Notebook SSDs are dead

by Robin Harris on Wednesday, 2 July, 2008

It’s all over but the shouting
The scoop: the gap between notebook SSD promise and performance has been growing steadily. Now a review in Tom’s Hardware puts the final nail in the coffin. The title says it all:

The SSD Power Consumption Hoax : Flash SSDs Don’t Improve Your Notebook Battery Runtime – they Reduce It

By as much as an hour. A winner with the stupid high-end notebook demographic. The Paris Hilton market.

Ouch. Oops. Who knew?

Or who should have known?

Details
There’s a longer piece with some detail at Storage Bits but here’s the summary:

  • A Crucial SSD – costing $25/GB – used more power – 1.6 W at idle – than any 2.5″ notebook drive requires.
  • A Memoright 32 GB drive used a full 2 W at idle
  • An Mtron 32 GB flash drive reduced battery life by almost an hour.
  • The slowest drive – a year old Sandisk SSD 5000 – almost equaled the Hitachi 7200 RPM Travelstar’s energy use. But the SSD offers fewer IOPS than the hard drive!
  • They tested against a 200 GB Hitachi Travelstar 7k200, but other 2.5″ 7200 RPM drives have similar power envelopes.

And, of course, a 5400 RPM drive is more efficient. And a 160 GB 1.8″ drive is even more efficient, roomier and cheaper than any of the SSDs TH tested.

My guess on the not-easily-or-quickly-fixed culprit? The flash control logic – disk translation layer – needs cycles for wear leveling, garbage collection, buffer and cache management, flash mux/demux and the SATA interface – with frequent background operations even when the drive is idle.

And don’t forget the 20 volts required to write a cell.

Tom’s singles out Crucial for special mention:

Users who purchase this drive because of Crucial’s statements such as “low power consumption” and the product being ideal for “users who want longer battery life” will most likely be disappointed. While the total battery runtime certainly depends on the workload — we used Mobilemark 07 — the minimum and maximum power consumption measurements prove that Crucial’s statements of low power consumption are in fact wrong: 1.6 W idle power is more than any 2.5” notebook hard drive requires.

Did anyone even think to check the facts? At least one engineer had to know – and he told someone.

What’s the dynamic?
Some will say I’m premature, like when I said HD DVD was dead a year ago. But think about the market dynamic:

  • Cool but costly new technology needs early adopters
  • Based on the marketing, hip high-end adopter spring for costly status symbol with claimed road-warrior features
  • But the supposed advantages don’t exist, so the early adopters feel like chumps
  • Word of mouth stops. Who wants to admit they were suckered?
  • Notebook SSDs slip into obscurity as enterprise and very low-end SSDs move into the spotlight

Making early investors/adopters look stupid is not a winning strategy.

The StorageMojo take
The notebook SSD vendors have dug themselves a very deep hole. How to fix?

  1. Stop digging. A month in detox would help. Some encounter group time with the HD DVD folks.
  2. Form a serious performance consortium and get real about performance, power and longevity.
  3. Do the hard work of getting notebook operating systems better optimized for flash. Use Linux and OS X to beat Microsoft into some semblance of cooperation. Do the engineering for Apple – they’re open source, right? If Apple does it, it’s cool – and you need cool.

What the SSD guys will do:

  • Deny and obfuscate. “Not representative. Slanted. Unfair. Conspiracy.”
  • Claim next gen will fix all problems.
  • Performance, performance, performance. Which is a weak reed as well.
  • Point to cost curves show that, without a doubt, flash overtakes disk in 5 years.

And then hope the smart, techy, affluent road warrior demographic has a short memory. Good luck with that.

Comments welcome, of course.

{ 24 comments… read them below or add one }

SSD engineer July 3, 2008 at 12:26 am

Please take a look at the comments on the Tom’s hardware guide report before you rush to build your basic logic foundation on that.

the storage anarchist July 3, 2008 at 3:46 am

Robin – couldn’t agree more: laptop flash drives have truly been over-hyped in the “green” space. Their only hope is that they figure a way to deliver sufficient performance to make it in the server or array storage markets – which I doubt will happen.

