Samsung ships the world's highest capacity SSD, with 15TB of storage

http://www.computerworld.com/article/3040208/data-storage/samsung-ships-the-worlds-highest-capacity-ssd-with-15tb-of-storage.html?nsdr=true

by elorant • 10 years ago

266 140 10 years ago

computerworld.com

NovaS1X• 10 years ago

Impressive. First question that pops into mind is how long the rebuild time would take if one of these failed in a RAID. I can imagine it'll take a while.

ymse• 10 years ago

It takes 8 hours 20 minutes to write 15TB over 500MBps. Why would you want to stick one of these in a RAID though?

..it would be more like Redundant Array of Prohibitively Expensive Drives.

(I'll show myself out)

rgbrenner• 10 years ago

article says 1200MBps.. so more like 3.5 hours.

rbanffy2• 10 years ago

RAPD is a catchy acronym.

jo909• 10 years ago

15.36TB sequential read and write speeds of up to 1200MBps

~4 hours to read/write a whole drive ain't too bad. Of course there might be other bottlenecks in your raid, or you only reserve so much speed for rebuild, but the drives size is not the problem.

astrodust2• 10 years ago

I'm kind of shocked that this drive is so slow. That's an impressive benchmark compared to your run of the mill consumer drives, but the new PCI-e drives Apple has been shipping are already beyond 2000Mbps.

http://blog.macsales.com/30725-owc-tests-speed-of-ssd-in-201...

I would've expected Samsung to be able to best that. They make the chips, after all.

cm21877• 10 years ago

I thought RAID and SSD were a terrible combination. As if both drives do exactly the same writes at the same time, chances are they will fail at the same time. Plus most RAID controllers don't support TRIM.

Has the thinking changed?

zanny• 10 years ago

A Raid 5 could work, since you are not writing the same data. You would need a hardware coprocessor to handle generating the checksums at 4GB/s though.

chipperyman573• 10 years ago

Yeah, the statistical chance of the same thing happening on both drives is the same, but the chance of it happening at the same time is pretty low.

sdesol• 10 years ago

I thought RAID and SSD were a terrible combination

Well there are different RAID settings. My product is an indexing engine (indexes Git repositories) and I've personally found RAID 0 reduces indexing time by about 30% compared to a single drive. I have a machine that has 4 SSDs running RAID 0 and I've found the performance gain after 3 SSDs is negligible.

It seems like 3 SSDs running RAID 0 is the best combination given my very limited sampling size.

gfody• 10 years ago

Software RAID or a host-based controller is the way to go with SSDs. Otherwise the raid controller becomes your bottleneck as whatever onboard ASIC or RAID-on-a-chip hits its throughput limit (and potentially overheats and dies if it's some crappy product designed for HDDs).

TRIM is well supported by modern host-based controllers designed for SSDs.

Sanddancer• 10 years ago

They'll have the same lifespan, but they won't fail at exactly the same time. Also, a lot of RAID now is done in software, and most software raid types support TRIM.

jlgaddis• 10 years ago

I just (like, about the time you were writing your comment) replaced a pair of spinning disks with a pair of 120 GB SSDs in one of my machines here at home. It's not technically RAID but a ZFS mirror instead. The operating system and related files will live on it while my data still lives on some spinning disks (simply because of the amount of data).

I've got pretty much the same setup (albeit with several more spinning disks) in a bunch of servers and have yet to have any problems (* crosses fingers *).

Freaky3• 10 years ago

As if both drives do exactly the same writes at the same time, chances are they will fail at the same time.

You mean due to exhausting their endurance? This is something you monitor, you should have plenty of time to replace the drives before it becomes a concern.

For other failures, how's it going to be any different from normal HD's? There's always the risk that having the same models/batches in the same conditions might lead to a cluster of failures, but RAID's still likely to save you from plenty of other failure modes.

For instance, the recent Google study[1] suggests that uncorrectable errors with SSDs are actually more common than HD's: "More than 20% of flash drives develop uncorrectable errors in a four year period... significantly higher rates .. than hard disk drives".

