Subject: Storage | February 14, 2016 - 02:51 PM | Allyn Malventano
Tagged: vnand, ssd, Samsung, nand, micron, Intel, imft, 768Gb, 512GB, 3d nand, 384Gb, 32 Layer, 256GB
You may have seen a wave of Micron 3D NAND news posts these past few days, and while many are repeating the 11-month old news with talks of 10TB/3.5TB on a 2.5"/M.2 form factor SSDs, I'm here to dive into the bigger implications of what the upcoming (and future) generation of Intel / Micron flash will mean for SSD performance and pricing.
Remember that with the way these capacity increases are going, the only way to get a high performance and high capacity SSD on-the-cheap in the future will be to actually get those higher capacity models. With such a large per-die capacity, smaller SSDs (like 128GB / 256GB) will suffer significantly slower write speeds. Taking this upcoming Micron flash as an example, a 128GB SSD will contain only four flash memory dies, and as I wrote about back in 2014, such an SSD would likely see HDD-level sequential write speeds of 160MB/sec. Other SSD manufacturers already recognize this issue and are taking steps to correct it. At Storage Visions 2016, Samsung briefed me on the upcoming SSD 750 Series that will use planar 16nm NAND to produce 120GB and 250GB capacities. The smaller die capacities of these models will enable respectable write performance and will also enable them to discontinue their 120GB 850 EVO as they transition that line to higher capacity 48-layer VNAND. Getting back to this Micron announcement, we have some new info that bears analysis, and that pertains to the now announced page and block size:
256Gb MLC: 16KB Page / 16MB Block / 1024 Pages per Block
384Gb TLC: 16KB Page / 24MB Block / 1536 Pages per Block
To understand what these numbers mean, using the MLC line above, imagine a 16MB CD-RW (Block) that can write 1024 individual 16KB 'sessions' (Page). Each 16KB can be added individually over time, and just like how files on a CD-RW could be modified by writing a new copy in the remaining space, flash can do so by writing a new Page and ignoring the out of date copy. Where the rub comes in is when that CD-RW (Block) is completely full. The process at this point is very similar actually, in that the Block must be completely emptied before the erase command (which wipes the entire Block) is issued. The data has to go somewhere, which typically means writing to empty blocks elsewhere on the SSD (and in worst case scenarios, those too may need clearing before that is possible), and this moving and erasing takes time for the die to accomplish. Just like how wiping a CD-RW took a much longer than writing a single file to it, erasing a Block takes typically 3-4x as much time as it does to program a page.
With that explained, of significance here are the growing page and block sizes in this higher capacity flash. Modern OS file systems have a minimum bulk access size of 4KB, and Windows versions since Vista align their partitions by rounding up to the next 2MB increment from the start of the disk. These changes are what enabled HDDs to transition to Advanced Format, which made data storage more efficient by bringing the increment up from the 512 Byte sector up to 4KB. While most storage devices still use 512B addressing, it is assumed that 4KB should be the minimum random access seen most of the time. Wrapping this all together, the Page size (minimum read or write) is 16KB for this new flash, and that is 4x the accepted 4KB minimum OS transfer size. This means that power users heavy on their page file, or running VMs, or any other random-write-heavy operations being performed over time will have a more amplified effect of wear of this flash. That additional shuffling of data that must take place for each 4KB write translates to lower host random write speeds when compared to lower capacity flash that has smaller Page sizes closer to that 4KB figure.
A rendition of 3D IMFT Floating Gate flash, with inset pulling back some of the tunnel oxide layer to show the location of the floating gate. Pic courtesy Schiltron.
Fortunately for Micron, their choice to carry Floating Gate technology into their 3D flash has netted them some impressive endurance benefits over competing Charge Trap Flash. One such benefit is a claimed 30,000 P/E (Program / Erase) cycle endurance rating. Planar NAND had dropped to the 3,000 range at its lowest shrinks, mainly because there was such a small channel which could only store so few electrons, amplifying the (negative) effects of electron leakage. Even back in the 50nm days, MLC ran at ~10,000 cycle endurance, so 30,000 is no small feat here. The key is that by using that same Floating Gate tech so good at controlling leakage for planar NAND on a new 3D channel that can store way more electrons enables excellent endurance that may actually exceed Samsung's Charge Trap Flash equipped 3D VNAND. This should effectively negate the endurance hit on the larger Page sizes discussed above, but the potential small random write performance hit still stands, with a possible remedy being to crank up the Over-Provisioning of SSDs (AKA throwing flash at the problem). Higher OP means less active pages per block and a reduction in the data shuffling forced by smaller writes.
