It’s true, the RealSSD™ C400 delivers some pretty impressive performance (415 MB/s reads and 260 MB/s writes). But all those IOPs and sequential performance specs are over the head of most consumers. To speak to them, we set up this simple demo of a C400 and stock hard drive running the same programs in identical laptops.
Of course the demo itself isn’t a novel concept by any means, but we think you’ll agree that results are staggering.
Dean Klein, Micron’s Vice President of Memory System Development, discusses the new RealSSD C400 drive and how SSDs will grow in 2011 with the influx of ultra-mobile notebooks. Dean also discusses how the leading-edge performance and capacities of the C400 are poised to serve this market.
Micron has a long history of providing networking customers with outstanding technical expertise, dedicated customer support, and some of the best low-latency memory solutions available. I’m proud that it has led to a technical collaboration with TPACK, now a subsidiary of AppliedMicro, that pairs our low-latency RLDRAM memory with their next-generation TPX5000 100G packet processor to create a highly customizable platform that maximizes networking system performance.
TPACK A/S chose our RLDRAM 2 memory for their new Ardbeg reference design for high-speed 100 Gb/s optical transport systems. It’s a high-performance design meant to help networking equipment meet the ever-increasing demand for bandwidth. TPACK has selected Micron’s RLDRAM 2 as it represents a cost-efficient, high-performing, and well supported memory solution for networking applications.
To match the requirements of the networking segment, Micron is committed to providing customers with supply stability and continuity of support. As a matter of fact, our popular RLDRAM 2 product line was recently migrated to a more advanced 50nm process technology and the Ardbeg platform makes good use of the new design. Moving to 50nm provides enhanced benefits, including improved system performance and lower power consumption. It also allowed us to move production of these parts to our most advanced fabs, which will help us maintain a much longer roadmap for RLDRAM 2 memory, enabling us to provide these parts for years to come.
The TPX5000 processor (the newest in TPACK’s line of TPX packet-transport switches) delivers a full 100 Gb/s line rate and is targeted at transport network applications. It provides extensive support of Ethernet and MPLS-TP protocols with HW-based SONET-style OAM and 50ms protection for thousands of individual packet flows. The addition of Micron’s RLDRAM 2 facilitates lookup in large Ethernet and MPLS-TP address tables at the 100 Gb/s line rate.
While we expect RLDRAM 2 be used in applications like this for years to come, we recently also announced our third-generation RLDRAM 3 memory that pushes performance even further with higher densities and lower latency. To learn more about our newest RLDRAM products, visit our Web site.
It’s a new NAND technology that’s going to enable new applications and continued NAND scaling. The difference is in the way it handles ECC. Get the details about what it is, how it works, and why it’s advantageous in this whiteboard overview.
We recently sat down with Glen Hawk, our VP of NAND Solutions, to talk about the company’s direction and vision for NAND Flash memory. It’s a candid look at the challenges that NAND technology presents and our response in terms of developing NAND solutions that can go the distance and continue to provide advantages and create exciting possibilities well into the future.
PCWorld recently published its list of top technology products of 2010, and coming in at No. 16 out of the 100 best in the world is the RealSSD C300 solid-state drive. In a year that has seen a fair share of breakthrough technologies – Apple’s iPad and iPhone 4, the latest version of the Amazon Kindle, and Google’s Android 2.2 smartphone operating system, just to name a few – earning a spot in the top 20 is a notable achievement. And in the Best of 2010 subcategory of “Storage and Useful Gadgets,” the C300 came in at the very top of the list.
In its description of the product, PCWorld noted that the C300 “is the first SSD we’ve tested to support 6-gigabits-per-second SATA, which enables faster write speeds because it can push more data through the pipe” and that it “delivered top-flight overall performance and the best write performance.”
PCWorld’s editorial staff evaluates hundreds of tech products throughout the year, spanning the full range of offerings from smartphones and PCs to software and storage and beyond. Only the best of the best make it onto the prestigious top 100 list each year, underscoring the truly standout capabilities of the RealSSD C300.
Some of the other products joining the C300 in the top 20 include Samsung’s Galaxy Tab PC (#5) and Epic 4G smartphone (#8), Google’s Chrome Web browser (#14) and Microsoft Office 2010 (#19). Make sure to read the full article for a complete list of PCWorld’s 100 Best Tech Products of 2010.
We just concluded a busy few days at the fall Storage Networking World in Dallas. Not surprisingly, a lot of the discussions centered on doing more with less. Nearly everyone is interested in what SSDs could do for their enterprise, but they’re cautious too, and not sure they know enough about this new technology to choose the best SSD for their application.
So how do you know which SSD will work best for you?
