![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
-
Some users have recently had their accounts hijacked. It seems that the now defunct EVGA forums might have compromised your password there and seems many are using the same PW here. We would suggest you UPDATE YOUR PASSWORD and TURN ON 2FA for your account here to further secure it. None of the compromised accounts had 2FA turned on.
Once you have enabled 2FA, your account will be updated soon to show a badge, letting other members know that you use 2FA to protect your account. This should be beneficial for everyone that uses FSFT.
To RAID, or not to RAID...
- Thread starter orkan
- Start date
anotherguy159
2[H]4U
- Joined
- Aug 7, 2002
- Messages
- 2,093
RAID 0 ONLY OFFERS PERFORMANCE INCREASES DURING SEQUENTIAL READS.
SEQUENTIAL READS ONLY EXIST DURING SYNTHETIC BENCHMARKS.
Do not RAID-0 two 74 gig Raptors.
SEQUENTIAL READS ONLY EXIST DURING SYNTHETIC BENCHMARKS.
Do not RAID-0 two 74 gig Raptors.
If that is true, then why is this situation true?:
I had one 120gb sata drive. I use photoshop a lot.
I bought another 120gb sata drive, and set them in RAID-0.
Photoshop ran considerably faster on the raid, than it did with a single drive. Games also loaded faster. My hdd access times were noticably faster with a RAID than without it.
Why is this not true with the raptors?
I had one 120gb sata drive. I use photoshop a lot.
I bought another 120gb sata drive, and set them in RAID-0.
Photoshop ran considerably faster on the raid, than it did with a single drive. Games also loaded faster. My hdd access times were noticably faster with a RAID than without it.
Why is this not true with the raptors?
bs, reads and writes are faster with RAID 0, but due to the increasing chance of failure, RAID 0 sucks. RAID 1 blows for performance, it will always be slower than 1 drive, even though it should have faster reads. But your data is safe with RAID 1. RAID 10 and RAID 5 are nice, but more expensive.
I ran with RAID 0 on my gaming system for about a year. At several LAN parties (where I was sitting side-by-side with non-RAID systems) I noticed that my game load times were slower than non-RAID systems. I had to search to find examples of where my RAID 0 system was faster. Two examples come to mind, running HDTach (a synthetic HD Benchmark) and when unzipping very large files, like 500MB+ in size. In those cases, I was nearly twice as fast as the non-RAID systems. But in terms of real world usage, those were edge cases. I didn't see any significant performance increase in day-to-day usage. So I eventually broke up the RAID into two separate drives because as I added more non-replacable data to my system. I couldn't justify the 50% reduction in MTBF just for the bragging rights of RAID 0 and little else in terms of tangible benefits.
Of course, if money was no object, I'd easily put 3 or more dives together for RAID 5 (SCSI required) or 4 SATA drives in RAID 0+1. But I doubt I'd ever do straight RAID 0 again.
Of course, if money was no object, I'd easily put 3 or more dives together for RAID 5 (SCSI required) or 4 SATA drives in RAID 0+1. But I doubt I'd ever do straight RAID 0 again.
I'm on Raid-0 36gb raptors and regardless of what people spout, its faster than a single drive, no ifs or buts. It's pretty much impossible for two Raid-0 to be outperformed by one of the drives seperately if the striping is done correctly. Everything, including windows and small programs load faster on my system now than they did with 1 36gb.
I dont know why people give Raid-0 so much shit, it's better in almost every way to a single drive if you use it for what it was designed for I.E productive work(photoshop, 3d imagery etc) and handling large files, it was designed for servers and mega-high-end workstations remember?. Gamers need to stop bagging it just cause it won't load they're programs faster than their friends at LAN parties.
If your going to use RAID-0, use it for what it was meant for.
I dont know why people give Raid-0 so much shit, it's better in almost every way to a single drive if you use it for what it was designed for I.E productive work(photoshop, 3d imagery etc) and handling large files, it was designed for servers and mega-high-end workstations remember?. Gamers need to stop bagging it just cause it won't load they're programs faster than their friends at LAN parties.
If your going to use RAID-0, use it for what it was meant for.
Lol, done correctly? As if there is a way to do it incorectly. There is no skill or technique to it, the hardware does it for you. If you are happy with it, great, I'm happy for you. But don't call my experiences "spouting", because your's are no different.