I've certainly seen enough IO errors from SSDs to know I want to defend against them. Not to mention the silent data corruption I've seen (and been protected from, yay ZFS).

1: http://0b4af6cdc2f0c5998459-c0245c5c937c5dedcca3f1764ecc9b2f...

rsync2• 10 years ago

"I thought RAID and SSD were a terrible combination. As if both drives do exactly the same writes at the same time, chances are they will fail at the same time."

This is a very good thought and you should be thinking it.

However, it is really only applicable to raid mirrors. Raid stripes will not have identical wear lifespans.

All of our mirrored boot devices are either:

- current intel enterprise SSD paired with the previous generation intel enterprise SSD

or:

- current intel enterprise SSD paired with current samsung enterprise SSD

That way if there is a usage related failure (or some weird firmware bug triggered by use pattern) they don't seize up identically.

antongribok• 10 years ago

If you use a fully distributed storage system, like Ceph (or 3Par?), then rebuild times during a drive failure would be significantly lower. Also, any self-respecting storage system (RAID or not) will lower rebuild priority over normal IO.

To give a specific example, let's say you have a 10-node Ceph cluster with 10 of these SSDs in each node filled up to 75% capacity. During a rebuild you throttle down to 50 MBps per SSD. After a single drive failure, the remaining 99 SSDs will work together to redistribute the data in about 40 minutes:

  $ python3 -c "print(round((15.36 * 1024 * 1024 *.75) / (50.0 * 99) / 60.0, 2))"

40.67

The 50 MBps per drive may seem like a low number, but that actually means each node has to move data at roughly 5 Gbps over the cluster network.

* Edited to fix the python line where the multiplication characters converted to italics formatting.

beagle3• 10 years ago

Isn't 50MBps closer to 0.5Gbps than 5Gbps? Where does the other factor of ten come from?

Animats• 10 years ago

It's awe inspiring to see fifteen terabytes of solid state storage in that little box.

imaginenore2• 10 years ago

It's still less dense than 200GB MicroSD cards.

(which are only $79.50 on Amazon)

jschwartzi2• 10 years ago

Does that density include the connectors and buses that would be required to access those cards?

userbinator3• 10 years ago

What's more amazing is that it's still a long way to go compared to biological storage: https://news.ycombinator.com/item?id=4396931

adevine• 10 years ago

But biological storage also doesn't have the random access characteristics and read/write speeds of this storage.

pcunite• 10 years ago

Bye bye mechanical ...

userbinator3• 10 years ago

The 15.36TB PM1633a drive supports one full drive write per day, which means 15.36TB of data can be written every day on a single drive without failure over its five-year warranty.

In other words, assuming good wear leveling, each bit can only be rewritten ~2K times, which is actually quite low for endurance... the capacity is high and they're using that to hide the fact that it's low, but if this was e.g. SLC flash with 100K endurance, you'd be able to write 50x more.

No mention of retention either, which makes me think SSDs today are more for temporary nonvolatile not-quite-RAM storage and not for more permanent applications.

altcognito• 10 years ago

SLC and MLC endurance aren't much different: http://hothardware.com/news/google-data-center-ssd-research-...

To reach a full rewrite a day, that's like 200Mb/s consistently over the span of a 24 hour day. If your loads/usage looks like this, you're probably expecting your drives to fail and are ready to replace them.

reitzensteinm2• 10 years ago

That link says they're not more reliable, which is not at all the same thing.

hyperpallium• 10 years ago

They're not guaranteeing failure in 5 years, but are pretty sure it won't fail. If "pretty sure" is 98%, then that's 100K (or something like it, probability maths is hard).