Unfortunately, it’s not such a simple question. All SSDs are not created equal, and you can’t rely solely on the quoted specs to tell them apart. If you do that, you’re setting yourself up for disappointment. Quoted specs are typically based on fresh-out-of-box (FOB) performance—the best performance the drive will ever achieve. Instead, you should look at steady state performance—the level the drive achieves once it fills to the point where write performance varies little over a relatively long period. Often, it’s at a significantly different level than the FOB performance numbers.
Why the difference? As the drive begins to fill, performance is impacted by a SSD phenomenon called write amplification—a multiplying effect that results from having to rewrite user data on the SSD. Inefficient garbage collection and wear leveling algorithms can further increase the write amplification of a SSD. You can compare the steady-state performance differences between our enterprise P300 drive and competitors in the benchmarks we posted back in August.
Steady state performance is an important matter that deserves more attention and education. That’s why one of my colleagues devoted much of his time here at SNW to conduct a hands-on lab that demonstrated how to set up an Iometer test that showed significant differences between the steady-state results of client and enterprise SSDs. We believe that the more people know about the specifics of SSD performance, the better the likelihood that they’ll make an educated choice and be pleased with the results they get.
We’re working on a whitepaper that explains more about the technical details behind SSD performance states. If you’d like to be on our mailing list when it comes out, sign up for a micron.com account (be sure to select an interest in SSDs).
Micron today announced that our 25nm NAND is being used in a new all-in-one storage and media solution from Hitachi-LG Data Storage (HLDS), called Hybrid Drive. The innovations Micron has made in NAND flash memory continue to spark new and compelling end-product designs from our customers – the Hybrid Drive being one of them. We wanted to learn a bit more about the Hybrid Drive, so we asked a few questions to HLDS’ Jack Lee, Senior Manager, and here’s what he had to say:
Analyst firm iSuppli recently issued a report forecasting a major surge in shipments of NAND flash memory, driven by the explosion in the tablet device market. Led by Apple’s introduction of the iPad™ earlier this year (and its immediate success with consumers), we’re now seeing a flood of competitive tablet offerings from OEMs spanning the industry–from the traditional PC manufacturers to smartphone OEMs and other mobile gadget makers.
It’s impossible to predict which tablet versions will catch on with consumers and which ones will quietly fade out of circulation, but we do know that NAND flash will be in high demand for this product segment.
According to the iSuppli report, shipments of NAND for tablet devices are expected to reach 1.7 billion (yes billion!) GB in 2011, which is nearly a 300 percent increase over this year’s expected 428 million GB shipped. And this rising tide isn’t expected to pull back anytime soon—the report forecasts 8.8 billion GB in NAND shipments by 2014.
NAND Flash represents the most advanced semiconductor technology in the world, and is approaching the atomic level where storage cells are separated by a countable number of electrons. Taken at face value, one might assume that these advances are creating a palpable opportunity for designers to capitalize on increased market demands for higher density applications in consumer electronic devices, computing platforms, and industrial systems. However, there are a few hurdles that we need to overcome before customers can seize this burgeoning opportunity.
In traditional system architectures using NAND, the host controller assumes responsibility for all NAND management functions, including block management, wear leveling, and error correction code (ECC). In these instances, the host controller has to take into account the NAND ECC requirements for potentially multiple generations (process nodes) and multiple cell technologies (1/2/3 bits per cell), as well as the requirements from multiple suppliers. Additionally, flash error management is rapidly moving beyond what today’s ECC schemes are capable of and in the near-future will require more advanced ECC and signal processing algorithms. On the other hand, the other NAND management functions are almost solely dependent on the application, something the host controller and system manufacturers are best suited to develop. The solution? Let the NAND manufacturers worry about the flash error management.
As a founding member of the ONFI Working Group, Micron has teaming up with other NAND Flash companies to create standards to ensure our customers can take advantage of the great strides we’re making in NAND flash without having to worry about the error management requirements. One of these standards is the EZ-NAND protocol, or ECC Zero NAND. In essence, EZ-NAND alleviates the burden of the host controller to implement and keep pace with the fast changing ECC requirements that is circumstance of NAND flash process shrink. Here’s a side-by-side comparison of a traditional NAND architecture compared to the EZ-NAND architecture:
| Traditional Architecture | EZ-NAND Architecture |
| Scaling complexity due to ECC and controller sizing issues | Abstracted storage for better error management and more scalable capacity |
| Increased requirements, both signal processing and ECC, cause controller implementations to be more tied to NAND Flash | Minimized controller cost by combining with high bus speeds and increasing capacity and performance per pin on the controller |
| Signal management is no longer contained at the interface boundary | Less burden on host controller to keep up with fast changing ECC requirements |
For more, check out this presentation from Pete Feeley of Micron that was given at the 2010 Flash Memory Summit. Or, drop me a question in the comments section and I’ll be in touch.
If you enjoyed the post, make sure you subscribe to the RSS feed.
© 2012
Powered by WordPress. Design by 