level next year.We don’t know exactly how small the NAND dies are going to get butDigiTimesreports Micron will get there next year. The company hasn’t even started mass-producing its25nm productyet and it’s pretty advanced. As Micron says, 25nm is 3,000 times [...]]]> Not content with shrinking its NAND flash process to 25nm, Micron is heading below that level next year.We don’t know exactly how small the NAND dies are going to get butDigiTimesreports Micron will get there next year. The company hasn’t even started mass-producing its25nm productyet and it’s pretty advanced. As Micron says, 25nm is 3,000 times smaller than the diameter of a human hair.
The company is buying rivalNumonyxand partnering with Intel in the IMFT flash fabrication operation.With the 25nm process. Micron is planning to build 8GB flash dies,with 2 bits per cell. A sub-25nm process might make 16GB, 2bit per celldies possible, doubling NAND product capacity.

SanDisk and Toshiba are reported to be getting ready to build 24nm NAND in the second half of this year, moving on from their current 32nm process, and slightly leap-frogging Micron’s 25nm process. They’ll have both 2 bits per cell and 3 bits per cell product.

Analysts think SanDisk and Toshiba see 3 bit multi-level cell (MLC) as their preferred route to increasing flash capacity whereas Micron and Intel think process shrinkage is better. However Micron will produce 3bit MLC product, thinking it could be useful for solid state drives (SSDs) rather than portable media players and the like.

Micron is working on adding EZ-NAND (Error Correction Code – ECC – Zero NAND) flash, in which the NAND product does the ECC work instead of the host controller, simplifying the job of host controller manufacturers. EZ-NAND is part of the Open NAND Flash Interface (ONFI) Working Group’s 3.0 specification. This should, Micron hopes, make its flash better suited for portable and consumer electronics applications. It says its 25nm, 8GB flash product could store 2,000 songs or 7,000 photos.

The sub-25nm product could increase that to 4,000 songs or 14,000 photos.

The current Micron flash product uses floating-gate technology transistors but Micron may change this to a charge trap-based technology. A floating gate in such a die is surrounded by highly resistive materials so, once charged, it retains its charge. The charge trap idea is to change this to a sandwich of layers with the inner layer having its charge trapped between the other layers. This technology can be made at smaller dimensions than floating gate flash

Random Posts

]]> http://www.hddwiki.org/index.php/2010/03/03/micron-to-make-even-teenier-nand-dies/feed/ 1 Memory Cards – Info About Different Types http://www.hddwiki.org/index.php/2010/01/20/memory-cards-info-about-different-types/ http://www.hddwiki.org/index.php/2010/01/20/memory-cards-info-about-different-types/#comments Wed, 20 Jan 2010 15:17:47 +0000 admin http://www.hddwiki.org/?p=651 The different types of flash memory cards for use in digital cameras are: Secure Digital (SD), CompactFlash (CF), Memory Stick (MS), MultiMediaCard (MMC) xD-Picture Card (xD) and SmartMedia (SM).

The type of memory card you use is dictated by which digital camera you buy. These cards are physically different and are -not- interchangeable.

NOTE — NOTE — NOTE [...]]]> The different types of flash memory cards for use in digital cameras are: Secure Digital (SD), CompactFlash (CF), Memory Stick (MS), MultiMediaCard (MMC) xD-Picture Card (xD) and SmartMedia (SM).

The type of memory card you use is dictated by which digital camera you buy. These cards are physically different and are -not- interchangeable.

NOTE — NOTE — NOTE — NOTE — NOTE — NOTE — NOTE — NOTE

eBay has become notorious for “fake memory card” sales, thousands of people have been scammed by sellers who are selling inferior or lower capacity cards that have been relabeled and repackaged to look like the real thing. Buyer beware – I recommend that you purchase your cards only from reputable online vendors or brick and mortar stores. If you must buy on eBay then check out this page at Overclockers.com.au for ways to visually identify some (not all) of the most popular fake cards. Also see ourpublic discussion forum for more info.