SixSigma4• 10 years ago

Thet aren't guaranteeing anything. They are saying they will replace it if it fails inside 5 years.

deelowe• 10 years ago

Given how SSDs work, this is actually pretty good reliability. No one is going to be writing the full disk of 15 TB of data every day.

medecau• 10 years ago

This disk looks just right for fatcache usage.

dcip6s• 10 years ago

Anyone care to take a guess how much the 15TB version will cost?

bluedino2• 10 years ago

10,000 USD?

sp332• 10 years ago

I don't see any SAS SSD's going for much less than $1,000/TB. Some of them are twice that! And I'm sure the density will let it command a premium by targeting users who really need more storage in tight spaces. My guess is $50,000.

tim333• 10 years ago

Samsung's 2TB SSD is $860 so maybe something like that.

http://www.amazon.com/Samsung-2-5-Inch-SATA-Internal-MZ-7KE2...

ryao8• 10 years ago

The projected cost when it was announced was $5000.

scurvy5• 10 years ago

Why SAS and not NVMe? You can buy all NVMe enclosures/servers now.

Is this solely aimed at replacing drives in existing enclosures? Not much good there as most of them will be SAS6. They'll still work, but not nearly as fast as SAS12 nor NVMe.

_wmd2• 10 years ago

You're asking for 2 major tech steppings in a single product. We don't even have many mass market NVMe drives yet, and that aside, they're undoubtedly pushing the limits of their existing controller tech just to handle the capacity. Has NVMe seen much enterprise adoption yet? No surprise they chose an interface to match their intended customers' existing product lines

scurvy5• 10 years ago

Has NVMe seen much enterprise adoption yet?

Have you tried to buy one in the past year? If you're not in the Fortune 100, fat effing chance! Here's how it goes, Samsung announces NVMe product, companies beat down their door screaming "take my money!" and Samsung conveniently "cancels" the product for sale. They made it. They just sold their entire production run to a handful of customers. Maybe, just maybe you can get a couple hundred units if you're willing to wait 3-4 months and someone returns some or changes an order after delivery (and you get the returns).

Even second tier (non Intel, non Samsung) suppliers are sold out. About the only thing you can buy now is HGST because no one wanted their stuff in the first place, and they jacked up their prices in response to other vendors' product shortages.

Yes, NVMe is on fire right now. Everyone wants it. I wouldn't put new tech like this into a 3-4 year old system because of a sunk cost fallacy. NVMe is not also exactly "new". They're already on version 1.2 (or 1.3) of the spec. Intel has gone through 2 major NVMe product revisions (with the third out in 3 months).

Also, Samsung isn't exactly breaking 3D NAND ground here. Novachips did it last year, and in a NVMe interface, too.

chime• 10 years ago

We put a couple of Intel 750s in our primary DB server and so many of our issues just went away instantly. Reading at 2GB/s is amazing. On a 10gb network, our network backups now happen at 1+GB/s. Of course we optimize our DB queries as much as possible but sometimes, you just hit a brick wall and can't speed things up because of how the data is structured. Instead of spending 100 developer/dbadmin hours on reorganizing our tables for some query that runs once a week, we put $2500 of drives and solved more problems than I imagined.

The easiest problems are the ones that go away if you throw money at them and NVMe drastically expands the set of such problems. Most small-mid-sized companies have DBs in the 10Gb-1TB range. If you have a single table that's 100GB in size, you can parse through every single row in just under a minute! This means you can actually use an easy to implement O(n) algorithm instead of trying to make O(1) or O(log n) fit your problem. NVMe SSDs are not that advantageous for companies that are built to scale horizontally on AWS. They are amazing when you have a monolithic DB that you can't partition/shard/cluster easily.

ryao8• 10 years ago

NVMe likely cannot scale to thousands of attached drives simultaneously due to limitations of PCI-E while SAS can. Scaling to thousands of attached disks was the motivation for QEMU's virtio-scsi because virtio-blk's 1 block device per PCI-E device meant that virtual machines using virtio-bulk could not scale past 32 drives. nVMe's 1 drive per PCI l-E device model should suffer from the same limitation.

Similarly, SAS supports multipath for HA while PCI-E does not appear to offer that ability. If it does, then I have yet to hear of an example of hardware implementing it.