Idle power is an interesting consideration. I”m not sure the culprit is exactly as you describe, but I’ll agree those things must at least be contributors.

Fortunately things are a little different in the enterprise flash drive space. At least, the drives typically do their housekeeping in parallel to normal I/O streams, since there really is no such thing as “idle time” for the overwhelming number of applications that are being deployed on the technology…and the “under load” power requirements are nominally lower than that of spinning rust (32% lower than the equivalent-sized 15K rpm disk for the STEC drives EMC is using, and that’s an all-in comparison including power conversion and A/C load). Still, in the near term you still won’t pay for and EFD on power savings alone :)

Jeff Darcy July 3, 2008 at 4:32 am

You misread the article, Robin. If you look at the power-consumption results on page 14, you’ll see that the Hitachi drive drew more power *at idle* than the Sandisk SSD did *under load* – and that doesn’t even count the difference in cooling needs. The MemoRight SSD also used less power at idle than the Hitachi did, and idle is where most notebook drives are most of the time. Those results are starkly at odds with the traffic-generating headline, and until the inconsistency is resolved I wouldn’t jump to any conclusions. What problems do exist with SSD power consumption are also more easily solved than you let on. Some functions can be moved back to the host, others to dedicated silicon which can do them very efficiently. It’s not like hard drives don’t have processors in them drawing power too, y’know. When somebody does a head to head comparison where the drives are idle 90% of the time and only reading 90% of the remainder, and properly accounts for the fact that whole-system I/O performance might not scale perfectly with drive performance, then it’ll be worth paying attention to.

DaveG July 3, 2008 at 6:19 am

Robin misread the article, Jeff? If anything he was more reserved in his judgement than Tom’s:

“results prove unmistakably that battery runtimes do not increase when using flash based SSDs… Battery runtime may improve if systems run idle, but most of the time they don’t, and this is where the good old hard drive shines brighter than before.”

And he clearly highlights the Crucial SSD’s idle power draw comparison, not the SanDisk (ref p.14). The point here is that flash is no panacea. Solid-state offers potential benefits, but shoving square peg technology into round hole problems as the industry is attempting with notebook SSD will only serve to delay and confuse the adoption of beneficial applications of flash storage.

Jeff Darcy July 3, 2008 at 8:27 am

I think Tom’s is flat-out wrong about notebooks running idle, but at least they added the caveat. Some couldn’t even bring themselves to do that.

I’ve composed a more detailed response at http://pl.atyp.us/wordpress/?p=1359 so I’ll forego further comment here. I was going to post-date it until tomorrow (because I like to space out my posts a bit) but, since everyone has been so polite and all, I decided to push it out sooner.

Steve Shockley July 3, 2008 at 8:28 am

They may still have a niche with ultra-quiet PCs.

Colin July 3, 2008 at 10:34 am

The article had a systemic flaw in it’s analysis… That is that because the SSDs had higher performance, the benchmarks were likely doing more work with them than they were with the HDD. So, the CPU was working harder, the graphics card was working harder, etc. etc. Tom’s should really redo the experiments and publish the disk stats and compare bytes read/written before battery death rather than simply battery runtime. It’s a glaring flaw that was immediately apparent and needs to be addressed before any conclusions can be drawn.

Robin Harris July 3, 2008 at 10:47 am

This is a corner case where the divide between marketing and engineering gets thrown into sharp relief. Yes, there are issues with Tom’s methodology. There are *always* issues with testing.

But the preponderance of ALL the testing is that notebook SSDs have either no or small power advantages over hard drives for most consumers. There have been a couple of tests that do show some advantage: Anandtech’s test that played music continuously – when most road warriors would use a player or a cell phone.

And there are good reasons for this: hard drives are pretty efficient and getting more so; and current displays, CPUs and graphics use a lot more power than the disks. There’s only so much juice you can squeeze out of this lemon.