Random Posts

]]> http://www.hddwiki.org/index.php/2010/01/20/memory-cards-info-about-different-types/feed/ 1 SD Technology Overview http://www.hddwiki.org/index.php/2010/01/20/sd-technology-overview/ http://www.hddwiki.org/index.php/2010/01/20/sd-technology-overview/#comments Wed, 20 Jan 2010 15:15:44 +0000 admin http://www.hddwiki.org/?p=649 SD memory cards are an innovative, always evolving bridge media powering the future of the digital world. SD memory cards deliver high-performance products that allow consumers to quickly capture video, photos, and sound in a reliable, easy-to-use format.

SD technology is the de-facto industry standard for mobile phones, digital cameras, MP3 music players, personal computers, printers, car [...]]]> SD memory cards are an innovative, always evolving bridge media powering the future of the digital world. SD memory cards deliver high-performance products that allow consumers to quickly capture video, photos, and sound in a reliable, easy-to-use format.

SD technology is the de-facto industry standard for mobile phones, digital cameras, MP3 music players, personal computers, printers, car navigation systems, electronic books, and other consumer electronic devices. SD technology is used in some 400 products across dozens of product categories and in more than 8,000 models.

SD memory cards are available in standard and high-capacity formats along with a variety of speed classes. A next-generation SD memory card specification, SDXC (eXtended Capacity), will soon be released to SD Association members who will create cards and devices that support its greater memory capacity and faster performance speeds. The cards have the broad interoperability and compatibility needed to ensure that its applications will continue to expand far into the future.

SDXC (eXtended Capacity) Cards: The Next Generation

The next-generation SDXC (eXtended Capacity) memory card specification, pending release in Q1 2009, dramatically improves consumers’ digital lifestyles by increasing storage capacity greater than 32 GB up to 2 TB and also increasing SD bus interface read/write speeds to 104 MB per second in 2009 with a road map to 300 MB per second. SDXC will allow your electronic devices – from laptops to cameras, camcorders and mobile phones – to store more content and to download content faster than ever before.

High-Capacity Cards

SD and SDHC memory card compatibility

The SD Association established a new specification rating for the minimum data transfer for standard and high-capacity cards, as well as both types of host products. With the ability to store between 4GB to 32 GB, you can do more at faster speeds across the entire SD line. Every high-capacity card is noted with “HC” to clearly designate this new capability.

With more storage, comes the need for faster speeds. The three Speed Classes allow host products to check the fragmented state in a card and calculate the write speed in each part of the card. This lets the host quickly determine where to write the data according to its speed requirement. The three SDHC speed classes — Class 2, Class 4 and Class 6 — transfer data at least 2, 4 and 6 MB per second, helping you to guarantee performance of your products. With the release of faster bus speed specifications, additional speed classes will soon be available to offer card and device manufacturers even higher guaranteed performance levels. SDXC, along with SDHC, Embedded SD and SDIO specifications, will benefit from the new performance specifications for SD interface speeds.

Copyright Protection

Ready for the Secured Digital Content

Content Protection for Recordable Media (CPRM), the content protection technology used for the SD card, is key to enabling a new distribution system for music and other commercial media. CPRM assures a high level of protection against illegal copying. The technology was developed by the 4C Entity, LLC (The digital contents copyright protection technology licensing organization of IBM, Intel, Panasonic, and Toshiba.)

Having a proven record in DVD, this protection is enhanced in SD memory cards through the use of “key revocation” technology built into each card.

The card’s control circuitry allows data to be read and written (in its protection area) only when appropriate external devices are detected. A check-out (copying) from a PC to the SD memory card is restricted to three copies in compliance with the SDMI specification. All SD-Audio products comply with SDMI.