Yeah, I could benchmark my RAID 0 setup and prove with data that it was faster than a single drive. But in the all important "how does it feel" test, I wasn't satisfied. Like I said, I ran with mine for about a year, and I really wanted it to be faster. But for me, the times when it was noticably faster were few, as in the examples I mentioned.
Also don't misunderstand my reasons for breaking up my RAID 0. I didn't break it up because it wasn't fast enough, or because I thought a single drive was faster. I did it because I was beginning to put more and more unreplacable data on the drives, and I was feeling more uneasy about that 50% reduction in MTBF. After about a year, for me, the risk of data loss out weighed the speed gain. Had the speed gain been more significant, I might have done something differently.
Yeah, I could benchmark my RAID 0 setup and prove with data that it was faster than a single drive. But in the all important "how does it feel" test, I wasn't satisfied. Like I said, I ran with mine for about a year, and I really wanted it to be faster. But for me, the times when it was noticably faster were few, as in the examples I mentioned.
Also don't misunderstand my reasons for breaking up my RAID 0. I didn't break it up because it wasn't fast enough, or because I thought a single drive was faster. I did it because I was beginning to put more and more unreplacable data on the drives, and I was feeling more uneasy about that 50% reduction in MTBF. After about a year, for me, the risk of data loss out weighed the speed gain. Had the speed gain been more significant, I might have done something differently.
The thing to remember is that there are two parts of drive speed that matter, and they matter in different ways.
Bandwidth, and access time.
With RAID 0 (really AID 0 as there is no Redunancy in RAID 0) Bandwidth always increases, this is a good thing. The bandwidth of the array is always the number of drives times the data transfer rate of the slowest drive, as long as the stripe size is small enough. This is because each drive has to deliver its stripe of the data chunk before the data can be transfered so the slowest drive (mulitpled times the number of drives) defines the speed.
Access time on the other hand, access time is always at least the same as a single drive or a little worse, never better. This is because all drives need to access their data for the data transfer, so you have to wait on the slowest drive to finnish.
What does this mean in the real world?
Well, the main measurement (Bandwidth) is improved with RAID 0 so data transfers faster from a RAID 0 array, but it takes the same amount of time (or a little longer) to start transfering the data. So if you are working on large files you can see a great benifit from RAID 0.. but if you are working with a lot of small files, the transfers are starting and stopping all the time, and suddenly the access time becomes more important, and the slight (very slight) penalty of RAID 0 can become noticable.. And suddenly you are not gaining much benifit, but are still running the increased risk associated with RAID 0.
So it all depends on how you use your computer if RAID 0 is worth it. For people who manipulate large files regularly (photoshop, video editing, publisher, quark, CAD, map building for games, etc.) then RAID 0 makes sence. For the rest of the world it usually is not worth it.
==>Lazn
Bandwidth, and access time.
With RAID 0 (really AID 0 as there is no Redunancy in RAID 0) Bandwidth always increases, this is a good thing. The bandwidth of the array is always the number of drives times the data transfer rate of the slowest drive, as long as the stripe size is small enough. This is because each drive has to deliver its stripe of the data chunk before the data can be transfered so the slowest drive (mulitpled times the number of drives) defines the speed.
Access time on the other hand, access time is always at least the same as a single drive or a little worse, never better. This is because all drives need to access their data for the data transfer, so you have to wait on the slowest drive to finnish.
What does this mean in the real world?
Well, the main measurement (Bandwidth) is improved with RAID 0 so data transfers faster from a RAID 0 array, but it takes the same amount of time (or a little longer) to start transfering the data. So if you are working on large files you can see a great benifit from RAID 0.. but if you are working with a lot of small files, the transfers are starting and stopping all the time, and suddenly the access time becomes more important, and the slight (very slight) penalty of RAID 0 can become noticable.. And suddenly you are not gaining much benifit, but are still running the increased risk associated with RAID 0.
So it all depends on how you use your computer if RAID 0 is worth it. For people who manipulate large files regularly (photoshop, video editing, publisher, quark, CAD, map building for games, etc.) then RAID 0 makes sence. For the rest of the world it usually is not worth it.
==>Lazn
