Subject: General Tech, Storage | July 18, 2013 - 04:56 PM | Scott Michaud
Tagged: Raspberry Pi, nvidia, HPC, amazon
Adam DeConinck, high performance computing (HPC) systems engineer for NVIDIA, built a personal computer cluster in his spare time. While not exactly high performance, especially when compared to the systems he maintains for Amazon and his employer, its case is made of Lego and seems to be under a third of a cubic foot in volume.
Image source: NVIDIA Blogs
Raspberry Pi is based on a single-core ARM CPU bundled on an SoC with a 24 GFLOP GPU and 256 or 512 MB of memory. While this misses the cutesy point of the story, I am skeptical of the expected 16W power rating. Five Raspberry Pis, with Ethernet, draw a combined maximum of 17.5W, alone, and even that neglects the draw of the networking switch. My, personal, 8-port unmanaged switch is rated to draw 12W which, when added to 17.5W, is not 16W and thus something is being neglected or averaged. Then again, his device, power is his concern.
Despite constant development and maintenance of interconnected computers, professionally, Adam's will for related hobbies has not been displaced. Even after the initial build, he already plans to graft the Hadoop framework and really reign in the five ARM cores for something useful...
... but, let's be honest, probably not too useful.
Subject: General Tech, Storage | July 18, 2013 - 02:29 AM | Tim Verry
Tagged: nand, micron, flash, 16nm
Micron recently announced that is has begun sampling 16nm NAND flash to select partners. Micron expects to begin full production of the NAND chips using the smaller flash manufacturing process in the fourth quarter of this year (Q4 2013). Drives based on its new 16nm MLC NAND flash are expected to arrive as early as next year. (PC Perspective's own storage expert is currently overseas, but I managed to reach out over email to get some clarification, and his thoughts, on the Micron annuoncement.)
The announcement relates to new NAND flash that is smaller, but not necessarily faster, than the existing 20nm and 25nm flash chips used in current solid state drives. In the end, Micron is still delivering 128Gb (Gigabit) per die, but using a 16nm process. The 16nm flash is a pure shrink of 20nm which is, in turn, a shrink of 25nm flash. In fact, Micron is able to get just under 6 Terabytes of storage out of a single 300mm wafer. These wafers are then broken down into dies in individual flash chips that are used in all manner of solid state storage devices from smartphone embedded storage to desktop SSDs. This 16nm flash still delivers 128Gb --which is 16GB-- per die allowing for a 128GB SSD using as few as eight chips.
A single 16nm NAND flash die with a SSD in the background
Micron expects the 16nm MLC (multi-level cell) flash to be used in consumer SSDs, USB thumb drives, mobile devices, and cloud storage.
The 16nm process will allow Micron to get more storage out of the same sized wafer (300mm) used for current processes, which in theory should mean flash memory that is not only smaller, but (in theory) cheaper.
A single wafer of 16nm NAND flash (just under 6TBs)
As Allyn further notes, the downside to the new 16nm NAND flash is a reduction in the number of supported PE cycles. Micron has not released specific information on this, but the new 16nm MLC flash is expected to have fewer than 1,000 P/E cycles. For comparison, 25nm and 20nm flash has P/E cycles of 3,000 and 1,000 respectively.
In simple terms, P/E (program-erase) cycles relate to the number of times that a specific portion of flash memory can be written to before wearing out. SSD manufacturers were able to work around this with the transition from 25nm to 20nm and still deliver acceptable endurance on consumer drives, and I expect that similar techniques will be used to do the same for 16nm flash. For example, manufactuers could enable compression that is used prior to writing out the data to the physical flash or over-provisioning the actual hardware versus the reported software capacity (ie a drive sold as a 100GB model that actually has 128GB of physical flash).
I don't think it will be a big enough jump that typical consumers wil have to worry too much about this, considering the vast majority of operations will be read operations and not writes. Despite the reduction in P/E cycles, SSDs with 16nm NAND MLC flash will still likely out-last a typical mechanical hard drive.
What do you think about the Micron announcement?
The full press release can be found below:
Subject: Storage | July 18, 2013 - 01:39 AM | Allyn Malventano
Tagged: ssd, Samsung, pricing, EVO, 840 evo
Subject: Storage | July 18, 2013 - 01:12 AM | Allyn Malventano
Tagged: tlc, ssd, slc, sata, Samsung, cache, 840 evo
Samsung's release of the 840 EVO earlier today likely prompted some questions, such as what type of flash does it employ and how does it achieve such high write speeds. Here is the short answer, with many slides in-between, starting off with the main differences between the 840 and the 840 EVO:
So, slightly increased specs to help boost drive performance, and an important tidbit in that the new SSD does in fact keep TLC flash. Now a closer look at the increased write specs:
Ok, the speeds are much quicker, even though the flash is still TLC and even on a smaller process. How does it pull off this trick? Tech that Samsung calls TurboWrite.