The SD card copyright protection function has the following features:

  • Access to an SD memory card must be enabled by authentication between devices
  • A random number is generated each time there is mutual authentication and exchange of security information

Random Posts

]]> http://www.hddwiki.org/index.php/2010/01/20/sd-technology-overview/feed/ 6 What is a memory card? http://www.hddwiki.org/index.php/2010/01/20/what-is-a-memory-card/ http://www.hddwiki.org/index.php/2010/01/20/what-is-a-memory-card/#comments Wed, 20 Jan 2010 15:05:44 +0000 admin http://www.hddwiki.org/?p=647 A memory card is a device offering an easy, fast and reliable way for storing and transferring digital files. It works like a portable hard disk drive but comes with some superior advantages, since almost every memory card is based on flash memory technology, of having much smaller form factor and being non-volatile and solid state. Making [...]]]> A memory card is a device offering an easy, fast and reliable way for storing and transferring digital files.
It works like a portable hard disk drive but comes with some superior advantages, since almost every memory card is based on flash memory technology, of having much smaller form factor and being non-volatile and solid state. Making flash memory cards much more durable and reliable than hard disk drives. Memory cards are made by a variety of manufacturers and can be found in different storage capacities and transfer speeds.

As a consumer, it´s easy to get confused by all those different memories. Which memory card should I use for this application? And is one memory cards comparative with other kind of devises?

Where can a memory card be used?
Memory cards can be used by digital cameras, cell phones, camcorders, portable audio, DVD and video players, PDAs, game consoles, laptops, desktops, printers etc. In the future it is expected that all electronic devices will be compatible with memory cards. The driving force, today, behind the growth in memory card demand is the usage of memory cards in mobile phones. In the near future as storage capacity and transfer speed will get higher and prices lower, memory cards will replace most of today’s storage mediums such as the hard drive.

Different memory card formats
The memory card industry is high competitive with many memory card standards and formats that are competing with each other. There are currently more than a dozen of memory card formats available, and they are often not compatible with each other.

The first memory card was introduced in 1990 by the name PCMCIA and latter on even called for PC Card and was mostly used by laptop computers. Thanks to the flash memory’s durability, reliability and low power consumption demand for memory cards grew rapidly. In 1994 the CompactFlash was introduced as a smaller alternative to the PCMCIA/PC Card, mostly to be used in digital cameras. In 1996 the SmartMedia Card was introduced as a competitive standard to CompactFlash also to be employed by digital cameras.

The popularity of memory cards grow rapidly in the beginning of 2000, since usage of memory cards expanded to other consumer electronic products like camcorders, portable audio players, mobile phones etc. This created also the need for much smaller memory cards, therefore the first small form factor was the MultiMediaCard standard followed by the SD Memory Card and later on the Memory Stick.
Since most portable devices were becoming smaller and the usage of memory cards in mobile phones were becoming more common some new smaller versions ware introduced. Those were the RS-MMC or MMCmobile, miniSD, and Memory Stick Duo.

In the last years the driving force in the memory card industry shifted to ultra small memory cards, even those mostly to be used in mobile phones. Existed memory card standards introduced ultra small versions, their third form factor, of their existing standards. Those were the microSD/TransFlash, microMMC and the under development Memory Stick Micro. A brand new memory card standard was also introduced, the xD-Picture Card, only to be used by digital cameras. The latest development in the memory card industry was the announcement of the development of the Memory Stick Micro by Sony and SanDisk.

During the above period some other memory card standards have been introduced, like the USB Flash Drive as a replacement for the floppy disk. Some memory cards have also been introduced specifically to be used by game consoles such PS One, PS2, Xbox, Xbox 360 and GameCube. The latest development in the new consoles is the usage of regular memory cards by PS3 and Nintento Revolution.

