Q&A: Tom's Hardware And Kingston On SSD Technology

Page 2 - Seeking answers? Join the Tom's Hardware community: where nearly two million members share solutions and discuss the latest tech.
Status
Not open for further replies.

obarthelemy

Distinguished
Feb 20, 2007
40
0
18,530
My question is:

Why don't SSDs work as caches for the main HD ? Their price is so high, and the usage patterns for which they are worthwhile are so specific, that at least in the consumer space it would make a lot of sense to simply use SSDs to cache the most read parts of the regular HD, instead of filling them with whole OSes and Apps, of which 50-75% probably don't deserve to be on the SSD. Windows already has ReadyBoost, which would require a very minimal amount of tweaking to work that way.
 

spinoza2

Distinguished
Dec 20, 2009
2
0
18,510
"I've had a negative experience with the Kingston V- series SSD."

I also have one of these drives (a 128GB) in my MacBook Pro. After several months of solid use, I can't say anything negative about it. I'm not getting the heat problems you mention, in fact, unlike the Western Digital drive it replaced, the fan almost never turns on now. Compared to the Western Digital, the performance has been nothing less than amazing, it's wicked fast, as they say. Without question this was one of the best $250 I've ever spent on a computing product. If this is what drinking the SSD "kool-aid" is about, then it's awfully sweet...
 

ordcestus

Distinguished
Feb 9, 2010
156
0
18,690
[citation][nom]ta152h[/nom] and I'm too old fashioned to feel something is a computer without some form of magnetic storage in it.[/citation]

Yeah i know how you feel. The only download manager i trust is wget, but strangely enough i don't have a computer without an ssd(Ocz too take that kingston)
 

lkaneshiro

Distinguished
Feb 9, 2010
4
0
18,510
Hello,
You're correct the answer really did not show up in the article. The wear difference between the random and seq comes about because the controller has to manage which data is good and bad and in a highly fragmented SSD with lots of random writes it takes more time and write cycles to move data around in the background to make room for the new incoming writes (garbage collection). This fragmentation of the data leads to higher write amplification meaning you're writing to more NAND cells than is needed. Purely sequential writes are not as taxing on the SSD. Of course in most client computing there is mix between randon and sequential writes.
 

lkaneshiro

Distinguished
Feb 9, 2010
4
0
18,510
[citation][nom]neiroatopelcc[/nom]Maybe it's just me, but I don't feel they properly answered the question of why there's a wear difference bewtween sequential and random ...[/citation]

You're correct the answer really did not show up in the article. The wear difference between the random and seq comes about because the controller has to manage which data is good and bad and in a highly fragmented SSD with lots of random writes it takes more time and write cycles to move data around in the background to make room for the new incoming writes (garbage collection). This fragmentation of the data leads to higher write amplification meaning you're writing to more NAND cells than is needed. Purely sequential writes are not as taxing on the SSD. Of course in most client computing there is mix between randon and sequential writes.
 

lkaneshiro

Distinguished
Feb 9, 2010
4
0
18,510
[citation][nom]El_Capitan[/nom]I like how they say, "The worst kind of writes that you can apply to an SSD are random. You will wear a drive out quicker that way". However, Kingston and Intel put all their advertising efforts into promoting the speed of their IOPS for their SSD's for server environments. That means they want you to buy their product to use it so it wears out quicker... which means you need to buy another one to replace it. Now that's a wicket smart business strategy.[/citation]

Actually SSDs have a better than expected endurance. I think what they were pointing out was that any kind of write is going to affect endurance, random writes are harder on a SSD than sequentials. Quoting Intel on the Kingston E series 160GB SSD on a purely 100% random 4KB write one could write 2 Petabytes total data to an SSD conversely if the writes are 100% sequential 4KB then endurance goes up to 10 Petabytes total over the life of the drive.
 

agibson2009

Distinguished
Apr 29, 2009
1
0
18,510
SSD? Intel went out of business in that market several times. Left their some of their customers high and dry.
For replacement purposes, 32GB drives are needed, performance is not an issue and over rated, just so long as it is faster than a Hard Drive.
 
G

Guest

Guest
Why would you put this in a USb enclosure? You know that's gonna be slow.. and as for usb3? and being boot=able.. well, thats what esata is for .. Laughs .. (while selling SHerbet to Cokefiends)
 

stutz lannigan

Distinguished
Feb 9, 2010
1
0
18,510
[citation][nom]kuffru[/nom]Why would you put this in a USb enclosure? You know that's gonna be slow.. and as for usb3? and being boot=able.. well, thats what esata is for .. Laughs .. (while selling SHerbet to Cokefiends)[/citation

I think the point was to illustrate what USB 3.0 is capable of at 4.8Gb/s
Would I put an SSD in an external enclosure? Probably not but with USB 3.0 I could. At USB 2.0 speeds forget it but I've already seen demos at CES where a proto USB 3.0 enclosure and a SSD were hitting over 200MB/s read speeds. As for eSATA, wave goodbye that was just a stop gap before USB 3.0 launches, with no real support for USB Flash drives, dodgy hot swappability and not much laptop adoption it will eventually fade out. The Kingston guy was saying with USB drives getting larger and larger USB 3.0 is needed. The dude that buys a 256GB USB flash drive is probably going to copy some a crap load of data over. Try copying a HDD image at USB 2.0 speeds. fugheadaboutit....
 
G

Guest

Guest
SSD is not reliable yet and also due to NAND flash's number of read/write limitation (Usable life). I would have go and buy Western Digial's SSD instead of Kingston's SSD.
 

lkaneshiro

Distinguished
Feb 9, 2010
4
0
18,510
Western Digital does not make their own NAND, they will buy from Toshiba, Micron, Samsung of maybe Hynix. Why would a WD SSD be more reliable than anyone elses using the same NAND the others do?
 

cart0181

Distinguished
Jun 2, 2007
24
0
18,510
[citation][nom]milktea[/nom]I like the idea that even if SSD goes bad, I can still recover my data possibly avoid paying thousands of dollars for the recovery. That is one thing that pulls me away from hard-disk into SSD.[/citation]

Yeah, that part IS great (not being sarcastic), but did you notice the much more common problem that you WILL encounter? If I read this right, if you delete a file using shift+del or empty your recycle bin, there will be no possible way for you to recover that file ever again. With HDDs that data stays there until something else overwrites it and can be recovered quite easily.
 

Gpimptastic

Distinguished
Mar 24, 2011
2
0
18,510
I have had 2 of 15 of these drives fail after only being in service for 3 months. One was installed in a desktop and the other in a laptop. I had the Kingston SSD v series drives in service for about a year with no problems. The V100 series seem to have reliability problems. I tried all of the tricks I use on mechanical drives but could not retrieve the user's data.
 
Status
Not open for further replies.