Freeing Up Capacity On An SSD With NTFS Compression

Status
Not open for further replies.

compton

Distinguished
Aug 30, 2010
197
0
18,680
0
I've been wondering about this very topic for a while now.


However, in the conclusion, it is stated that compression ends up writing more vs. uncompressed NTFS, thus consuming more PE cycles. Shouldn't the opposite be true? When writing to the file system, if a file is compressible it should take up less space and therefore conserve more PEs (though actually compressing the files for the first time should result in more writes).

Why does on-the-fly compression result in more writes even though the amount to be written is smaller?
 

husker

Distinguished
Oct 2, 2009
921
8
18,985
0
An interesting article, but seems a bit contradictory. Kind of like buying a Ferrari and then worrying about the gas mileage.
 
G

Guest

Guest
because when you modify even just one byte of a file that is compressed, you can end up changing a significant portion of the file, not just that byte. it's good if you can fit the change in one block erase; what if you can't? you'll end up writing more info on the "disk" then.
 

clonazepam

Distinguished
Jul 10, 2010
2,625
0
21,160
119
I'd been wondering how I would fit the 20+gb necessary for sw:tor on the ssd (i said this previously, in the recent games on ssd vs hdd article). It'd be interesting to see the titles tested in that article with full drive ntfs compression on and off.
 

Marcus52

Distinguished
Jun 11, 2008
619
0
19,010
9
[citation][nom]clonazepam[/nom]I'd been wondering how I would fit the 20+gb necessary for sw:tor on the ssd (i said this previously, in the recent games on ssd vs hdd article). It'd be interesting to see the titles tested in that article with full drive ntfs compression on and off.[/citation]

Keep in mind, whatever storage option you use, you need room to install updates on top of installing the game, most especially for MMOGs. This means room to download the update AND install it.

;)
 
G

Guest

Guest
Why windows are using ntfs instead of zfs or ext4 which are far superior than ntfs?
 

BrightCandle

Distinguished
Dec 24, 2008
11
0
18,510
0
Presumably this negatively impacts Sandforce based drives more than the Samsung by making the data stored compressed? Any chance you can do this with a Sandforce drive to see the impact?
 

acku

Distinguished
Sep 6, 2010
559
0
18,980
0
[citation][nom]cruizer[/nom]because when you modify even just one byte of a file that is compressed, you can end up changing a significant portion of the file, not just that byte. it's good if you can fit the change in one block erase; what if you can't? you'll end up writing more info on the "disk" then.[/citation]

Correctomundo. Compression involves replace repeated occurrences of data with references to a single copy of that data existing earlier in the input (uncompressed) data stream. That's why it's not right to think of a compressed archive as a container that stores any given file into a discrete space. If anything, the files kind of overlap in a big mixing pot.

When you compress on the fly, you have to completely decompress all the files in an archive and recompress it when you're done. Hence it's all random transfers for the most part.

[citation][nom]BrightCandle[/nom]Presumably this negatively impacts Sandforce based drives more than the Samsung by making the data stored compressed? Any chance you can do this with a Sandforce drive to see the impact?[/citation]

It's not a sequential transfer. Plus it's already precompressed data. Nothing SandForce can do about it. SandForce, Samsung, it's not going to make a difference.

Cheers,
Andrew Ku
TomsHardware.com
 

chesteracorgi

Distinguished
Apr 11, 2011
1,956
0
20,460
206
Do you have any data on how much using compression shortens the life cycle of the SSD? It would be most helpful as those with smaller SSDs are the likely candidates for using compression.
 

DanglingPointer

Distinguished
Dec 1, 2009
16
0
18,510
0
The final conclusion mentions increased write cycles with compression shortening the life-span of the SSD!?!?!? Wot the...??? LoL
There is less write cycle since the file is smaller! The increase is in the time needed to compress the file before writing to the SSD. Therefore Using compression lengthens the life of the SSD since u are writing less since the files are smaller!
The author must have been sleepy...
 