Related Posts

]]> http://www.hddwiki.org/index.php/2010/01/20/what-is-a-memory-card/feed/ 12 Micron’s new flash dies live longer http://www.hddwiki.org/index.php/2009/10/27/microns-new-flash-dies-live-longer/ http://www.hddwiki.org/index.php/2009/10/27/microns-new-flash-dies-live-longer/#comments Tue, 27 Oct 2009 15:03:26 +0000 admin http://www.hddwiki.org/?p=584 Micron is sampling new flash dies that can write data many more times than other flash chips before they expire.

Micron calls its new products Enterprise NAND and is using a 34nm process to build them. It has a new 16Gbit single-level cell (SLC) die capable of up to 300,000 write cycles. Micron says this is three [...]]]> Micron is sampling new flash dies that can write data many more times than other flash chips before they expire.

Micron calls its new products Enterprise NAND and is using a 34nm process to build them. It has a new 16Gbit single-level cell (SLC) die capable of up to 300,000 write cycles. Micron says this is three times more than what it calls standard NAND chips. It can be combined in a package to produce a 16GB NAND unit.

There is a 2-bit multi-level cell (MLC) die with a 32Gbit capacity and a 300,000 write cycle rating. This is claimed to be a sixfold increase on standard NAND chips. A 32GB package is possible by combining the 32Gbit dies.

Micron says both chips support the ONFI 2.1 synchronous interface.

Solid state drive (SSD) controller startup SandForce offers a complimentary quote in a Micron release, saying it is “very excited to work with Micron and enable cost effective, reliable, high-performance SSD solutions that support stringent enterprise lifecycle requirements”. This implies Micron and SandForce are working together to produce complete NAND flash drives or modules.

In July Micron said it was going to build PCI-e-connected flash cards, working with IDT and its PCIe and memory interface technology. It looks like the cards are going to use SandForce controllers.

Micron is sampling the new chips now, with general availability slated for early 2010

Micron is sampling new flash dies that can write data many more times than other flash chips before they expire.

Micron calls its new products Enterprise NAND and is using a 34nm process to build them. It has a new 16Gbit single-level cell (SLC) die capable of up to 300,000 write cycles. Micron says this is three times more than what it calls standard NAND chips. It can be combined in a package to produce a 16GB NAND unit.

There is a 2-bit multi-level cell (MLC) die with a 32Gbit capacity and a 300,000 write cycle rating. This is claimed to be a sixfold increase on standard NAND chips. A 32GB package is possible by combining the 32Gbit dies.

Micron says both chips support the ONFI 2.1 synchronous interface.

Solid state drive (SSD) controller startup SandForce offers a complimentary quote in a Micron release, saying it is “very excited to work with Micron and enable cost effective, reliable, high-performance SSD solutions that support stringent enterprise lifecycle requirements”. This implies Micron and SandForce are working together to produce complete NAND flash drives or modules.

In July Micron said it was going to build PCI-e-connected flash cards, working with IDT and its PCIe and memory interface technology. It looks like the cards are going to use SandForce controllers.

Micron is sampling the new chips now, with general availability slated for early 2010

