Card has a maximum sequential read speed of 80 megabytes per second and could be a good fit for LTE smartphones and tablets

By Mikael Ricknäs

Samsung Electronics has started mass producing a microSD card that uses an Ultra High Speed-1 (UHS-1) interface to improve data transfer speeds in LTE smartphones and tablets.

The microSD HC card stores up to 16GB and has a maximum sequential read speed of 80MBps (megabytes per second), according to internal tests conducted by Samsung. That is more than four times the read speed of today’s advanced microSD cards, which have speeds up to 21MBps, Samsung said.

What real-world speeds that will translate into remains to be seen, although Samsung said that the card will be a good fit for LTE smartphones and tablets.

Cards that can store more than 16GB and offer the same are speeds on the way, according to Samsung.

Samsung didn’t say when the card will start shipping or what it will cost.

Task force aims at interoperability rules

By Lucas Mearian

Fifty industry representatives have signed on to a task force created to address interoperability issues related to a plethora of form factors and connectivity issues for PCIe solid-state drives (SSDs).

The Solid State Storage Initiative (SSSI) PCIe SSD Task Force, which is being organised under the Storage Networking Industry Association (SNIA), will deal with current standards, any standards gaps that need to be filled and end user concerns over interoperability between products.

Solid-state storage devices come in so many shapes and sizes that the electrical signal standards and OS drivers can vary from manufacturer to manufacturer. Unlike SAS and SATA interconnects, which only increase in speed with each new generation of product, PCIe is a good fit for SSDs, as it allows the interface speed to be increased quickly by adding PCIe lanes, according to Paul Wassenberg, chair of SNIA’s SSSI.

But not every SSD or systems manufacturer is speaking the same language when it comes to PCIe, Wassenberg said.

For example, standard SAS and SATA drives plug into a south gate on an Intel motherboard. If you plug that gate into a PCIe bus, it doesn’t automatically show up as a mass storage drive on the computer system. A software driver is required for the operating system to recognise the PCIe SSD as a mass storage device, according to Eden Kim, chair of SNIA’s Solid State Storage Initiative’s technical group.

In addition, in array environments where multiple drives are used for higher capacity and data resiliency, PCIe SSDs do not act the same way as SAS or SATA drives. For example, they cannot be swapped out without disrupting the system. And, in consumer computers, laptops and desktops, there is no easy way to gain access to PCIe cards, Kim said.

“That’s a motherboard layout issue,” Kim said. “So now you can have a cable coming from your card in the PCI slot on the motherboard to your drive bay that allows you to plug in a PCIe 2.5-in form factor SSD. That’s the Micron product.”

Currently, the SSSI recognises three categories of PCIe devices: low-end SSDs for tablets, laptops and desktops that may use one or two PCIe lanes, small PCIe cards that use four lanes and full-height cards that use eight lanes.

“This is about getting the nomenclature set so that people can talk and don’t have a Tower of Babel,” Kim said. “It’s more about organising what’s out there into cognitive blocks, and providing guidance to the SSD marketplace on PCIe SSDs. This can take the form of educational materials, best practices documents and SNIA standards.”

For example, there are already a number of standards groups focused on SSD electrical signalling, including NVMHCI, SATA Express and the Proposed PCI class code assignment for SOP.

The task force will hold its first meeting April 9 and meet on a bi-weekly basis after that. It is structured as an open industry forum for the first 90 days, and thereafter as a SSSI committee. No NDA is required and thus no confidential information is expected to be shared among task force members.

Firefox 12 will be last browser version to run on operating system

By Gregg Keizer

Mozilla on Friday announced that next month’s Firefox 12 will be the last version to run on early editions of Windows XP and the 12-year-old Windows 2000. The company also reiterated that it will stop serving security updates for 2010’s Firefox 3.6 in April.

Starting with Firefox 13, the browser’s minimum requirements will be XP Service Pack 2 (SP2). Firefox 13 will not work on Windows 2000, Windows XP RTM (release to manufacturing, the original mid-2001 build) or XP SP1.

Firefox 12, set to ship April 24 and due to be replaced by the next edition on June 4, will be the last that supports the three older Windows. “This support change allows us to significantly improve Firefox performance on Windows by using a more modern build system,” Mozilla said.