A segment of the TLC flash is accessed by the controller as if it were SLC flash. This section of flash can be accessed (especially written) much faster. Writes are initially dumped to this area and that data is later moved over to the TLC area. This happenes as it would in a normal write-back cache - either during idle states or once the cache becomes full, which is what would happen during a sustained maximum speed write operation that is larger than the cache capacity. Here is the net effect with the cache in use and also when the cache becomes full:
For most users, even the smallest cache capacity will be sufficient for the vast majority of typical use. Larger caches appear in larger capacities, further improving performance under periods of large write demand. Here's the full spread of cache sizes per capacity point:
So there you have it, Samsung's new TurboWrite technology in a nutshell. More to follow (along with a performance review coming in the next few days). Stay tuned!
Subject: Storage | July 17, 2013 - 09:06 PM | Allyn Malventano
Tagged: Samsung, ssd, sata
Good morning from Seoul, Korea!
We're covering the 2013 Samsung Global SSD Summit, and the press embargo has just been lifted on a new SSD - the 840 EVO:
The EVO will push 10nm-class (1x nm) flash, promises increased (2x-3x) write speed improvements over the 840, and will be available in capaities as high as 1TB:
Full press blast after the break, and more to follow as the Samsung SSD Summit continues.
Subject: Storage | July 15, 2013 - 04:05 PM | Jeremy Hellstrom
Tagged: sandisk, Extreme II series, ssd, mlc, slc
SanDisk has done something interesting with their new Extreme II SSD series, they have used both SLC and MLC flash in the drive to attempt to give users the best of both worlds. The drive still has a DDR cache sitting between the flash storage and the controller, but there is an nCache between the MLC flash and the DDR comprised of ~1GB of SLC flash. The idea is that the SLC can quickly accumulate a number of small writes into a larger single write block which can then be passed to the MLC flash for storage. Don't think of it as a traditional cache in which entire programs are stored for quick access but more as a write buffer which fills up and then passes its self to the long term storage media once it is full. The Tech Report put this drive through their tests and found it to be a great all around performer, not the fastest nor the best value but very good in almost any usage scenario.
"With MLC main storage and an SLC flash cache, the SanDisk Extreme II is unlike any other SSD we've encountered. We explore the drive's unique design and see whether it can keep up with the fastest SSDs on the market."
Here are some more Storage reviews from around the web:
- Corsair Neutron GTX 240GB SSD RAID 0 Performance @ Legit Reviews
- SanDisk Extreme II @ SSD Review
- OCZ Vector 256GB @ LanOC Reviews
- Silicon Power Velox V55 240GB SSD @ NikKTech
- Western Digital Se 4TB Review @ TechwareLabs
- Seagate Laptop Thin SSHD 500 GB HDD Review @ Hardware Secrets
- Seagate Desktop HDD.15 4TB / Barracuda XT 4TB @ Hardware.info
- Western Digital SE 4TB Hard Drive @ hardCOREware
- Toshiba Nearline MG03ACA400 4TB SATA III HDD @ NikKTech
- Western Digital Sentinel DX4000 16TB RAID5 4-Bay NAS @ eTeknix
- Icy Box IB-WF200HD @ Rbmods
- Sandisk Extreme microSDXC UHS-I 64GB Memory Card Review @ Legit Reviews
- Patriot SuperSonic Mini 16GB USB 3.0 Flash Drive @ NikKTech
- 49 SD and MicroSD cards tested: there's a difference @ Hardware.info
- Mach Xtreme MX-FX 32GB USB 3.0 Flash Drive @ NikKTech
Subject: Storage | July 3, 2013 - 03:04 PM | Jeremy Hellstrom
Tagged: toshiba, THNSNH 256GB, ssd
If you buy a machine with an SSD installed inside of it there is a good chance it is from Toshiba and it might even be the THNSNH 256GB model. As these drives are not sold separately but only inside OEM machines it is not often benchmarked. [H]ard|OCP wants to change that and put this drive and its internally designed controller up against some of their favourite retail drives. Their testing revealed a mixed bag of performance as in some tests it came close to beating out Samsung's 840 series but in other testing ended up at the bottom of the pack. Still, as this drive will end up in many mobile devices it is good to get an idea of the performance you can expect from it.
"Toshiba's massive foundry capabilities allow it to develop some of the leading SSDs for the OEM market. These SSDs come pre-installed in the latest computers with the option for an SSD, and today we look at the Toshiba THNSNH in comparison to current top-flight enthusiast-class SSDs."
Here are some more Storage reviews from around the web:
- OCZ Vertex 450 256GB @ [H]ard|OCP
- OCZ's Vertex 450 solid-state drive reviewed
- Seagate 600 Series SSD 480 GB ST480HM000 @ techPowerUp
- SanDisk Ultra Plus Solid State Drive @ X-bit Reviews
- Strontium Hawk (240GB) @ AnandTech
- SanDisk Extreme II SSD 240GB @ TechSpot
- SanDisk Extreme II SSD Review (240GB) @ SSD Review
- Plextor M5 Pro PX-256M5P 256GB SSD @ NikKTech
- Angelbird SSD2Go 480GB External SSD @ SSD Review
- Monster Digital OverDrive 3.0 1TB External SSD @ SSD Review
- Seagate Backup Plus 1 TB External HDD Review @ Hardware Secrets
- Silicon Power Diamond D20 500GB USB 3.0 Portable Hard Drive @ NikKTech
- QNAP TS-221 and TS-220 @ Legion Hardware
- SilverStone DS322 Compact USB 3.0 RAID Enclosure Review @ Pro-Clockers
- Patriot Memory Tab 16GB USB 3.0 Flash Drive Review @ Madshrimps
- Icy Dock MB882SP and MB982IP Hard Drive Converter Reviews @ Legit Reviews
- Nine internal DVD burners and Blu-ray drives tested @ Hardware.info
- Hard Disk Drive Myths Debunked! @ TechARP
We caught wind of a leaked Intel SSD Roadmap over at VRZone. The slide shows their rough release plans into early 2014:
Starting bottom-up, the old 320 Series (cropped slide bottom) and 330 Series are being phased out in light of the newer 500 series entrants. The 335 Series, driven by a SandForce controller and 20nm flash, may drop in capacity to only an 80GB model in order to drive customers towards the new 530 Series, which will replace both of the SandForce-driven 520 (SATA) and 525 Series (mSATA) offerings. The new 530 Series will be available in 80-480GB and connect via SATA, mSATA, and the newest M.2 SATA interfaces. You can learn more about M.2 by reading the first 6 or so slides from Paul Wassenberg's presentation from Storage Visions 2013. Here's a closer look at an M.2 unit:
From CES 2013, a Micron mSATA SSD (above) and M.2 SATA SSD (below).
With the 530 appearing to become Intel's big mainstream consumer push, they will also introduce a Pro 1500 and 2500 Series. I suspect Intel's own SATA 6Gb/sec controller will be lifted from their SSD DC S3500 and S3700 Series and trickled down into the Pro Series and possibly even into the 530 Series, though that is only speculation on my part.
For the enterprise, Intel will be further juggling their enterprise models around a bit, discontinuing the SSD 710 and possibly even the (25nm) S3700 in favor of the (20nm) S3500 Series, which will also see large gains in available capacity upwards of 800GB and even 1.6TB crammed into a 2.5" SATA unit. Intel's PCIe SSD 910 will eventually be replaced by what appears to be a quad-SSD-RAID variant of the current S3500 and S3700 Series units, dubbed P3500 and P3700, respectively. These models should show a substantial gain over the SSD 910, which did not perform spectacularly when compared to the newer SATA models available.
Subject: General Tech, Storage | June 28, 2013 - 01:24 PM | Jeremy Hellstrom
Tagged: western digital, Arkeia, backup
IRVINE, Calif. – June 27, 2013 – WD®, a Western Digital (NASDAQ: WDC) company and world leader in digital storage solutions, today unveiled the fourth generation of WD Arkeia™ network backup appliances, delivering an all-in-one backup and disaster recovery solution for small- and medium-sized businesses (SMBs).
The new line up consists of four rack-mount appliance models with larger internal disk capacities, faster processors, increased memory, and integrated solid state drives (SSDs) to shorten backup time and accelerate data recovery. The bundled WD Arkeia v10.1 software delivers new support for “seed and feed” technology to support hybrid cloud backups. This allows companies to move backups offsite via network replication rather than shipment of tapes.
“In announcing their fourth generation of purpose-built network backup appliances, WD Arkeia is delivering comprehensive SMB backup solutions that go beyond simple raw capacity increases,” said Liz Conner, senior research analyst, IDC. “WD Arkeia offers easy- to-use data protection that takes the guess work out of purchasing, configuring and managing data backup and recovery, while also embedding features such as deduplication and hybrid cloud backup for small- and medium-sized businesses or remote offices." “WD is committed to providing the growing SMB marketplace with a comprehensive suite of storage solutions,” said Jim Welsh, executive vice president and general manager of WD’s branded business unit. With unique features, these next-generation WD Arkeia network backup appliances offer solutions providers and their customers a simple, smart way to protect their data.
