Cherryville - SSD 520

[Computurd]

the samsung is going to change things just in time...OEM drives will make things better for everyone. Just have to let that one develop. But the speeds it gets with no compression/decompression or smoke and mirrors is nice. With M4 up there as well there are some real solid SSDs out there right now pushing big numbers, without compression, that is extremely important for competition, especially in the future.

As to the intel....no sense in making any 'statements' about it. couldnt say either way, confirm or deny, so it doesnt matter.
Sounds silly actually. someone saying "i know something but cant say" because then no matter what when it comes out, they will just say, "See, i was right!" with no way to know if they were in fact right or privy to insider knowledge. so i guess i will shut up.
i do think it would be safe to say that if one door opens for one, a door closes for another. kinda how things work lol.

[squick3n]

I'm very curious, b/c judging from Ao1's trace analysis, the speeds of these drives is mostly unused. I'm testing out my Z68's caching function just to see how it goes, and I wonder if that sort of thing is the future. Controller and firmware working to make frequently used random files "behave" like large sequentials.

[bluestang]

Well, the 4th of November is here....

[Ao1]

Where did the 4th come from? I heard around the middle of the month when the X79 comes out. Intel have already posted some details of the DX79SI on their web site.


It’s only going to be a stop gap until the next gen Intel controller comes out anyway

[Anvil]

The 4th was a date that I believe was mentioned at some site that frequently reports about Intel SSD's , it has later been said to be released more or less toghether with the X79, which is said to be released the 14th
(we'll just have to wait)

[Ao1]

Looks like it’s taken Intel a little longer than anticipated to iron out the bugs in the SF controller. Cherryville ETA Q1 2012.
20nm nand coming up in Q3 2012.



Uploaded with ImageShack.us

http://www.xbitlabs.com/news/storage...ce_Points.html

[Anvil]

They are late but it's not the end of the world, I wonder what changes they are doing vs other SF2 series drives.
(I expect they are able to "tune" the drives to their liking)

[some_one]

This quote is interesting although how much substance there is to it is unknown by me.

also interesting that Intel is choosing to slow the 2281 controllers max throughput on their version(520 series) of these drives as well.

[johnw]

This quote is interesting although how much substance there is to it is unknown by me.

Completely worthless without substantiation.

[Anvil]

From this PDF
(can be found by searching for one of the part#)

There is nothing that points to Intel will be slowing down the drives, they may have chosen other tools for measuring iops/bandwidth.

Intel® SSd 520 SerIeS
Performance SSD: High-performance storage for 2nd gen Intel® Core™ processorbased notebook and desktop PCs with a 6.0 Gb/s SATA interface.
SATA Interface SATA 6.0 Gb/s
Form Factor 2.5” drive
480 GB / 240 GB / 180 GB / 120 GB / 60 GB
IOPS Reads: up to 40K IOPs 3,4
maximum 85K IOPs 3,4
Writes: up to 70K IOPs 3,4
Sequential Bandwidth (Sustained) Reads: up to 550 MB/s 4
Writes: up to 520 MB/s 4

5 years minimum useful life with up to 20 GB/day written 3

1 IOPs and endurance specifications are for server-based workloads.
2 With 20 percent over provisioning.
3 IOPs and endurance specifications are for client-based workloads.
4 Performance values vary by capacity.
5 Designed to work with any Intel® Solid-State Drive as the target drive for cloning. Other brands of SSDs are not supported as target
drives. Extra cables or other hardware items are required for complete data migration solution.
6 Desktop installation kit contents: SATA power and signal cables, 3.5-inch desktop drive bay to 2.5-inch SSD adapter bracket, screws,
installation guide, and warranty documentation.
All products and dates specified are preliminary based on current expectations and are subject to change without notice. Availability
in different channels may vary.
Intel, the Intel logo, and Intel Core are trademarks of Intel Corporation in the U.S. and other countries.
* Other names and brands may be claimed as the property of others.
Intel®

[bluestang]

Is the thought still that it is a SF controller in the 520?

I was researching some info about bad blocks on SSD and ran across this (source - almost at the bottome of page):

"Intel invests in SSD controller company

Editor:- November 16, 2010 - Anobit today announced that it closed a new funding round of $32 million led by Intel Capital - bringing Anobit's total funding to over $70 million.

Editor's comments:- unlike another well known SSD controller company - Anobit is a company which until today has been known by a very small number of analysts - and competitors and maybe even some SSD companies who might be using their products. I'm sure that - as a result of today's announcement - they will get better known too."

Also found on Intel's site here.

Just grasping at staws at this point while I await the realease of the 520.

[Ao1]

Interesting. Nice find.

According to the specs below they match those quoted for the 520. All the indications are that Intel will use SF for the 520, however the specs Intel quote for the 520 do not match up with a SF controller. That said it’s hard to know how Intel would have established performance if compression is taken into account.

“Separately, Anobit announced more Genesis SSDs for network and direct attached storage. The new Genesis SSDs perform at 70,000 IOPS on random read using 4KB blocks and 40,000 IOPS for writes using 4KB blocks. They have an endurance of 50,000 program erase cycles”.

http://www.networkworld.com/newslett...1911stor1.html

[Anvil]

I was just about to post a link on the Genesis SSD, could very well be possible

Link

MSP vs 48bit ECC
MSP=memory signal processing

[Computurd]

Drama!!! very interesting, i must admit my interest is piqued!

[flamenko]

Nice find...could I be wrong?!?!?!?!?

Credit where credit is do and I hope you guys dont get offended at my linking the site and bluestang...

http://thessdreview.com/latest-buzz/...rce-based-ssd/

If you check out the article...you just might find an interesting pik with interesting firmware listed though!!!

[Christopher]

There's more drama than Boardwalk Empire.

At least it won't be JASFD (Just another SF drive) in any even. Most pleasing.

[bluestang]

@flamenko
Wow, thanks!

Edit: In the pic, the FW looks like 31Hi to me.

[Computurd]

MSP v ECC. very interesting....

http://www.anobit.com/uploaded/IPA12112-Web.pdf

http://www.anobit.com/uploaded/FMS_2010_MSP_vs_ECC.pdf

[flamenko]

A little background today and I learned that Anobit, for the most part, has been accustomed to multiple PCBs in their form factor SSDs and their speeds quite probably are not achieved through any single controller which would make it impractical and not financially viable as a consumer SSD.

Could be wrong though...

[Computurd]

i think SF + MSP = 520

[bluestang]

Would MSP + any other controller out there = 520 specs? Marvell (M4), Samsung (830), Intel (a new one?), etc. ???

[Computurd]

That im not sure of. What is interesting is the concept of MSP as a whole, with it working alongside ECC in many cases...just seems to do so much to relieve the error rates. I guess saying using ECC v MSP is a bit of a misnomer, as from their documentation they seem to work in tandem.

[Christopher]

I know that Anobit aquired some of their [DSP related] IP from a company called Cosmic Circuits earlier this year. From the reading I did on the subject, it seems like something you'd find on my Asus Xonar Essence STX.

[Christopher]

It looks like Apple may be buying Anobit.

http://techcrunch.com/2011/12/13/app...n-500-million/

But I see mention of Apple already using their stuff in the MacBook Air, and I don't think that's true. The Macbook airs use Toshiba and Samsung controllers.

[Computurd]

funny thing, everyone was screaming that Intel was buying SF and they were in talks, but that was actually LSI. i think that you may see a bit of a similar situation here. Not Apple, but someone else? who knows, thats purely speculative, i have notta freaking clue.
I do know this, why in the world would Intel sink all that money into them...then let apple scoop em up?