The decision wasn’t a surprise: Mozilla has been discussing the change for at least three years. And the company actually pulled the trigger two months ago, when Asa Dotzler, director of Firefox, explained the firm’s reasoning.

“Our developers have not been able to take advantage of new compiler features [in Visual Studio 2010] and have had to struggle to keep valuable optimizations from breaking, including having had to back out and ultimately delay some important new features like SPDY,” said Dotzler in late January. “Our users have suffered a slower Firefox than would be possible as both direct and indirect results of moving to a more modern compiler.”

SPDY, for “speedy,” is a Google-crafted protocol that promises faster and more secure page loading. Mozilla added support for SPDY in Firefox 11, the March 13 release.

By switching to Visual Studio 2010, Mozilla will not be able to build Firefox for the older operating systems, said Dotzler. But it’s not as if Mozilla has jumped the gun, as Microsoft retired all three editions years ago. Windows 2000 fell off Microsoft’s support list in mid-2010, and XP and XP SP1 were dumped in 2004 and 2006, respectively. Microsoft doesn’t even support Windows XP SP2.

The only version of Windows XP still backed by Microsoft with security updates, including patches for Internet Explorer 6 (IE6), is SP3, which released in 2008 and has two years of support life left.

Mozilla advised Firefox users still running Windows XP RTM or XP SP1 to migrate to a newer operating system, Windows XP SP3 is a free upgrade, and urged Windows 2000 customers to do the same.

Dotzler also steered Windows 2000 users toward a rival. “If you’re a Windows 2000 user and you simply cannot upgrade your PC to a more modern Windows version, I’m sure the good folks over at Opera [Software] will be happy to help you out,” said Dotzler. Opera runs on Windows 2000, but its Norwegian maker recommends XP or later.

Firefox 3.6.28, which Mozilla shipped March 13, is the last planned update for the two-year-old browser. Between now and April 24, Mozilla will only release fixes to 3.6 if developers uncover critical security or stability issues.

Mozilla advised Firefox 3.6 users to upgrade to the current edition, or failing that, to Firefox ESR (Extended Support Release), the build that targets enterprises leery of upgrading browsers every six weeks.

Become an expert on flash memory

By Matt Prigge

Solid-state disks (SSDs) are hardly new, but their growing usage represents a significant shift in the primary storage landscape. SSDs have been increasing in capacity and decreasing in cost at an accelerating rate, so the chances that you’re going to bump into them in the wild are climbing as a result. However, SSDs are not perfect. A solid understanding of their history and differentiating factors will help you debunk some of the hype and leverage them more effectively in your environment.

The idea of a solid-state disk has been around for a very long time. Essentially, an SSD is persistent storage media constructed using transistors rather than an electromechanical disk or tape. SSDs have been holding the firmware for our switches, routers, cell phones, calculators, and just about any other kind of non disk persistent memory for easily 30 or more years.

What is different today is that we’re well down the path of using these SSDs in our enterprise primary storage environments — either augmenting traditional disks or replacing them completely. This type of application for SSD hasn’t been possible until recently due to the tremendous difficulty involved in constructing very large SSD memory modules that are cheap, reliable, and fast and that have a long lifetime. We’re still working to overcome some of these challenges, and being aware of them is key to implementing them successfully.

Volatile versus non-volatile SSDs

The biggest distinction to make right off the bat when talking about SSDs is whether they are volatile DRAM-based devices (RAM storage) or non-volatile NAND memory (flash storage) devices. They both often fall under the SSD moniker, so it’s easy to get them confused.

DRAM-based devices essentially use the same type of memory that makes up the primary system memory of your server; they are both extremely fast and susceptible to total data loss if power is interrupted for some reason. To combat this, most DRAM-based SSD devices require a battery backup to power the memory and ensure data integrity until power is restored.

In some cases, this battery backup is a super capacitor that can power the device for a few days; this is common in very high-performance DRAM SSDs that ship in a PCIe card factor. However, in the event that power isn’t ever restored, your data probably won’t be, either. In other cases, the DRAM is paired with an equal-capacity array of hard disks or slower NAND flash memory in a rack-mount chassis that is used to stage and de-stage the DRAM memory during power up and power down (with an internal bank of batteries or capacitors providing enough power to perform the de-stage operation in the event of an unexpected power outage).

