Last week’s Flash Memory Summit was a whirlwind of tutorials, announcements and networking, including a presentation by our very own CEO Roy Sherrill, but if you missed the announcement by ONFI (the Open NAND Flash Interface organization) that its new ONFI 2.3 specification will include the EZ-NAND protocol, take a few minutes to soak it in. The news about EZ-NAND, which stands for error correction code (ECC) Zero NAND, is especially welcome for manufacturers of smart phones and other high-performance, price-competitive embedded devices. The EZ-NAND protocol dictates that ECCs are done by the NAND itself rather than by the NAND controller or device driver, a much more efficient method (as we’ve been  blogging for nearly two years now). Also, in this era of fast-changing ECC NAND requirements, not having to keep up with the rapidly changing NAND ECC requirements by using costly high-end NAND controllers is a major competitive advantage for OEMs.

In current raw NAND implementations, the host controller is responsible for managing ECC and all other NAND functions like wear leveling, bad block management and compaction. ECC is notoriously difficult to implement in software, and EZ-NAND decouples it from the rest of these functions, allowing device designers to use the system processor for these functions and avoid the use of expensive NAND controllers. This setup, combined with a high-performance flash manager like FlashFX Tera, offers OEMs the ultimate combination of performance, flexibility and low cost to make those high-octane smart phone processors (MIPS, PowerPC, iMX-31 and OMAP families, etc.) really shine. While eMMC, with its hardware-based ECC seems to offer good compromise, the reality is that the other NAND functions (wear-leveling, bad block management and compaction) have to contend with the inexpensive controller found in the eMMC device. Translation: less-than-optimal use of the flash and seriously bogged-down performance in many use cases.

We would love to hear what you thought of this year’s Flash Memory Summit, what you learned, what surprised you and what you liked or didn’t like.  Also, let us know if you have comments or questions about EZ-NAND using the comment field below.

Earlier this week we announced XCFiles, a design-ready exFAT-compatible file system for next-generation extended capacity SD (SDXC) cards. Visit Datalight.com to read the full announcement or learn more about our exFAT implementation

Bothell, WA, – June 15, 2010 – Today Datalight announced support for Micron Technology’s 4-gigabit (Gb) 34-nanometer (nm) NAND flash with on-die error correction code (ECC) within its popular flash managers, including FlashFX Pro and FlashFX Tera.

Read the full story here: http://bit.ly/9xn2DW

While the software that we develop integrates at system level in devices, we do keep an eye open for developments happening at the application level primarily because the heavy data storage use most modern applications make. Understanding what kind of applications are making headway in the market helps us determine the type of data profile to expect at the file system and flash management level. This helps is in creating new system level functionality that benefits all applications. One example of this is the work done in Reliance Nitro. We observed a lot of device applications and how they use files. We found that apart from some data logging / music playback apps, most apps do not indulge in sequential reads and writes that often. Most of the time, apps are primarily engaged in file operations (file open being the most common). Hence we designed the tree-based metadata architecture in Reliance Nitro to provide radical performance improvements in the file operation speeds.

When we talk about apps on devices, we cannot ignore the 800-pound gorilla – the Apple Appstore. The appstore has created a huge ecosystem for apps on the iPhone and the iPod touch by having a one-stop shop for discovering, reviewing and purchasing apps. Combine this usability with the popularity of the hardware, the app ecosystem has blossomed which has forced rivals to pay attention and start something similar of their own. Google’s Android marketplace is probably the next biggest player while Palm, Microsoft and Nokia are just getting started. While Apple does indeed have a head start of 12-14 months, what is interesting to note is the huge difference between it and its closest rival. This article by Fierce Developer highlights the significant difference in revenue that app developers are seeing on the Apple v/s the Android platform.

Source: AdMob’s July 2009 Mobile Metrics Report

What do these stats mean for device developers? In order to catch up or surpass Apple, innovation needs to happen on multiple fronts

1. Device hardware: Apple innovated with the capacitive touchscreen and no-keyboard design. OEMs need to find the next game changer on the hardware front

2. Device user experience: Hardware on its own does not win over users. There have been several iPhone competitors who have had better hardware of paper but without matching software capabilities, the devices have not met with spectacular success. Using an optimized data storage stack can help improve device responsiveness which users rank as one of their top requirements. Creating new simple user interface paradigms will also be critical.