A 25nm flash memory die. Note the support logic (CMOS) along the upper left edge.
One final thing helping out Micron here is that their Floating Gate design also enables a shift of 75% of the CMOS circuitry to a layer *underneath* the flash storage array. This logic is typically part of what you see 'off to the side' of a flash memory die. Layering CMOS logic in such a way is likely thanks to Intel's partnership and CPU development knowledge. Moving this support circuitry to the bottom layer of the die makes for less area per die dedicated to non-storage, more dies per wafer, and ultimately lower cost per chip/GB.
Samsung's Charge Trap Flash, shown in both planar and 3D VNAND forms.
One final thing before we go. If we know anything about how the Intel / Micron duo function, it is that once they get that freight train rolling, it leads to relatively rapid advances. In this case, the changeover to 3D has taken them a while to perfect, but once production gains steam, we can expect to see some *big* advances. Since Samsung launched their 3D VNAND their gains have been mostly iterative in nature (24, 32, and most recently 48). I'm not yet at liberty to say how the second generation of IMFT 3D NAND will achieve it, but I can say that it appears the next iteration after this 32-layer 256Gb (MLC) /384Gb (TLC) per die will *double* to 512Gb/768Gb (you are free to do the math on what that means for layer count). Remember back in the day where Intel launched new SSDs at a fraction of the cost/GB of the previous generation? That might just be happening again within the next year or two.
Subject: Systems, Storage | February 10, 2016 - 03:34 PM | Jeremy Hellstrom
Tagged: asustor, AS5002T, NAS, htpc, baytrail
Being in the market for a Plex server and running low on patience and spare hardware I have been sniffing around NAS servers, which is why you are now reading about the ASUSTOR AS5002T. Missing Remote just picked this NAS up for review, powered by a dual core Celeron J1800 clocked at 2.4GHz instead of an ARM processor. The reason that matters is the inclusion of Intel HD Graphics onboard for real time encoding when streaming to remote devices. On the other hand it is not the most modern of processors and the AS5002T also showed some peculiarity with drive sizes. The processor is not going to be able to push 4k over some interfaces but HDMI 1.4a, IR control capability and broad support for the usual selection of HTPC programs does make this NAS a good fit for many. Read the full review to get a better idea of the capabilities of the ASUSTOR AS5002T.
"The ASUSTOR AS5002T is the first Intel based network attached storage (NAS) device tested at Missing Remote. So, I was very curious to see how its dual-core 2.4GHz Celeron J1800 would stack up against the strong showing we’ve seen from ARM Cortex-A15 based systems recently."
Here are some more Storage reviews from around the web:
- PNY CS1311 @ The SSD Review
- PNY CS2211 SSD @ TechwareLabs
- Micron M600 512GB SSD Review @ NikKTech
- OCZ Trion 150 240GB and 480GB SSD @ Kitguru
- SanDisk Extreme 900 480GB Portable USB Type-C SSD @ Kitguru
- WD My Passport Ultra 3TB USB 3.0 Portable Hard Drive Review @ NikKTech
- Kingston HyperX Savage 128GB USB 3.1 Gen 1 Flash Drive @ Modders-Inc
- Kingston HyperX Savage 128GB USB 3.1 Gen 1 Flash Drive Review @ NikKTech
Subject: Storage | February 3, 2016 - 03:31 PM | Allyn Malventano
Tagged: Trion 150, toshiba, tlc, ssd, slc, sata, ocz, A15nm
*Note* This piece originally stated 'A15nm', however this was an error in the Trion 150 spec sheet at OCZ. It has been corrected in this article (as well as at the OCZ web site).
2015 was a bit of a rough year for OCZ, as their integration with parent company Toshiba ran into a few performance bumps in the road. First was the Vector 180 launch, which saw some particularly troublesome stalls during writes and TRIM operations. The Trion 100 launch went a bit smoother, but we did note some inconsistencies in caching performance of those TLC/SLC caching SSDs.
OCZ hopes to turn things around by kicking off 2016 with some updates to their product lines. First up is the just announced Trion 150:
Looking at the spec sheets of the Trion 100 and 150, it may be difficult to spot any differences. I’ll save you the trouble and say that only *one digit* changes, but it is an important one. The Trion 150 will use Toshiba 15nm TLC flash (the Trion 100 used A19nm). What is interesting about this is that the Trion 150 carries the same endurance rating as its predecessor. A flash memory die shrink typically comes with a corresponding reduction in endurance, so it is good to see Toshiba squeeze this likely last die shrink to their planar flash for all of the endurance they can. Further backing up that endurance claim, the Trion 150 will carry OCZ’s ShieldPlus warranty, which offers shipping-paid advance-RMA (without receipt) of this product line for three years!
