Intel's 34nm NAND SSDs launch in two weeks

[Da_maniaC]

Yeah, i'm making the switch to W7 once im replacing my PC.
This won't be anytime too soon though (Im waiting at least for new Intel i7's and SATA interfaces with higher throughput etc).
I decided to restore images since i simple have 40 pcs different programs and about 60 games installed (divided over 2 partitions software/games).
It will take up another weekend once i buy new hardware to install from scratch and migrate all current settings etc.

[fritzman]

I know what you mean.

You've inspired me though... just ordered a 2nd Intel... I miss my Matrix Raid!

[Da_maniaC]

Haha, you're not gonna be dissapointed.
CrystalDiskMark just hit 506 MB/s Sequential read here!

[fritzman]

Haha, you're not gonna be dissapointed.
CrystalDiskMark just hit 506 MB/s Sequential read here!

Cheers.

Did you by chance try CrystalDiskMark on a single at all?

I got 262 MB/s when I first installed my single Intel. I believe they scale, so should expect around what you've got. Back to bed (it's 2:20am here!)

[Ao1]

Imagine you use a horse and cart to get from A to B. Someone then gives you a Porsche and getting from A to B now takes a fraction of the time. Now imagine you tie two Porsche’s together to try and go even faster.

Theoretically it should go a lot faster as it doubles the spec, but in reality it will not go any faster. In some rare instances it may be a benefit but more often than not it will actually be a handicap.

Now consider a horse and cart as a conventional hard drive, a Porsche as an SSD and two Porsche’s tied together as Raid 0.

The cost premium for SSD is worth it, but the cost premium for raid 0 ssd is defiantly not worth it for desktop users.

The only people than benefit from raid 0 are those that benchmark and the manufactures that get to double their sales in a currently limited market.

[Levish]

I think you are right :p There's alot of testing involved in optimizing RAID for a application and most users wouldn't bother, furthermore most users wouldn't bother with a HBA (or even a few different ones) that might be needed to make those tuning options even feasible. Particularly since those HBA's are fairly costly.

[Ao1]

In a way an ssd is a raid 0 system within itself, i.e. you have 10 nand channels that are optimised to work in parallel. That is more than enough for me

[Da_maniaC]

I still feel i have to disagree with you on that one.
Windows and Application startup time also decreases siginificantly compared to a single SSD. (I've been measuring this for the past few days).
For me this is alone is reason enough to still buy a second SSD. (Another reason is my System and Games partition just need 2 x 160GB space + a little overhead).
I can definetly see your point though, especially because the X-25M's cost a lot of money.

[Levish]

If my meager understanding of RAID/Stripe Size/Stripe Width is correct:

If file size is small enough the file is only read/written to a single drive.
If you mean reading files over 1MB in size sure, you are correct.
If you mean the smallest of files you'd be wrong you have single drive performance there no matter how big the array (in RAID0).

[Da_maniaC]

Doesn't that depend on what value you are using for blocksize on your array though?

[Computurd]

there are other benefits of raid, such as write speeds. the write speeds on some ssd are subpar, raiding them up gives you actual tangible measurable results. when you are loading applications and games there isnt a single ssd out there that will match the speed of raid ssd's. this point has been debated and re-hashed so many times in this forum it is ridiculous. it has been proven a number of times that you can and will gain real results with raid and ssd. the main thing to take from this is that you reach a diminishing point of returns very quickly with raid and ssd. once you get over 3 your performance increase with each additional device drops sharply. however, that does not equate to writes. they still scale very very well.

it also depends upon what type of hardware that you are going to use, etc. HBA , full featured raid card, onboard? there are so many setups and personal user patterns that no one can definitively tell you that raid and ssd is a waste of time.

you have to see through the forest to see the trees. many peoples perceptions of raid and ssd is limited to game load times and easily measured obvious things of that nature. after you add MORE than a few ssd's you will not see measurable change in the load times. true. however, that is a pinhole view of computing if what you base your testing on is one area in general, such as game load times. what about performance while in that game? what about overall speed of things such as video transcoding, video editing, searching for music, searching for anything, period, file transfers etc etc etc?
http://www.anandtech.com/storage/sho...px?i=3403&p=14
if you look at that page you will notice the difference in actual performance of your system, the same types of performance increases are also evident with ssd in raid, however it is not easily tested and shown like game load times, so that is ignored. the speed and snappiness of these devices sometimes is not easily measured, and comes down to you, and your perception of how the system 'feels'.
i think basically if you can afford it, do it. if you cant, dont.

[Ao1]

^ No one to date has demonstrated significantly better real life performance with a raid 0 array, however the bottom line is that people should go with what they think works for them. All I am trying to point out is that people who go for raid 0 for additional performance are most likely going to be xtremely disappointed.

As of date there is not a raid card out there that is optimised for SSD. As a consequence there is an inherent loss of efficiency.

Even an optimised SSD PCi solution like the IOXTREME failed to deliver significant real life performance benefit as demonstrated by lowfat

An example of no real benifit with raid 0 can be found here
and another example can be found here

Desktop use is by default general. You can't get one stripe size that works for everything and at the end of the day you are one person making minimal demands in comparison to true high workloads that involve multiple people on a server making demands simultaneously.