In a consumer driven world the fine points get lost. Until a major notebook vendor can advertise something like “Get an extra 60 minutes battery life with our high-performance SSD” the notebook SSD market is toast.

The standard engineering take is: give us more time and money and we will make it better. And I’m sure they can and will make it better. The question is: better enough to matter?

Here’s the marketing take. When you charge more for less capacity and the only benefit is performance – and even that seems limited to startup and some file operations – you do not have a value proposition that consumers will buy.

And BUYING, my friends, is the name of the game.

Flash will still play an important role in data storage – just not in the notebook market.

Robin Harris July 3, 2008 at 2:06 pm

From Micron Technology – why they can’t comment like everyone else . . . .

To be attributed to Dean Klein, vice president of memory system development for Micron:

“The controllers analyzed in the Tom’s Hardware review are early-generation, multi-chip and in some cases even use FPGA’s, which can be quite power hungry. As with many other first and second generation drives, these drives are not delivering on the full potential of the NAND and are not delivering properly on the performance promise.

There is another factor to be aware of. If the CPU spends 25 million clock cycles waiting for random HDD data, but only part of that waiting for SSD data, the actual increase in notebook power consumption may be in the CPU. A useful metric is how much processing gets done per watt. If you are willing to scale back performance to that of an HDD-based system, an SSD-based system should deliver significantly longer battery life.

Finally, consider that many of today’s applications and operating systems are not optimized for SSDs, but for rotating media. As an example, Vista has a background defrag utility that is not needed, and in fact is not desired for SSDs.”

Wes Felter July 7, 2008 at 11:51 am
Gayle July 7, 2008 at 12:37 pm

Seems odd that the company that was first to the market with MLC SSD’s and recently knocked Samsung out of a lucrative Apple contract should have been tested. I think Apple probably ran as many tests as TOM”S. STEC SSD’s are superior and it directly relates to their controller technology and testing procedures .

Joe Kraska July 7, 2008 at 8:39 pm

I think it might behoove the various readers of Robin’s article to go over to the Tom’s Hardware original story and read the comments in detail. It seems to me that we’re not readily able to discern much here quite yet.

Joe

Fazal Majid July 9, 2008 at 7:19 am

I have a MacBook Air with the SSD option. It boots so fast I am considering getting a SSD as the boot drive for my next desktop.

antonio July 14, 2008 at 3:32 pm

hi, I find strange you say ssd is dead. Startup times when loading programs or documents can be very good => they can make you win time and the money you wasted if you work much at your pc/laptop. If they were cheaper they would be preferred absolutely over hdd by the regular users. Many who use the pc to work not in multimedia doesn’t care about the capacity. And even the capacity could be not a problem if ssd gets bigger. Everything is going smoothly, and prehistoric spinning hard drives are the ones who are dead. I don’t understand why are you hesitant about a new good technology. Most people seems to be excited. May be even samsung in its next offering will aready lower the prices fast, even more han ocz recently did. Am waiting for it. NAND prices keep going down and SSDs too.

Gary July 18, 2008 at 8:39 pm

Antonio: “NAND prices keep going down and SSDs too.”

Do you think HDD prices and performance are standing still. There just isn’t going to be parity in these competing products any time soon (next 5-8 years i.m.h.o.) So when either side claims a superiority, it seems fundamental to question and attempt to prove the claim. Flash (SSDs) will improve with maturity, without doubt. And inherent qualities in flash will bring improvements to users. But this whole debate has been attended by much hype and the usual caveats should be heeded.