Mark Heath

Distinguished
Apr 28, 2010
837
0
19,010
14
[citation][nom]DanglingPointer[/nom]The final conclusion mentions increased write cycles with compression shortening the life-span of the SSD!?!?!? Wot the...??? LoL There is less write cycle since the file is smaller! The increase is in the time needed to compress the file before writing to the SSD. Therefore Using compression lengthens the life of the SSD since u are writing less since the files are smaller!The author must have been sleepy...[/citation]

The above comment shows a lack of reading and/or comprehension skills!?!?!?!? Wot the...??? League of Legends
The question was already asked and answered in previous comments! I can't think of what to put in this sentence. Therefore Using compression can potentially shorten the life of the SSD since u are potentially writing more [see above comments]!
The poster must have been stupid...
 

blackened144

Distinguished
Aug 17, 2006
1,051
0
19,280
0
My 60gb ssd drive died.. Im back running Windows 7 64bit on a 30gb Vertex drive.. It works fine but it does just barely fit the OS and all my programs.. All of my games and steam are installed to another raid0 array.. And to be honest Ive never had an issue with this setup.. I flux between about 5-10gb free depending on what Ive been doing and how long I last ran disk clean up..
 

Lord Captivus

Distinguished
Sep 13, 2011
139
0
18,680
0
I liked the review but the conclusion: "So long as you have a fast-enough processor, there's no reason to not consider enabling it", should have been tested including at least another CPU (Dual-Core).
Thats just my opinion, just to check that Microsoft is right about that performance issue with NTFS.
(Excuse my english).
 
I've been running an 80GB Intel X25M for quite a while now. I think 80GB should be the minimum amount most people would want to consider, as that allows me to have plenty of apps and even 1 or 2 games that I'm playing at the time. 60GB is just too small.

I did use file compression on it, but only on the AppData and ProgramData folders, because those guys suck and can just randomly start becoming monstrously huge. Every now and then I go in there and find some programs like to put a couple gigs in there for no good reason (Adobe Premiere stored several movie files like wtf, and Code Master games (F1 2010, Dirt 3) like to store 1gb replays in there).

However, Black Friday gave me a great opportunity, so I bought a 120gb Corsair Fore Series GT for $110 off (~$150)! This drive is strickly for my games... damn it's nice. Skyrim boots up and loads locations in seconds.
 

hixbot

Distinguished
Oct 29, 2007
818
0
18,990
1
This article reminds me of 1990 when I used a DOS compression program, Stacker, to increase the capacity of my 20MB hard drive to a whopping 40MB.
 
G

Guest

Guest
I don't think this would be as effective for the sandforce 2200 based SSD's because they already do some sort of compression - which is where they get their speed from. You can't compress compressed data, so NTFS compression would have no benefit for sandforce based SSD's

"Like other SandForce controllers, the SF-2281 features a technology called DuraWrite, which uses data compression to lower write amplification and extend the life of the drive by reducing the number of program-erase cycles. This data compression also plays a big part in the controller's performance. The more the data can be compressed, the faster an SSD like the HyperX is able to read and write."

from:
http://cdrlabs.com/Reviews/kingston-hyperx-120gb-solid-state-drive/All-Pages.html
 

TeraMedia

Distinguished
Jan 26, 2006
904
1
18,990
3
I disagree with the way that compression ratios were measured for 7-zip vs. Windows NTFS. You stated that system files were often not compressed. This is probably because Windows couldn't compress an executable that was in-use. But then you compared the 17% compression ratio of the system drive to the ratios of specific folders compressed by 7-zip. You seem to have ignored the fact that you didn't compress a significant number of files in the system folder(s).

You might try the following:
1) Take your SSD with the system partition and connect it to another computer that already has a bootable drive.
2) Boot into Windows on that other computer, go to the attached SSD, and re-compress the system folder as well as any contents in Program Files and Program Files (x86) that were not compressed already.
3) Put the SSD back into your test machine, re-evaluate the compression ratios, and re-test boot times, shutdown times, and various other system tests.
 

MasterMace

Distinguished
Oct 12, 2010
1,151
0
19,460
68
Those using SSDs and storage hdds should make sure that they set up their virtual memory so that none is being used on the SSD. Instead have the hdd set up for virtual memory.
 

clonazepam

Distinguished
Jul 10, 2010
2,625
0
21,160
119
[citation][nom]Marcus52[/nom]Keep in mind, whatever storage option you use, you need room to install updates on top of installing the game, most especially for MMOGs. This means room to download the update AND install it.[/citation]

Oh I know it. I freed up 30gb total and that's not going to cut it =/

:: the excuses to go raid 0 begin to formulate ::
 

Traciatim

Distinguished
Nov 11, 2006
2,057
0
20,460
310
I was reading a guide recently by the guys who make Ultimate Defrag (which is probably the best defrag I've ever seen for those that want to Google it) and they had mentioned that moving unused data away from your beginning portion of your drive (to make room for the good stuff), and then enabling NTFS compression on all EXE and DLL files on your machine is actually quite a large performance gain since new machines sit idle most of the time anyway and spindle disks simply can't feed the data fast enough anyway.

Think Toms could do some research in to that claim? I lack the skills and the hardware to properly test it, but for those of us who value space and price over raw performance I think it's a worthy question to answer.
 

ProDigit10

Distinguished
Nov 19, 2010
585
1
18,980
0
I've been using NTFS compressed format for many years now!
It's not very CPU intensive, like the article mentions.
Even on a netbook the CPU has more than enough horsepower spare to copy/move files around.
Copying them at the maximum bandwidth of the drive and CPU bus, usually raises CPU usage from 1-3% to 20% for slower processors. But still, you have enough CPU spare to do other jobs, even on a Pentium M 800Mhz cpu!

Also NTFS compression causes MUCH faster reads, but also MUCH slower writes!
For instance, my Intel 710 SSD drive can get speeds of upto 3GB/s, write speeds as low as 80MB/s.
The drive is rated at a nominal speed of 5xxMB/s / 4xxMB/s Write.

Another: NTFS compressed harddrives are not recognized by Linux. Many Linux operating systems can read them, but not all write to them, and if they do, they will cause a mess on the harddrive, overwriting clusters of other files!
Therefor, many linux operating systems like those based on Android, or other media players for instance, will read the disk TOC, but will not read the file's content, as they can't deal with the compression algorythm.
 

ProDigit10

Distinguished
Nov 19, 2010
585
1
18,980
0
[citation][nom]Traciatim[/nom]I was reading a guide recently by the guys who make Ultimate Defrag (which is probably the best defrag I've ever seen for those that want to Google it) and they had mentioned that moving unused data away from your beginning portion of your drive (to make room for the good stuff), and then enabling NTFS compression on all EXE and DLL files on your machine is actually quite a large performance gain since new machines sit idle most of the time anyway and spindle disks simply can't feed the data fast enough anyway.Think Toms could do some research in to that claim? I lack the skills and the hardware to properly test it, but for those of us who value space and price over raw performance I think it's a worthy question to answer.[/citation]
I have UD since 2008, and their claim is justified. They base their thought on the surface speed on the outter rings of a disk is faster than the inner speeds.
It would be the same if the engine would be able to rotate the platter from 7200rpm to 10.000rpm on the inner disk, and slow down on the outter, but since hardware manufacturers want to keep cost low, they equip their drives with a single speed motor.

It only works for harddrives, not for ssd's, and speed could go up from 18MB/s read speeds on the inner rings, to 80MB/s on the outter.

There have been many reviews of HDTACH etc done on harddrives on TH, proving this.
 
Status
Not open for further replies.

ASK THE COMMUNITY

TRENDING THREADS