Benchmarks don't really mean jack, especially when it comes to ssd. As an example, I believe you pointed out, benchmarks that claim to represent gaming experience work better on large stripes, but in reality smaller stripe sizes are better.

Check out the synthetic gaming workload scores from Anandtech. SSD's are all around the same ballpark. The Kingston SSD is performing as well as the G2, yet the reads on the Kingston SSD Now are only 100MB/s in comparison to a G2 with a read speed of 250MB/s.

Look at what the gaming benchmark actually tests. Four games are run (simultaneously?) for 30 minutes. That does not happen in real life. The game loads for a short duration and then disk activity is limited to nearly zero, with only a few (small) periodical writes occurring before the next level loads. If they made a benchmark that reflected a single game load over 30 seconds with small writes for the next 5 minutes it would be even harder to tell the difference between SSD's.

A G2 benchmarks twice as fast as a Vertex but that is not apparent in desktop use so why would something twice as fast as a G2 seem any faster.

The most important differentiators between good ssd's is not performance it is reliability, cost and write durability. The differentiator between a single ssd and multiple ssd's in raid 0 is extra space. If you don't need the extra space my advice is don't waste your money.

[Levish]

Even an optimised SSD PCi solution like the IOXTREME failed to deliver significant real life performance benefit as demonstrated by lowfat

If you are looking for a particular result thats what you will probably find :p

http://www.tgdaily.com/hardware-feat...indows-in-36-s

[Ao1]

If you are looking for a particular result thats what you will probably find :p

http://www.tgdaily.com/hardware-feat...indows-in-36-s

Fair comment (even though they are comparing hdd to ssd) and Comp is right in that one cap does not fit all. All I'm trying to say is millage will vary and for most people it will disappoint.

[Computurd]

well i dont think that they ran the games concurrently in the anand link, i think it was consecutively.(empasis added)

Our final test focuses on actual gameplay in four 3D games: World of Warcraft, Batman: Arkham Asylum, FarCry 2 and Risen,in that order.

however, that is a technicality.
I agree with much of what audienceofone states, it is truly up to the end user.
alot of the links provided by audienceofone that show no benefit in raid usage, however, focus on load times. i think sometimes load times is kinda like 4k. people just think it is a definite statement of performance, and you definitely want to be careful of making assessments on a few factors only.
for instance, the photshop loading times. what about when you are manipulating images and such when you are actually using the program? there is no data for that. same as the autocad loading times assessment. both of those programs use VERY large files that benefit tremendously from enhanced throughput, however, it is hard to bench such things. when you are manipulating images, is there hangs and lags? how does actual in-program performance handle. how many people just sit around and load autocad repeatedly? unless they are some sort of weirdo, no one. they load the program to USE it. with very large files that are acceseed and manipulated constantly.
one group of people who can definitely attest to this is anyone who has manipulated video files and things of that nature, especially professionals. that is why they go for the z-drives and the pcie solutions for these types of usages. they know that the load time is the most IRRELEVANT thing that they do when they use a program. a developer loads a video. takes 15 seconds. then manipulates/edits that image for HOURS> that is where the performance advantages of the raid arrays really shine., and we arent even speaking of the writing of these types of large documents and files, etc, after they are done working with them.

about the only benchmark that is truly indicative in alot of storage capacities anymore is pcmarkvantage. and that is because it doesn't load synthetics when it does the assessment, it actually performs tasks that end users do on their computers every single day.
such as watching a high def video and also performing other tasks in the background. such as checking mail or working on a project.
importing digital photos into libraries and media players. cataloging those libraries?
what about when you transcode a highdef video to a portable media player? have you ever done that? i have with my HD zune player. takes time.
what if you are streaming to a xbox 360 or ps3 off your rig while you are also recording video from a television source on that same computer? time shifting? what if you are even watching HD video on that computer yourself while these other things are going on as well? do you think that is going to play smoothly?
how many times have you went through levels in your game and noticed lags and hangs when you are going through to other areas of the map? in alot of the newer games streaming data from the hdd comes into play alot, especially MMO and RTS games.
encryption and compression of ANYTHING is also taxing on your storage subsystem.
the computer is becoming more and more of a hub for all activities that are media related in the home. anyone who watched steve ballmers keynote at CES could tell you that. they are showing the direction of computing that it is going for the average end user, and it is multiple things at once and tons of HD vid. what happens if your weekly virus scan kicks in while any of the above things are happening??
pcmark vantage tests 26 different things of this nature in the default benchmark. of the 26 tasks performed only one is application loading. it is important, but not that important. performance DURING use is key.
you can look at pcmark vantage top users and see the real truth. not one person near the top of that is using a single ssd. probably not in the top hundred. they simply do not have the performance. it is all RAID SSD and acards in RAID.
of course this is a heavy heavy use of your computer that is being simulated, but it is USAGE testing. not loading testing and it is the type of usage that everyday users do now, and will do increasingly in the future.