Anonymous October 5, 2008 at 7:55 am

Just a remarkable example, beyond extreme speed increase by booting and random IO, SSDs are more robust devices, I am working in hard enviroemnts and this is the 5th time in 2 years that I changed my 2.5 inch rotating notebook drive because of bad sector and malfunctioning because of vibration. With SSD no trouble even when working near a press brake vibrating at 20HZ

james braselton December 15, 2008 at 8:27 pm

Hi there you are worng ssd are just now taking off sony and alien ware offer 256 gb ssd for $1,300 and sony apple and dell have 128 gb ssd for about $500 soo soner or later the hard core gamers advantaly put hard drive sells too zero i say in about 22 years from know no one will know what a hard drive is just like the floppy drive and dail up modems

james braselton February 5, 2009 at 12:28 pm

Hi there Apple 17 Inch Mac Book Pro has a 256 GB SSD for $900 and at New Egg you can buy a 2.5 inch 256 GB SSD $499 soo SDD are not done for yet.

james braselton March 4, 2009 at 8:13 am

Hi there Apple has added a 256 GB SSD flash drive for their 13 inch MacBook and their 15 inch MacBook Pro has the 256 GB SSD option as well.

james braselton March 14, 2009 at 12:48 pm

Hi there. Sega is going to release the first solid state flash drive gaming console which will probably take millions of hard core games away from Sony and Microsoft

Serious Sidius August 17, 2009 at 10:31 am

I agree that SSDs are not quite the “better storage” we were looking for. They also want to claim “data integrity” but keep dropping the bar from 100 years with SLC to 10 or even 5 years with MLC. So I can keep my hard disk in a box for 20 years and come back to find my data intact, not so with newer flash.

“Hi there. Sega is going to release the first solid state flash drive gaming console which will probably take millions of hard core games away from Sony and Microsoft”
You mean like the master system or genesis?
Cartridges are eeprom-based (flash), you realize that?

Daniel Watkins February 6, 2010 at 11:10 pm

THIS BLOG ENTRY IS INVALID!
Tomshardware.com has admitted that the “The SSD Power Consumption Hoax” article was not done correctly. They admitted this in a follow-up article named:”Flash SSD Update: More Results, Answers” URL::
http://www.tomshardware.com/reviews/ssd-hard-drive,1968.html

Exercpt::
“First of all, we want to take this opportunity and apologize to our readers, for we made a procedural mistake when we compared battery runtime of various Flash SSDs, which we used to replace a 7,200 RPM hard drive on a business notebook in an effort to compare battery life of SSDs versus a conventional hard drive. As it was commented by our readers (see comments of that article) and other sources (thanks, George), part of the test procedure was inaccurate because of varying workload. This may cause other system components such as the CPU to be used more intensively, hence contributing to draining the battery earlier than on a slower drive.

The conclusion, however, that Flash SSDs are often misleadingly presented as energy savers to increase your battery mileage on notebooks, is not invalidated. The truth is that more and more Flash SSDs will be increasingly efficient. But many conventional hard drives can also be more efficient than today’s Flash SSDs in the scenarios some of you were demanding: when providing data under a defined workload such as video playback or in idle until the notebook battery runs empty.”

Observe how power efficient the Intel G2 MLC SSD is and the performance it delivers, not to mention the slew of new SSDs from OCZ, Memoright and others. These solid state storage devices will only get more efficient in time.

Dennis April 6, 2012 at 9:43 pm

Well, history has proven this blog entry dead wrong. Personally, I run all my personal computers (desktops and laptops) from Flash based drives, and the firm I work at now uses only SSD-only hi-end PCs attached to a fast NAS over gigabit Ethernet.

HDDs are good for cheap systems, bulk storage and backup.
Like is too short to use them as your main drive if you can afford not to.

KD Mann April 12, 2012 at 12:03 pm

@Dennis,

To the contrary, this prediction was and is startlingly accurate.

In 2008 the analysts forecast 26% of notebooks would be shipping with SSDs in 2011.

http://www.forward-insights.com/present/FMS%20Jim%20Elliott.pdf

Today SSD penetration in the notebook market is still under 3%. Even the flooding of Thailand didn’t boost uptake of SSDs.

http://www.isuppli.com/Abstract/P23626_20120320155355.pdf

What killed SSD adoption in notebooks? Smarter filesystem caching, and more plentiful DRAM means that random IOPS are less and less important. 95% of the performance advantage between SSD and HDD is in random reads, but the need for random IOPS is greatly diminished in a post-XP world…

Leave a Comment

{ 2 trackbacks }

Previous post:

Next post: