Spare Area and Redundant NAND

Intel's controller is a 10-channel architecture and thus drive capacities are still a little wonky compared to the competition. Thanks to 25nm NAND we now have some larger capacities to talk about: 300GB and 600GB.

Intel sent a 300GB version of the 320 for us to take a look at. Internally the drive has 20 physical NAND devices. Each NAND device is 16GB in size and features two 64Gbit 25nm 2-bit MLC NAND die. That works out to be 320GB of NAND for a drive whose rated capacity is 300GB. In Windows you'll see ~279GB of free space, which leaves 12.8% of the total NAND capacity as spare area.

Around half of that spare area is used to keep write amplification low and for wear leveling, both typical uses of spare area. The other half is for surplus NAND arrays, a RAID-like redundancy that Intel is introducing with the SSD 320.

As SandForce realized in the development of its controller, smaller geometry NAND is more prone to failure. We've seen this with the hefty reduction in rated program/erase cycles since the introduction of 50nm NAND. As a result, wear leveling algorithms are very important. With higher densities however comes the risk of huge amounts of data loss should there be a failure in a single NAND die. SandForce combats the problem by striping parity data across all of the NAND in the SSD array, allowing the recovery of up to a full NAND die should a failure take place. Intel's surplus NAND arrays work in a similar manner.

Instead of striping parity data across all NAND devices in the drive, Intel creates a RAID-4 style system. Parity bits for each write are generated and stored in the remaining half of the spare area in the SSD 320's NAND array. There's more than a full NAND die (~20GB on the 300GB drive) worth of parity data on the 320 so it can actually deal with a failure of more than a single 64Gbit (8GB) die.

Sequential Write Cap Gone, but no 6Gbps

The one thing that plagued Intel's X25-M was its limited sequential write performance. While we could make an exception for the G1, near the end of the G2's reign as most-recommended-drive the 100MB/s max sequential write speed started being a burden(especially as competing drives caught up and surpassed its random performance). The 320 fixes that by increasing rated sequential write speed to as high as 220MB/s.

You may remember that with the move to 25nm Intel also increased page size from 4KB to 8KB. On the 320, Intel gives credit to the 8KB page size as a big part of what helped it overcome its sequential write speed limitations. With twice as much data coming in per page read it's possible to have a fully page based mapping system and still increase sequential throughput.

Given that the controller hasn't changed since 2009, the 320 doesn't support 6Gbps SATA. We'll see this limitation manifest itself as a significantly reduced sequential read/write speed in the benchmark section later.

AES-128 Encryption

SandForce introduced full disk encryption starting in 2010 with its SF-1200/SF-1500 controllers. On SandForce drives all data written to NAND is stored in an encrypted form. This encryption only protects you if someone manages to desolder the NAND from your SSD and probes it directly. If you want your drive to remain for your eyes only you'll need to set an ATA password, which on PCs is forced by setting a BIOS password. Do this on a SandForce drive and try to move it to another machine and you'll be faced with an unreadable drive. Your data is already encrypted at line speed and it's only accessible via the ATA password you set.

Intel's SSD 320 enables a similar encryption engine. By default all writes the controller commits to NAND are encrypted using AES-128. The encryption process happens in realtime and doesn't pose a bottleneck to the SSD's performance.

The 320 ships with a 128-bit AES key from the factory, however a new key is randomly generated every time you secure erase the drive. To further secure the drive the BIOS/ATA password method I described above works as well.

A side effect of having all data encrypted on the NAND is that secure erases happen much quicker. You can secure erase a SF drive in under 3 seconds as the controller just throws away the encryption key and generates a new one. Intel's SSD 320 takes a bit longer but it's still very quick at roughly 30 seconds to complete a secure erase on a 300GB drive. Intel is likely also just deleting the encryption key and generating a new one. Without the encryption key, the data stored in the NAND array is meaningless.