3. Developer support: Use of standard tools and languages for writing apps along with a low cost of entry is critical to attract app developers to your platform

4. Discovery and purchase of apps: How do users of your devices find and purchase apps? How can you do it better and faster? What is the average price-point for the buyer profile of your device. All of these are critical questions to answer before creating an app store.

It will need strong execution on all of the above fronts for device manufacturers to catch up to the Apple success. We will be keeping an eye open to developments here; look out for further postings on this topic in the future.

Attendance at FMS broke records for the fourth year in a row – truly remarkable, given our current economic situation. I think even the organizers were pleasantly surprised by the confirmation that FMS is now the venue of choice for flash industry leaders to come together and for the rest of the industry to learn what is new in the world of flash technology. As a sponsor of the show since its inception, Datalight contributed to the Summit by organizing a half-day executive update, presenting on wear-leveling, moderating a round-table on endurance at the beer & pizza forum, and delivering a product demonstration at the FMS Theater.
Once again SSDs, thought by many to be the savior of the beleaguered flash industry, dominated the agenda. This year talk of application-specific SSDs was everywhere, specifically optimizations for interfaces like SAS, SATA, and fiber channel.

Presumably, the idea behind application-specific SSDs is to drive faster adoption by tuning them for the use cases presented by different applications. For example, in a random write intensive application, additional RAM may be added to mitigate lackluster random write performance by buffering writes.
As expected, there was a fair amount of preaching to the choir about the benefits of SSDs over HDDs, including read/write performance and the power savings. Intel touted 17% longer battery life and reduced failure rates (less than .5% for SSD vs. 5% for HDD). Other manufacturer’s boasted instant-on boot time, smaller footprint, and lower DRAM requirements.

On the topic of barriers to adoption, SSDs were declared still too expensive for mass market appeal. Also, many of the benefits promised by SSDs are being undone by bottlenecks in the OS (and file system!). As a result, it was postulated that SSDs will cannibalize 15k RPM HDDs, where customers are willing to pay a premium for speed, rather than replacing HDDs as a whole, at least in the short term,
In the emerging technologies track, 3-D NAND was promoted by both Micron and SanDisk, in part because they are able to leverage the technology used to build DRAM for manufacturing 3-D NAND, making it easier and cheaper to produce. It was particularly interesting to hear SanDisk say that 4 bit per cell will have limited application in the future, especially since it had been speculated that the 4 b/c technology was a key reason for them to acquire M-Systems. PCM and FeRAM were also discussed as next generation flash technologies. Overall, we were impressed by the level of participation from across the industry and we’re already looking forward to next year’s Summit!

Denali’s Lane Mason has a very good article on NAND flash pricing. Strongly recommended reading.

http://www.denali.com/wordpress/index.php/dmr/2009/07/16/nand-forward-prices-rate-of-decline-will

It’s been over 20 years since NAND flash was supposed to take over the world, so why is the ‘ultra-portable’ laptop I’m writing this on still using HDD? And why, nearly 30 years after the invention of flash, are we still debating its applications vs. the rotating platter? Given the performance advantages of flash and the mechanical shortcomings of HDD – things like wear and susceptibility to shock and vibration, not to mention the latency associated with spinning up for reads – it’s no wonder that flash has long been the darling of memory technology futurists. So why after all this time has flash not replaced older technologies? I’ll give you a hint: It’s all about the money. But it’s not only about the money – check out our list of five top barriers to flash adoption:

1. Cost – Despite years of oversupply in the flash market, and the corresponding reductions in price, flash is still relatively expensive when compared to HDD, especially on a $/bit basis. To make matters worse, the current economic climate has taken its toll on the flash industry, spurring several rounds of consolidation and requiring flash vendors to curb manufacturing costs by shrinking portfolios and closing fabs. Predictably, these changes in the supply landscape are causing prices rise in many cases, making the cost factor an even bigger problem for flash.

2. Shrinking lithography = lower endurance – One way for flash manufacturers to remain competitive is to use smaller die size to reduce raw material costs. Just a couple years ago, the vast majority of NAND flash was manufactured with 90nm lithography. Most vendors are now planning to move to 30nm technology either this year or next. An unfortunate side-effect of smaller lithography is significantly decreased endurance. SLC NAND, which had 100K + erase cycles, is now predicted to be in the 50-70k range. The biggest impact is on MLC NAND where the endurance has gone from 10k erase cycles to around 3k (a 70% reduction!).