Related Posts

]]> http://www.hddwiki.org/index.php/2009/10/27/microns-new-flash-dies-live-longer/feed/ 0 Balancing act for servers that flash the cache http://www.hddwiki.org/index.php/2009/09/28/balancing-act-for-servers-that-flash-the-cache/ http://www.hddwiki.org/index.php/2009/09/28/balancing-act-for-servers-that-flash-the-cache/#comments Mon, 28 Sep 2009 11:18:54 +0000 admin http://www.hddwiki.org/?p=546 Multiple flash locations in the server-storage stack are upsettingbalanced I/O conventions and making overall system design much moredifficult.Designing and implementing server-to-storage systems is going tobecome very much harder, because existing assumptions about server andstorage I/Os per second (IOPS) handling are being swept aside.Virtualised, multi-core, cache-enhanced servers can digest and generatedata faster than many hard drive storage [...]]]> Multiple flash locations in the server-storage stack are upsettingbalanced I/O conventions and making overall system design much moredifficult.Designing and implementing server-to-storage systems is going tobecome very much harder, because existing assumptions about server andstorage I/Os per second (IOPS) handling are being swept aside.Virtualised, multi-core, cache-enhanced servers can digest and generatedata faster than many hard drive storage arrays can deliver and receiveit. How do you build balanced server-storage systems when this ishappening?
Let’s look at the basic server-to-external-storage stack. We startwith a server motherboard containing four, six, even eight-core serverswhich talk via memory and a PCIe bus to an I/O adapter. This linksacross a network to a storage array controller, which is connected tohard disk drive shelves. Imagine these are virtualised servers and itseasy to see, put crudely, a server engine room capable of handlingthousands of I/Os per second (IOPS) talking to a storage array onlycapable of handling hundreds of IOPS.To get the storage array IOPS capacity up to the server’s IOPS levelwill require a large increase in array spindles or solid state drive(SSD) use, and increased controller I/O capability. This can be done byusing flash memory caching but, unfortunately, flash storage can alsobe used as a server cache, in various places on the server side of thestack, to increase the server’s IOPS rate, thus unbalancing thingsagain.We can envisage server motherboards coming with flash modules onthem, acting as a cache between the server engine and its data store.Intel’s Braidwood technology is heading this way, it appears, and willincrease a server’s hunger for data. Then there could be a flash storeconnected to the server’s PCIe bus: another cache. Fusion-io and ViolinMemory both have such flash-based I/O acceleration cards, as doesOracle’s Sun-powered Exadata server. This caching also increases theI/O burden on the connected storage.Next there could be a flash cache on the I/O adapter used to connectto the storage. Adaptec is showing the way with its data conditioningideas and an Intel-supplied flash cache addition to its 5000 and 2000RAID controllers. We have three potential tiers of flash cache here allincreasing the burden on the storage array.Skipping across the network link to the storage array we could havea flash I/O accelerator card in the array controller. This is whatNetApp’s PAM (Performance Acceleration Module) is, admittedly in DRAMform initially, but now with a flash version announced, too. That makesfour flash locations with the fifth being SSDs actually replacing harddisk drives and providing a so-called tier zero of storage. Virtuallyevery storage array supplier is doing this, with the majority usingSTEC SSDs and a few settling on Intel.Finally the sixth flash location is a replacement for the hard drivearray itself. This is what Texas Memory Systems (RamSan), Sun(FlashFire), and also Violin Memory (1010 Memory Appliance with networkhead) are doing, and also what Fusion-io might do with its ioSAN. Thesecan be viewed as flash data stores, not caches, coming in eitherdirect-attached flash (flash DAS) or network-attached (a flash SAN)form.It seems a nonsense to have flash in all six locations. We might saywe could sensibly add a flash cache to servers and/or have either aflash store or a flash-enhanced storage array. The flash-enhancedstorage array would have a cache for hot data or else incorporate aflash store to augment a high-capacity and cheap bulk data SATA harddrive store.A system vendor – say Cisco (soon), Dell, HP, IBM or Oracle/Sun -would design systems to provide server and storage IOPS-handling anddata capacity in the most balanced, scalable and cost-effective way.Oracle’s Exadata 2 is an example of this. A server and storage systemintegrator and VAR will also need to do it, but not have the sameresources as tier one systems providers, and probably do a clumsierjob. A data centre operator buying servers and storage separately willalso have a complicated balancing act to accomplish balanced IOPShandling across its servers and storage. Tools are going to be needed.Its not clear where these tools will come from and it’s likely thatthe need to accomplish a technically different balancing act may helpdrive a focus on integrated server and storage systems from suppliersand away from the idea of treating servers and storage as relativelyseparate purchases. This aspect of flash could prove to be quitedisruptive