Fourth-Generation Performance and Ease-of-Use
Fourth-generation WD Arkeia backup appliances deliver performance and ease-of-use at an affordable price for SMBs. These new appliances extend the upper range of WD Arkeia appliances and complement available lower-range appliances. WD Arkeia R120s and R220s, both with optional LTO4 tape drives, integrate dual-core Atom and quad-core Xeon processors, respectively. Existing appliances deliver disk capacities from 2TB to 12TB and will also bundle WD Arkeia v10.1. The new fourth-generation appliances offer: o Increased Backup and Recovery Speed: New features include integrated LTO5 tape drives, processor upgrades to a maximum of 2 hex-core Intel® Xeon®, integrated SSDs on select models, and memory up to 96 GB to allow for increased data backup and recovery speeds of both files and disk images. WD Arkeia’s patented Progressive Deduplication™ technology accelerates backups by compressing data at source computers before transfer over local area networks (LANs) or wide area networks (WANs).
- Higher Storage Capacity: Storage capacity doubles from the third generation, with raw capacity now ranging up to 48 TB, configured in RAID-6.
- Improved Ease of Use: Version 10.1 of WD Arkeia software, delivered with the new generation, includes an on-boarding wizard to streamline the appliance setup process.
- Storage Reliability: All new WD Arkeia appliances feature WD enterprise-class WD RE™ hard drives for maximum data integrity.
- Simplified Tape-free, Offsite Storage: Version 10.1 of WD Arkeia software extends support for hybrid cloud backup capabilities to the full line of WD Arkeia appliances. “Seed and feed” capabilities allow administrators to supplement network replication of backup sets offsite by using USB-connected hard drives to transfer initial and large backup sets and also to size WAN bandwidth for the replication of nightly incremental backups.
Pricing and Availability
WD Arkeia fourth generation network backup appliances – models RA4300, RA4300T, RA5300, RA6300 – will be available in July 2013 through select DMR’s and WD-authorized value-added resellers (VARs) in the US, Canada, and Europe. Manufacturer’s Suggested Retail Price, including hardware and software, begins at $9,990 USD. WD Arkeia network backup appliances are covered by one year of unlimited access to technical support, one year of software updates, and a one-year limited hardware warranty.
Introduction and Specifications
Intel has pushed out many SSDs over the years, and unlike many manufacturers, they have never stopped heavily pushing SSD in the enterprise. They did so with their very first push of the X25-M / X25-E, where they seemingly came out of nowhere and just plunked down a pair of very heavy hitting SSDs. What was also interesting was that back then they seemed to blur the lines by calling their consumer offering 'mainstream', and considering it good enough for even some enterprise applications. Even though the die-hard stuff was left to the SLC-based X25-E, that didn't stop some consumers from placing them into their home systems. The X25-E used in this review came from a good friend of mine, who previously had it installed in his home PC.
With several enterprise class models out there, we figured it was high time we put them all alongside each other to see where things are at, and that's the goal of this particular piece. We were motivated to group them together by the recent releases of the DC S3500 and DC S3700 drives, both using Intel's new Intel 8-channel controller.
|X25-E||SSD 320||SSD 710||SSD 910*||DC S3500||DC S3700|
|Capacity||32, 64GB||40, 80, 120, 160, 300, 600GB||100, 200, 300GB||400, 800GB||80, 120, 160, 240, 300, 480, 600, 800GB||100, 200, 400, 800GB|
|Write (4k)||3.3k||23k (8GB span)||2.7k||18.7k||11k||32k|
- Since the SSD 910 is subdivided into 4 or 2 (depending on capacity) physical 200GB volumes, we chose to test just one of those physical units. Scaling can then be compared to other units placed into various RAID configurations. 910 specs were corrected to that of the single physical unit tested.
- All other listed specs are specific to the tested (bold) capacity point.
Starting with the good old X25-E, which pretty much started it all, is Intel's original SATA 3Gb/sec 10-channel controller. Despite minor tweaks, this same controller was used in the X25-M, X25-M G2, SSD 320 and SSD 710 Series. Prior to Intel releasing their own 6Gb/sec SATA controller, they filled some of those voids by introducing Marvell and SandForce controllers with the 510 and 520, respectively, but those two were consumer-oriented drives. For the enterprise, Intel filled this same gap with the 910 Series - a PCIe LSI Falcon SAS RAID controller driving 2 or 4 6Gb/sec SAS Hitachi Ultrastar SSDs. Finally (and most recently), Intel introduced their own SATA 6Gb/sec controller in the form of the DC S3500 and DC S3700. Both are essentially the same 8-channel controller driving 20nm or 25nm IMFT flash, respectively.
More to follow on the next page, where we dive into the guts of each unit.
Get notified when we go live!