OCZ has Trion 150 samples on the way to us, and we will get a full performance review of them up as soon as we can! Full press blast follows after the break.
NVMe was a great thing to happen to SSDs. The per-IO reduction in latency and CPU overhead was more than welcome, as PCIe SSDs were previously using the antiquated AHCI protocol, which was a carryover from the SATA HDD days. With NVMe came additional required support in Operating Systems and UEFI BIOS implementations. We did some crazy experiments with arrays of these new devices, but we were initially limited by the lack of native hardware-level RAID support to tie multiple PCIe devices together. The launch of the Z170 chipset saw a remedy to this, by including the ability to tie as many as three PCIe SSDs behind a chipset-configured array. The recent C600 server chipset also saw the addition of RSTe capability, expanding this functionality to enterprise devices like the Intel SSD P3608, which was actually a pair of SSDs on a single PCB.
Most Z170 motherboards have come with one or two M.2 slots, meaning that enthusiasts wanting to employ the 3x PCIe RAID made possible by this new chipset would have to get creative with the use of interposer / adapter boards (or use a combination of PCI and U.2 connected Intel SSD 750s). With the Samsung 950 Pro available, as well as the slew of other M.2 SSDs we saw at CES 2016, it’s safe to say that U.2 is going to push back into the enterprise sector, leaving M.2 as the choice for consumer motherboards moving forward. It was therefore only a matter of time before a triple-M.2 motherboard was launched, and that just recently happened - Behold the Gigabyte Z170X-SOC Force!
This new motherboard sits at the high end of Gigabyte’s lineup, with a water-capable VRM cooler and other premium features. We will be passing this board onto Morry for a full review, but this piece will be focusing on one section in particular:
I have to hand it to Gigabyte for this functional and elegant design choice. The space between the required four full length PCIe slots makes it look like it was chosen to fit M.2 SSDs in-between them. I should also note that it would be possible to use three U.2 adapters linked to three U.2 Intel SSD 750s, but native M.2 devices makes for a significantly more compact and consumer friendly package.
With the test system set up, let’s get right into it, shall we?
Subject: Storage | January 29, 2016 - 04:49 PM | Jeremy Hellstrom
Tagged: pny, CS2211, CS1311, tlc, mlc, phison, xlr8
Over at the SSD Review you can check out PNY's newest SSDs, the TLC based CS1311 and the faster MLC based CS2211 which offers ECC RAM and extra data security features as well as a copy of Acronis. Inside the CS2211 which is the drive featured in this review, you will find an 8-channel Phison PS3110-S10-X controller and 15nm Toshiba MLC, the cache is DDR3L-800, 256MB on the 240GB model and 512MB on the 480GB. This replaces PNY's original Silicon Motion powered XLR8 and it improves upon performance as well as offering a 4 year warranty. Check out all the benchmarks right here.
"Just last week we announced PNY's latest SSD products for the new year, the CS1311 and CS2211. It just so happens that today we have some in our hands for review."
Here are some more Storage reviews from around the web:
- Lexar USB 3.0 Portable SSD (256GB) @ SSD Review
- Synology DiskStation DS1515+ Network Attached Storage @ Modders-Inc
- Synology DS416 4-bay NAS @ techPowerUp
Subject: Storage | January 26, 2016 - 04:41 PM | Jeremy Hellstrom
Tagged: kingston, HyperX Savage, 128GB USB drive, usb 3.1
Once USB drives were everywhere, they weren't particularly fast nor large but they were more portable that HDDs and much more durable. With the arrival of SSDs, flash storage moved from slower thumb drives to SATA which has now become the bottleneck for speed as the drives themselves can actually exceed the transfer capabilities of SATA. That leaves the USB drive out in the cold, with prices matching or even exceeding lower end SSDs and a form factor only slightly more portable than an SSD in an enclosure.
Kingston's Digital HyperX Savage 128GB USB drive is $86 and Kitguru saw sequential reads topping 400MB/s and writes around 200MB/s which comes close to the limits of the USB 3 connection it uses. The question is, does the smaller size and admittedly attractive packaging draw you to choose this over a low cost SSD and enclosure?
"Kingston has earned a reputation with its HyperX brand over the last few years. Today, we are taking a look at the HyperX Savage 128GB USB drive, which supports first-generation USB 3.1 technology and promises ‘blazing fast’ read and write speeds. How does it hold up? Let’s find out!"