Related Posts

]]> http://www.hddwiki.org/index.php/2009/09/28/balancing-act-for-servers-that-flash-the-cache/feed/ 1 Could Micron buy Numonyx? http://www.hddwiki.org/index.php/2009/09/08/could-micron-buy-numonyx/ http://www.hddwiki.org/index.php/2009/09/08/could-micron-buy-numonyx/#comments Tue, 08 Sep 2009 15:06:28 +0000 admin http://www.hddwiki.org/?p=507 NAND flash chip manufacturer and Intel partner Micron could belooking to buy Intel offspring and NOR flash manufacturer Numonyx,according to anEE Timesreport.This would enable Intel to get out of Numonyx, Micron to get into the NOR flash business, and get its hands on Numonyx’sphase-change memorytechnology. Numonyx is a joint venture between Intel, which owns 45 per cent,and [...]]]> NAND flash chip manufacturer and Intel partner Micron could belooking to buy Intel offspring and NOR flash manufacturer Numonyx,according to anEE Timesreport.This would enable Intel to get out of Numonyx, Micron to get into the NOR flash business, and get its hands on Numonyx’sphase-change memorytechnology.
Numonyx is a joint venture between Intel, which owns 45 per cent,and STMicroelectronics, which has 49 per cent. Financial servicecompany Francisco Partners owns the rest, and invested $150m whenNumonyx was formed in 2008. The flash operations contributed by bothIntel and STMicroelectronics were unprofitable. It is not known ifNumonyx is profitable.NOR flash is typically used in mobile phones, whereas NAND flash isused more in general computing products. Smart phones are tending toconsume more and more NAND.Intel and Micron are partners in IM Flash Technologies which is aNAND chip foundry. At one time it was thought that Intel wanted to beshot of the flash business altogether. The setting up of Numonyx and IMFlash Technologies was seen as laying stepping stones to a flash-freestate.Now Intel is selling disk drive replacement X18 and X25 flashmodules as well as developing flash-on-motherboard Braidwoodtechnology. Its partner Micron is developing PCIe-connected flash solidstate drives, and we hear no more flash divestiture talk from Intel.However, it may be quite happy to buy flash dice from arms-lengthfoundries and trusted partners.Phase-change memory (PCM), which stores binary digits via statechanges in cells rather than as a magnetic direction, can scale down tomuch smaller process geometries than NAND flash which is forecast tohit a wall in a few years time. PCM also has the promise of beingfaster to access than flash and be byte-addressable instead of havingto be written to and erased in blocks.The NAND flash market has recently seen a firming up of prices after foundry over-capacity has been mothballed.Both Micron and Numonyx say they don’t comment on rumours.

Related Posts

]]> http://www.hddwiki.org/index.php/2009/09/08/could-micron-buy-numonyx/feed/ 0 IBM beefs up midrange storage arrays with 8 Gig FC, iSCSI, self-encrypting drives http://www.hddwiki.org/index.php/2009/08/27/ibm-beefs-up-midrange-storage-arrays-with-8-gig-fc-iscsi-self-encrypting-drives/ http://www.hddwiki.org/index.php/2009/08/27/ibm-beefs-up-midrange-storage-arrays-with-8-gig-fc-iscsi-self-encrypting-drives/#comments Thu, 27 Aug 2009 12:22:38 +0000 admin http://www.hddwiki.org/?p=468 IBM has upgraded its midrange disk array storage systems, rolling out the IBM System Storage DS5020 Express with support for 8 Gbps Fibre Channel (FC), iSCSI and self-encrypting drives.

The DS5020 Express is based on LSI Corp.’s Engenio 4900 that IBM sells through an OEM deal. It replaces the DS4700, which IBM plans to sell through the [...]]]> IBM has upgraded its midrange disk array storage systems, rolling out the IBM System Storage DS5020 Express with support for 8 Gbps Fibre Channel (FC), iSCSI and self-encrypting drives.

The DS5020 Express is based on LSI Corp.’s Engenio 4900 that IBM sells through an OEM deal. It replaces the DS4700, which IBM plans to sell through the rest of 2009 and support for at least five years.