Lastly, the idea is likely to appeal to those that care about density. For that purpose, it is better to lower costs and requiring a multitude of PCI express slots is not a great way of doing that. Motherboards with more PCI-E slots are expensive due to the layers needed to accommodate the circuits required by them. Assuming JBODs for such things exist (using a chip to share lanes among multiple devices), they would be expensive and also add to cost.

There is also a decent write up on this here:

http://www.datacenterjournal.com/nvme-connected-ssds/

scurvy5• 10 years ago

NVMe doesn't require a PCIe slot. It just requires some lanes. You can use the 2.5" form factor with the same performance. It works really well.

NVMe scales just fine. As for the livelihood of those selling these storage appliances, well check out NetApp stock price lately.

Also,I wouldn't want to run thousands of these SAS drives in today's storage appliances. Even with high performance, the control plane and rebuild/rebalance stuff would be a nightmare. Trying to imagine a 50 node Isilon cluster with these things... not pretty.

hendry• 10 years ago

On Amazon UK, the highest capacity Samsung SSD I can find is 2TB for 620GBP. So I am wondering when this 15TB we actually hit stores and what's the sweet spot price wise between 1 & 15TB.

bluedino2• 10 years ago

It's a SAS device which isn't meant for retail sales and probably only available to select enterprise customers and OEMs

cm21877• 10 years ago

Samsung is expected to release a 4TB consumer SSD this summer. Might be a better bargain that this enterprise SSD.

http://www.anandtech.com/show/9652/samsung-details-3rd-gen-v...

fakename• 10 years ago

and a 2015 macbook pro still ships with 128gb

thecatspaw• 10 years ago

Not sure what you're talking about, mine shipped with 1tb, my work has some with 500gb and 1tb as well.

vlasev• 10 years ago

The maximum on the Apple site is 512GB standard but you can purchase 1TB to make it 500 dollars extra.

philjohn• 10 years ago

If you cheap out and buy the lowest priced one, yes.

antocv4• 10 years ago

Samsung can try to sell me 100TB SSD, and Ill decline thanks.

Their EVO drives suffer from read degradation slowly over long periods of time. See EVO 840, after about a year it was down to 20MiB/s of continuous read speed. An SSD they call that thing. They refused warranty and released a "Speed up flash warez tool now.exe" as help, which all it did was move the data around on the SSD in the background so as to keep up the charade that their SSDs dont suffer from manufacturing/design error.

pilsetnieks2• 10 years ago

This is a completely different category of product than their low end consumer Evo drives, though.

antocv4• 10 years ago

Its the same company.

They marketed it as warranty so and so, speed such and such, and in fact its a piece of shit, and Samsung does not stand by its own product.

Even their newest Samsung EVO 850 is marketed as write speed ~500mb/s, but that is only true for the first 3gb, due to their usage of "TurboWriteCache", which means it has some kind of DRAM or faster real flash disk, and the rest of the disk is below 100mb/s in write speed.

Perhaps, they should not offer warranty on their "low-end consumer" drives, or lie in marketing/product specifications?

So that we as low-end consumers can make informed decisions? I bought SanDisk Extreme Pro instead, they are very fine disks, and did not cost much more than the evo. Recommend SanDisk strongly.

discreditable• 10 years ago

Their Pro series drives have not been affected by any of the EVO issues to my knowledge.

astrodust2• 10 years ago

The Evo Pro drive was the last one standing in a number of SSD burnout tests: http://techreport.com/review/27909/the-ssd-endurance-experim...

pkaye• 10 years ago

What likely happened is that memory NAND degrade it contents with time and a correction algorithm takes place but it can take multiple iterations of reads and this slows the whole thing down. What the controller should be doing is periodically refreshing the data to prevent this from getting overwhelming.

izzydata• 10 years ago

I hope I can get a 1TB SSD for ~$100 in the near future. I'd love to replace all my HDD.

Samsung ships the world's highest capacity SSD, with 15TB of storage

Related Stories