NAND-based (flash) devices use the same general breed of memory found in cell phones and USB sticks. These memory devices do not require power to hold their state. Thus, they don’t require a battery backup of any kind to ensure data integrity. On the other hand, they’re several times slower than DRAM-based devices, though their speed is improving as the devices and their controllers mature.

MLC versus SLC SSDs

NAND devices come in two major flavours: MLC (multilevel cell) and SLC (single-level cell). MLC devices are so named because they can store a few bits of data in the same cell, whereas SLC devices can store only a single bit of data per cell. SLC devices are much more expensive to make because they require more transistors to store the same quantity of data, but they’re significantly faster and have a longer lifespan than MLC devices.

Most consumer-grade SSDs, such as the one in your fancy new laptop, are likely MLC devices. In those applications, low cost, lower power usage, and higher reliability when dragged off your coffee table by your dog are key concerns, while raw performance is not. Any enterprise-grade NAND-based storage device is likely to be SLC-based – and much more expensive as a result.

It’s all about the controller

As with traditional primary storage devices, NAND flash-based SSDs live or die by the functionality delivered by their controllers. The strength of the on-device controller represents a significant cost in delivering an enterprise-grade SSD, but the controller is also responsible for providing exceptional (or less than exceptional) reliability and performance. In addition, it’s where significant technology growth and innovation is still taking place.

Unlike DRAM-based SSDs, flash-based SSDs suffer from long-term reliability issues. Individual single-bit SLC SSD cells usually wear out after about 5 million write cycles. Multibit MLC cells become unreliable after just 500,000 to 1 million write cycles.

To combat this issue, the SSD controller performs a technique called write-levelling that tries to spread writes across the cells that make up the SSD, to ensure relatively equal write-load distribution. Additionally, some SSD controllers use slack (unprovisioned) space that can take over from cells that are near the end of their expected lifetime.

Some cheaper controllers perform this write-levelling without regard to how much load the device is under, while others wait until the device is under lower load before running. This is one of many reasons that benchmarking performance on SSDs is difficult: Performance often looks stellar for a few hours until the write-levelling algorithm starts running, but then it crashes and burns.

Write-levelling can also have some unintended security side effects. Let’s say you have an unencrypted file and then decide to encrypt it. As you do this, your server reads the unencrypted file from disk, encrypts it, and writes the encrypted file over the unencrypted file — usually deleting it in the process. Due to how some write-leveling techniques work, it’s possible for your server to believe that the unencrypted files have been overwritten when in fact they have not. Some controllers honour these requests and erase the blocks, while others do not.

Putting it all together

As you start digging into SSDs and deciding whether they’re right for your primary storage environment, remember that they are a completely different animal than traditional spinning disks and are still undergoing growing pains. To be sure, the enormous performance potential of SSDs will ensure that they will be an option for IT for many years. Just don’t charge into SSD usage without understanding how they work, so you don’t have nasty surprises in your production environment.

By Rachel King

Summary: Intel sells a pair of its Flash memory factories to Micron as part of agreement to advance their NAND Flash memory joint venture relationship.

Intel and Micron’s relationship took a new turn this week in the name of advancing Flash memory technology.

Notably, Intel has agreed to sell two of its Flash memory factories to Micron for the price of $600 million.

Intel will get back approximately half of the consideration amount in cash. The rest will be deposited with Micron, which may be refunded or applied to Intel’s future purchases under the NAND Flash supply agreement.

News of the deal comes as part of the announcement that Intel and Micron plan to expand their existing NAND Flash memory joint venture relationship.

The two factories in question are located in Virginia and Singapore. There is a third factory in Utah that operates under joint venture, but its involvement will remain unchanged.

Goals for the continuing partnership include Micron increasing its share of the global NAND Flash market. Micron will also continue to supply Intel with NAND Flash memory from its facilities.

Robert Crooke, corporate vice president and general manager of Intel’s Non-Volatile Memory Solutions Group, explained in prepared remarks that “the new NAND Flash supply agreement with Micron gives Intel better flexibility to meet growing demand for SSDs and other products.”