Harold Pike, IBM’s DS5020 offering manager, said the new unit is designed for virtualized environments that support multiple applications. “When you do that, what you really need is a product that has tremendous mixed workload capabilities,” he said.

The System Storage DS5020 Express supports up to 112 FC or SATA drives with EXP520 expansion units, and up to eight 8 Gbps FC or 1 Gbps iSCSI host ports. The DS4700 only offered up to four 4 Gbps FC host connections. The 5020 also provides auto-negotiate Fibre Channel connection speeds so it can be integrated into an existing 2 Gbps or 4 Gbps environment, and adjust when 8 Gbps FC is implemented.

Mark Peters, an analyst at Milford, Mass.-based Enterprise Strategy Group, said storage vendors are under pressure to produce more powerful systems in virtualized environments without large price increases. “There’s a big emphasis in the midrange right now on adding functionality without adding price,” he said. “The nature of the world is that things get bigger and faster and more capable, and this is bigger and faster and more capable than the DS4700 was.”

The System Storage DS5020 Express also supports up to 4 GB of cache memory, self-encrypting drives from Seagate Technology Inc., hot-swappable capacity expansion, up to 128 storage partitions, as well as dual-redundant, hot-swappable cooling fans.

The midrange system competes primarily with EMC Corp.’s Clariion CX4 Model 120 and Hewlett-Packard (HP) Co.’s StorageWorks 4400 Enterprise Virtual Array.

While IBM has added 8 Gbps FC capability to this model, iSCSI could be the protocol to grow most in customer adoption. The firm’s Pike said iSCSI interest is “really starting to hit now” in the midrange. “It’s a segment of the market that we think is very, very important and growing very rapidly,” he said.

The IBM System Storage DS5020 Express is available for shipment on Sept. 4. Pricing starts at $22,500.

Related Posts

]]> http://www.hddwiki.org/index.php/2009/08/27/ibm-beefs-up-midrange-storage-arrays-with-8-gig-fc-iscsi-self-encrypting-drives/feed/ 0 In the spin of SSDs on database servers http://www.hddwiki.org/index.php/2009/08/25/in-the-spin-of-ssds-on-database-servers/ http://www.hddwiki.org/index.php/2009/08/25/in-the-spin-of-ssds-on-database-servers/#comments Tue, 25 Aug 2009 11:23:39 +0000 admin http://www.hddwiki.org/?p=460 Interest in solid state drives (SSDs) is growing as their initially terrifying price plummets: Intelrecentlyannounced a 60 per-cent reduction in the cost of its X25-M Mainstream SATA SSD since its launch last year.SSDs are attractive because they radiate less heat than traditionalspinning disks, they’re smaller, quieter and consume less energy. And,of course, they’re fast. So naturally [...]]]> Interest in solid state drives (SSDs) is growing as their initially terrifying price plummets: Intelrecentlyannounced a 60 per-cent reduction in the cost of its X25-M Mainstream SATA SSD since its launch last year.SSDs are attractive because they radiate less heat than traditionalspinning disks, they’re smaller, quieter and consume less energy. And,of course, they’re fast. So naturally we bung them into laptops.
However, there’s growing use of SSDs in servers, and particularlydatabase servers. Teradata, one of the world’s leading businessintelligence companies, recently unveiled a prototype data warehouserunning entirely on SSDs.Do the same reasons reasons for using SSDs on a laptop also apply todatabase servers? And what do SSDs mean for the applications that willaccess the data they hold?Performance improvement is certainly top of the list of appealingfactors, but uptake is also being driven by energy issues. Keepingservers cool is an ongoing battle so anything that reduces the heatgeneration is welcome, so how do the two technologies stack up?A 600GB HDD might use about 16 watts when active: that’s about 30Wper terabyte. The X-25M has an active power consumption of 150mW.Running six of the 160GB SSDs gives a figure of about 1W/TB – 30 timesless. If that wasn’t reason enough, the cooler disks can be packed moreclosely together saving space, another server-room plus point.So, what of the speed? To read from a particular place on disk theread/write head in an HDD must move to the right position, which takes”seek” time – quite variable but let’s say an average of fourmilliseconds. Once the head is in position there is a “latency” as thesector containing the data crawls painfully around to it’s appointedposition under the head.A 15,000 rpm HDD might have an average latency of two milliseconds.An SSD with no moving parts doesn’t really have equivalents but latencymay be quoted – for example a recent SSD was rated at 65 microseconds.Quick serviceAll the advantages of using SSDs for general servers hold true fordatabase use, but there are other potential advantages. Databases havealways, if you ignore punched cards, used rotating media and whilstHDDs aren’t serial access, they certainly aren’t random access either,so all manner of database mechanisms have been developed to improvedata access performance. I’ll pick on just two: indexing and column/roworiented databases.A non-clustered index is often a sorted list of pointers to data andfor performance reasons should be kept on a separate spindle. Of courseif the table is small enough, and particularly if it’s marked asread-only – as it may well be for analysis – any database engine worthits salt will run a full table scan, pop the table into memory andignore the index. Why? Because memory is much faster than disk.Intelligent DBAs understand all this and modify their indexingstrategies accordingly.Moving down the scale, there are doubtless DBAs who index all fieldsneeding fast access regardless of table size, those who do this andkeep the indexes on the same disk as the data and even those who don’tknow about indexes.

