![[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.
15K Raid 6 performance
- Thread starter tangoseal
- Start date
I added a raid6 array to my WHS for bulk storage
five seagate barracuda LP 2TB on an adaptec 5805. BBU installed.
More disk efficient than DE and the power management of the 5805 spins down the array
Below are the crystal disk mark results of benchmarking the array
five seagate barracuda LP 2TB on an adaptec 5805. BBU installed.
More disk efficient than DE and the power management of the 5805 spins down the array
Below are the crystal disk mark results of benchmarking the array
Go and read on the differences of raid 10 and Raid 6 and you will self answer. Also read what I posted:
-=:Below you can see the basic benchmark for the throughput. Now keep in mind this is Raid 6 with a way more intense parity calculation going on than raid 5 would have. Also you lose half the phys disk worth of storage due to the extra parity but you can withstand 2 disk failures in any order unlike raid 10 which can have an unlucky 2 be a total loss.:=-
- Joined
- Aug 3, 2004
- Messages
- 24,315
not likely, the drives them selves cant go above 3gbps so it doesnt matter.
drescherjm
[H]F Junkie
- Joined
- Nov 19, 2008
- Messages
- 14,933
Delete. Someone already said what I was going to say..
I just don't see the point of raid 6 with 4 drives. 12 drives, sure, you don't want to take a capacity hit of 6 drives when you can just lose 2, but a total 4 drive array and you use raid 6? You are giving up a lot of performance for the (very) small likelihood that two drives will die at the same time. Doesn't make any sense to me but whatever, it's your server.
- Joined
- May 7, 2007
- Messages
- 13,442
nvm
Last edited:
Why in the hell did i share this if people can't just say shh cool and move on. There are ppl freaking defending my choice to use raid 6 when it shouldn't even be necessary. Some of you have no IDEA why I chose raid 6, if it's even permanent, and the history of the server, the consultation process with the customer. Does every server need 12tb of storage???
Any ways as the op of this I never recall asking for inputs on why I should have used raid x instead.
Any ways as the op of this I never recall asking for inputs on why I should have used raid x instead.
- Joined
- May 7, 2007
- Messages
- 13,442
Sorry for defending your case. How the fuck are we supposed to know if it's temporary or not when you don't explain jack shit. 
Hello and welcome to the internets.
I wasnt saying those who defend my setup shouldnt have. I was saying there should have been no reason for people to go to bat for the setup defending it and all because other people quickly assumed that this was a dumb setup designed by an idiot. And that happens on these forums all too much.
Sorry for defending your case. How the fuck are we supposed to know if it's temporary or not when you don't explain jack shit.
I appreciate your original comments. And wasn't shooting at you.
My original post sais' at the very bottom:
"Please keep in mind I am only sharing this for fun and to add to the community. I am not going to review these disk anymore really. They are in a production server at a client I do work for and they are very pleased thus far with the performance."
Yeah I compared it to the original first post and was like wow what is going wrong here??? Then I also seen how the speeds reduce as he was testing larger and larger file sizes. Which im actually glad he did post those results.
- Joined
- Aug 3, 2004
- Messages
- 24,315
Sorry for people being honest and correct something you may think, which can be wrong, which in turn could mis-lead your client like thinking 6gbps would make the system faster when drives cant go that fast.
People will comment on decision others make because believe it or not, we don't always make the best decisions even though we often think we do.
People will comment on decision others make because believe it or not, we don't always make the best decisions even though we often think we do.
Only 115 IOPS? That seems painfully slow. Is this just because your benchmark isn't queuing requests? What are the queue depth counters in Windows showing you?
I think it is just HD tune. HD Tune from what I have been reading is not a good benchmark. Its pretty but thats about it.
Im going to run Crystal Disk mark next time im logged into this server.
Is there anything else I should run?
Once again I did not say they were going to push 6gbps. I did NOT say that. Read the flippin post people.
I said it could be higher. Could be a questionable statement. Means maybe but doubtful. The disk are rated for 6 but the interface they are on is 3.
Is this any different than claiming that a GTX580 is 16x preferred but is only running on 8x link? In benchmarking you want everything to deliver 100% as designed on all interfaces and connections.
So yeah Im not sorry if you are sorry.
CDM has exactly the same problem -- it only issues one outstanding I/O request.
iometer
- Joined
- May 7, 2007
- Messages
- 13,442
I know you are getting sick of people stating these kinds of things, but you need to rework your original post to show your main intent and show any other technical details.
I know you think you did this, but when many people are calling you out on things you didn't intend for, it may be time for a re-write.
I do appreciate you posting your results though, it's definitely good-to-know info.