The Test

CPU

Intel Core i7 965 running at 3.2GHz (Turbo & EIST Disabled)

Intel Core i7 2600K running at 3.4GHz (Turbo & EIST Disabled) - for AT SB 2011, AS SSD & ATTO

Motherboard:

Intel DX58SO (Intel X58)

Intel H67 Motherboard

Chipset:

Intel X58 + Marvell SATA 6Gbps PCIe

Intel H67
Chipset Drivers:

Intel 9.1.1.1015 + Intel IMSM 8.9

Intel 9.1.1.1015 + Intel RST 10.2

Memory: Qimonda DDR3-1333 4 x 1GB (7-7-7-20)
Video Card: eVGA GeForce GTX 285
Video Drivers: NVIDIA ForceWare 190.38 64-bit
Desktop Resolution: 1920 x 1200
OS: Windows 7 x64

 

The Same Controller Random & Sequential Performance
Comments Locked

194 Comments

View All Comments

  • bji - Tuesday, March 29, 2011 - link

    I think that garbage collection refers to a process of deferring block erases to a later time to be done when the drive is otherwise idle. I.e., if you need to rewrite a block you don't re-write it in place, you write it to a block from the spare area that is already cleared thus saving yourself the time of having to erase the old block before rewriting it. You still mark the old block as needing to be cleared and put into the spare area (to replace the block that was taken out of the spare area during this process), and you do that later during 'garbage collection'.

    There may also be some aspects to which individual blocks from an erase region (my understanding of the terminology is a bit off but I am pretty sure that flash memory can write to smaller regions than it can erase) are moved around during 'garbage collection' to consolidate them into single blocks; this takes blocks that are interspersed with dead area and collapses them down to a smaller fully populated region, then takes all of the now-free blocks and then erases them and puts them in the spare area.

    Having TRIM makes both of these processes more efficient because it tells the drive that it can just mark blocks as ready-for-erase-and-put-into-the-spare-area immediately rather than having to be tracked and managed, and also increases the overall spare area available which means that more already-erased blocks are ready to be used for writes. Having to erase a block before writing it is the performance killer of SSDs and TRIM, along with intelligent algorithms listed above, in addition to things I haven't even thought of most likely, are what allow SSDs to get around the erase block performance penalty and to have such killer performance.
  • randomlinh - Monday, March 28, 2011 - link

    I was excited to see this back when it was "announced." I was hoping we'd be closer to $1/GB for the mainstream performance around now, but looks like I've still got to wait.

    Hoping the 2011 round of controllers push intel to compete with pricing. I'm happy with the performance honestly, but need pricing.

    Or maybe this will drive the x25-Ms down in price and I'll just RAID-0 a pair of 80GB's...
  • Ushio01 - Monday, March 28, 2011 - link

    So all we get from Marvel and Intel are there old controllers working as they should of from the beginning with only Sandforce actually innovating, pathetic.
  • darckhart - Monday, March 28, 2011 - link

    "should have been" maybe. but we all know that's not how business works. sell it, revise it, sell it, revise it, ad nauseum. in any case, know this: they are getting comparable SF-12xx performance WITHOUT realtime compression and dedup which is mighty impressive in my book.
  • darckhart - Monday, March 28, 2011 - link

    oh i forgot to mention they're doing this on 25nm.
  • Vlad T. - Monday, March 28, 2011 - link

    Planned obsolescence or built-in obsolescence.

    http://en.wikipedia.org/wiki/Planned_obsolescence

    That is even more obvious considering how Intel trashes G2 with the same controller.
  • thudo - Monday, March 28, 2011 - link

    My gawd how the mighty have fallen? Doesn't remotely hold its own against the mighty Vertex 3 (SATA3). 120Gb Vertex 3 as now showing up in Canada for ~$290 -- a frick'n steal considering my boot drives have always been ~100-150Gb+ (all you need) and the performance increase is so well worth it. Shame Intel.. shame..
  • davepermen - Monday, March 28, 2011 - link

    you know that intel has the 520 ssds, too? those are to fight vertex3.

    not that i would ever consider ocz an ssd worth buying anyways, but lets not discuss that. anand loves them after he hated them. i still can't (as even after anand has forgiven them, they continue the same crap they did before).

    so for a sata3 system, it's 520. for a sata2 system, the 320 is fine, actually nearly perfect.
  • sean.crees - Tuesday, March 29, 2011 - link

    Say what you will of OCZ, at least they listen to their consumers, and attempt to make legitimate changes in their business practices to satisfy their existing customer base.

    Intel has it's advantages, but appeasing it's current customers are not one of them. Ask the numerous amounts of people that jumped on Intel's 1st gen SSD bandwagon to then be shunned from TRIM support forever, which would require nothing but a firmware upgrade. 2x 80gb for $500 each, and no TRIM support. These things have slowed to almost HDD performance.
  • shatteredx - Monday, March 28, 2011 - link

    Performance is fine, but Intel isn't pricing these drives cheaply enough.

    Whatever happened to the prediction that 25nm drives would cost half as much as their 34nm siblings?

Log in

Don't have an account? Sign up now