Related Posts

]]> http://www.hddwiki.org/index.php/2009/08/25/in-the-spin-of-ssds-on-database-servers/feed/ 11 Best practices for solid-state drive storage technology http://www.hddwiki.org/index.php/2009/08/15/best-practices-for-solid-state-drive-storage-technology/ http://www.hddwiki.org/index.php/2009/08/15/best-practices-for-solid-state-drive-storage-technology/#comments Sat, 15 Aug 2009 16:11:34 +0000 admin http://www.hddwiki.org/?p=433 Until the costs drop even further, performance-boosting solid-state drives (SSDs) won’t make economic sense for every type of application, so it’s important to follow best practices to ensure they are working for your storage shop. Here are 10 SSD best practices to follow:1. Identify I/O-intensive applications that will benefit from faster data storage.

Types of applications [...]]]> Until the costs drop even further, performance-boosting solid-state drives (SSDs) won’t make economic sense for every type of application, so it’s important to follow best practices to ensure they are working for your storage shop. Here are 10 SSD best practices to follow:1. Identify I/O-intensive applications that will benefit from faster data storage.

Types of applications that may be well suited to SSD technology include databases, data mining, data warehousing, analytics, trading, high-performance computing, server virtualization, Web serving and email systems.

A check of how many enterprise-grade 15,000 rpm and 10,000 rpm hard disk drives are in use and how much money has been spent on DRAM for performance will help to determine if SSDs will be worth the investment.

Study application workloads and assess hot vs. warm vs. cold data sets. Active data can be directed to the flash solid-state drives, and the less frequently accessed data can go on Fibre Channel (FC) drives or SATA drives.

“If you have a good understanding of that, then you can understand how much solid-state storage you’re likely to need to be able to optimize the performance of your system,” said Jim Handy, an analyst who focuses on memory chips and SSDs at Objective Analysis.

2. Ensure that applications, especially those that are custom-written, can handle the faster solid-state drives.

“For most applications, this should not really be a problem, but depending on coding and timings, you can have the chance to have things done too quickly which can throw off timing a bit, as well as the processors actually jumping due to it not having to wait for the disk anymore,” said Jon-Carlos Mayes, IT director at CCP hf, a Reykjavik, Iceland-based online game producer. CCP uses both DRAM and NAND flash SSD technology from Texas Memory Systems Inc.

Random Posts

]]> http://www.hddwiki.org/index.php/2009/08/15/best-practices-for-solid-state-drive-storage-technology/feed/ 0