Here are some more Storage reviews from around the web:
- Basic Solid State Drive Features Explained @ eTeknix
- QNAP TS-451+ @ Legion Hardware
- Synology DiskStation DS216play 2-Bay Multimedia-Optimized NAS @ eTeknix
- Adam Elements iKlips @ Kitguru
Subject: Systems, Storage | January 19, 2016 - 09:44 PM | Scott Michaud
Tagged: M.2 SATA, M.2, LIVA, ECS
Back in November, Sebastian reviewed the ECS LIVA X2. While the device always had an M.2 slot, its storage options were soldered eMMC chips with capacities of their 32GB or 64GB. They were also pretty slow, with 150MB/s reads and 40MB/s writes in his testing. To exceed that, you need to install your own M.2-based SSD, which was a bit of a difficult process.
According to Links International, via FanlessTech, we are now seeing options that include M.2 SSDs without eMMC. In this case, they are using an Intel-based, 120GB drive. Its signal is M.2 SATA though, which is slower than M.2 PCIe, but a device with this performance characteristic will probably not care about that extra bump in performance. You probably couldn't do much high-bandwidth data crunching with the Braswell processor, and just about every other way on or off of the device is limited to less than or equal to a gigabit of bandwidth. You might be able to find a use case, but it's unlikely to affect anyone interested in this PC.
The jump from eMMC, on the other hand, might.
Subject: Storage | January 15, 2016 - 01:53 PM | Jeremy Hellstrom
Tagged: silicon power, Slim S55, 240gb, Phison PS3110-S10, tlc
At 7mm the Silicon Power Slim S55 is perfect for older ultraportables that need a drive upgrade, though they will certainly slip into a 2.5" bay in any system. The drive uses the Phison PS3110-S10, found in a variety of drives which Al compared last summer. The controller is paired with a 128MB cache of Nanya DDR3 and TLC NAND, which lowers the price to an impressive $65 for the 240GB model. It also performs decently, eTeknix saw 556MB/s in ATTO and 530MB/s in CDM; you can check out more tests in their full review here.
"Silicon Power’s Slim series of solid state drives all come with a 7mm thickness, making them perfect for ultrabooks and similar portable computers that require this form factor. Traditional 2.5-inch mechanical drives mostly come with a 9.5mm thickness, ruling them out as an option. The Slim S55 SSD is the little brother in this series, but it doesn’t need to be ashamed of that."
Here are some more Storage reviews from around the web:
- ADATA Premier SP550 240GB @ eTeknix
- WD Blue SSHD (WD10J31X) 1TB @ TechARP
- Synology DiskStation DS416 4-bay High-Performance NAS @ eTeknix
- Thecus W4000+ Windows Server NAS Review @ Madshrimps
- QNAP TurboNAS TS-451+8G NAS Server Review @ NikKTech
- Synology DS216se 2-Bay Entry-Level and Cloud NAS @ eTeknix
- Toshiba 16GB TransMemory U401 USB 2.0 Flash Drive Review @ Madshrimps
Got a high bandwidth video camera that fills a piddly 4TB SSD in too short of a time? How about a 13TB SSD!
Fixstars certainly gets cool points for launching such a high capacity SSD, but there are a few things to consider here. These are not meant to be written in a random fashion and are primarily geared towards media creation (8k RAW video). Filling at saturated SATA bandwidth, these will take about 7 hours to fill, and just as long to empty onto that crazy high end editing machine. But hey, if you can afford 13TB of flash (likely ~$13,000) just to record your video content, then your desktop should be even beefier.
The take home point here is that this is not a consumer device, and it would not work out well even for pro gamers with money to burn. The random write performance is likely poor enough that it could not handle a Steam download over a high end broadband link.
Subject: Storage | January 13, 2016 - 02:38 PM | Allyn Malventano
Tagged: Seagate, helium, hdd, enterprise, 3.5, 10TB
Seagate updated their Enterprise Capacity line of HDDs this morning with a monster of a 10TB unit:
To achieve this capacity, Seagate switched over to a sealed, Helium filled design (similar to what HGST has been doing for a few years now). Since filling the space of a HDD with Helium helps reduce head flutter and platter thickness, Seagate was able to fit seven platters into a standard 3.5" housing. As an additional note, this drive uses the same PMR (Perpendicular Magnetic Recording) as other recent generation units, and not the SMR (Shingled) employed in their recent 8TB Archive HDD. PMR is a good thing here, as it enables random write access without the performance penalty incurred when attempting the same on an SMR drive.
The Helium filling pushes the MTBF up to 2.5 million hours. Unfortunately the release was light on the other details, and we do not have pricing as of yet, but we will certainly be keeping an eye on this one. Seagate states they are 'shipping to select customers', but given that those customers are ordering by the truckload, it may be some time before we see them in the OEM aftermarket channels.