3. Increasing ECC – Another side-effect of shrinking lithography is an increase in error rates for flash, requiring stronger correction codes. Most SLC NAND flash today requires 1-bit correction. That number is predicted to increase to 4-bit on 30nm NOR parts. And the ECC outlook for MLC NAND is even worse, requiring ECCs greater than 12-bit (compared to 4-bit or 8-bit today). These increased ECC requirements mean the controller design for managing flash will become more complicated, and more difficult for OEMs to implement. Performance will also be impacted, especially if the ECC is done in software running on the host processor.

4. Vendor volatility – Churn or volatility in the flash market, the products of a difficult economic climate, are making it difficult for OEMs to find a reliable source of flash parts. Examples are everywhere; A major flash supplier is currently under Chapter 11. There are merger talks happening between SanDisk and Samsung. Asian vendors have been hit especially hard, particularly those also in the DRAM business. OEMs are rightfully concerned about interruptions to their production cycles in the midst of all this turmoil.

5. Lack of killer application – While NAND flash densities have continued to increase, the industry is still waiting for the killer application to gobble up these immense quantities of flash. For long SSDs have been viewed as that application but they have not taken off as fast as the flash industry would have liked.

In spite of the obstacles faced by the industry, flash remains a strong and growing choice for data storage and has put breakthrough devices like MP3 players and smart phones (iPhone!) into the hands of millions of consumers. Early adopters of SSD technology in laptop computers, netbooks and enterprise applications are making a solid case for mass market potential there, which should significantly drive flash adoption in the next few years. Visit the FlashFX Tera page to learn how Datalight is making flash easier and more competitive.

We have talked about managed NAND in a few blog posts before. Usually a combination of raw NAND flash (SLC or MLC) combined with a hardware controller that performs flash management features like bad block management, ECC and wear leveling is referred to as managed NAND. The term covers a huge spectrum of flash-based storage devices so in this post we will try and highlight some of the more prevalent types of managed NAND

The following is an enumeration of some of the popular managed NAND form factors. Please note that the list covers flash technologies used for resident storage and does not cover removable storage like USB flash, SD, etc.

•    eMMC
•    eSD
•    CompactFlash
•    Solid State Drives
•    BA NAND
•    Adaptable NAND
•    Specialized
–    Specially designed controller + raw flash

CompactFlash is included here because it is used both as resident and removable storage. CF comes with a Fixed-drive option which allows it to be used a resident managed NAND.

The above technologies differ from each other on several attributes

•    Form factor – managed NAND can come is several form factors. An SSD may sport a standard 2.5” drive enclosure whereas a CF card will take a 1.0” card form factor.
•    Plug-in interface: What interface does the managed NAND use to connect to the device platform
–    MMC
–    SD
–    ATA
–    Custom
•    Cost: Cost depends on several elements
–    Type of flash used: SLC is much more expensive than MLC
–    Type of controller used: consumer grade controllers (used for consumer grade CF for example) are much cheaper than specialized industrial grade controllers
•    Performance
–    Performance varies depending on the flash type, the controller attributes and the interface.

Some of the big players in the managed NAND business are

•    eMMC
–    Micron, Numonyx
•    eSD
–    SanDisk, Toshiba
•    BA NAND
–    Toshiba
•    Solid State Drives, CompactFlash
–    Too many players in these markets

This was a brief view of the managed NAND landscape. If there is interest, we will do a follow up going in details about the specific categories and interfaces

Once again confirming our suspicion that he doesn’t sleep, Datalight Director of Engineering Ken Whitaker has just published his second book on managing the software development process, this time with a focus on incorporating the ‘agile’ project management techniques that have been so successful at Datalight and elsewhere in Ken’s long distinguished career. At over 400 pages, we haven’t read it yet, but it’s bound to become a favorite of technology management gurus around the world. Here’s some praise from one of Ken’s colleagues:

“Whitaker explains how to run development as the critical business function that it is. Get Principals of Software Development Leadership if you want to lead a technical team to success – or buy a copy for your boss if you want to work on a successful team.” – Steve Johnson, VP, Pragmatic Marketing