FAO , nick Hows the head ;) , prepare to be shocked........
15 August 2002, 10:43 PM
#34
Scooby Regular
iTrader: (7)
Join Date: Jun 2001
Location: Stalking Kate Beckinsale
Posts: 4,265
Likes: 0
Received 0 Likes
on
0 Posts
Obviously you have far, far more knowledge than me on this miles......
Have you look at the performance differences between software and hardware RAID using today's IDE RAID controllers?
Si is using 0 purely for speed, not for speed and redundancy.
We are not talking RAID 5 or anything like that, just simple raid for normal users.
Not meaning to have a dig, just no need to be arsy mate.
Cheers,
Nick
Have you look at the performance differences between software and hardware RAID using today's IDE RAID controllers?
Si is using 0 purely for speed, not for speed and redundancy.
We are not talking RAID 5 or anything like that, just simple raid for normal users.
Not meaning to have a dig, just no need to be arsy mate.
Cheers,
Nick
15 August 2002, 10:48 PM
#35
Scooby Senior
Join Date: Mar 1999
Location: Norwich
Posts: 1,835
Likes: 0
Received 0 Likes
on
0 Posts
<do you reckon this could go to the muppet forum.>
Yes! Set him free, he'll be happier amongst his own kind there! It must be nasty cooped up here with sensible, informed people who can communicate easily between one another in a globally recognised language.
Yes! Set him free, he'll be happier amongst his own kind there! It must be nasty cooped up here with sensible, informed people who can communicate easily between one another in a globally recognised language.
15 August 2002, 10:48 PM
#36
Scooby Regular
Sorry, my apologies. I was coming from a server perspective and was blinkered thinking RAID=fault tolerance and forgetting the performance advantages of RAID 0.
15 August 2002, 10:49 PM
#37
Scooby Regular
Join Date: Nov 2001
Location: Leeds - It was 562.4bhp@28psi on Optimax, How much closer to 600 with race fuel and a bigger turbo?
Posts: 15,239
Likes: 0
Received 1 Like
on
1 Post
Have you look at the performance differences between software and hardware RAID using today's IDE RAID controllers?
Hardware is the only one worth bothering with... but si will have no benefit from it as he will break it next week
David
15 August 2002, 10:56 PM
#38
Scooby Regular
iTrader: (7)
Join Date: Jun 2001
Location: Stalking Kate Beckinsale
Posts: 4,265
Likes: 0
Received 0 Likes
on
0 Posts
Hehe, David, true about Si 
. But at a worstation level nowadays as far as IDE RAID controllers are concerned, they really, really are not slow at all. The CPU hit is very small with the speed of chips that we run now. Out of the AMI(LSI), Promise and Highpoint software controllers, the Hp 370 onwards range really do perform very strongly mate. The Adaptec and Promise etc hardware versions really do not shine at all above these guys. The IDE's are a slightly different kettle of fish to the SCSI fellows in my experience.
Cheers,
Nick
(Pcmark 1542 HD performance, Sandra 54002kb/S HD sustained HD throughput)
. But at a worstation level nowadays as far as IDE RAID controllers are concerned, they really, really are not slow at all. The CPU hit is very small with the speed of chips that we run now. Out of the AMI(LSI), Promise and Highpoint software controllers, the Hp 370 onwards range really do perform very strongly mate. The Adaptec and Promise etc hardware versions really do not shine at all above these guys. The IDE's are a slightly different kettle of fish to the SCSI fellows in my experience.Cheers,
Nick
(Pcmark 1542 HD performance, Sandra 54002kb/S HD sustained HD throughput)
15 August 2002, 11:00 PM
#39
Scooby Regular
What I was referring to was software RAID as in NT Mirroring/Striping. No doubt dedicated RAID cards (IDE or SCSI) will be better that NT's in built offering.
But actually, I'd be interested to hear about (non-NT) RAID in terms of:
the difference between RAID 0 and RAID 1 benchmark wise;
the response to the user.
I suspect it's not that great, as in I and most people won't notice the difference but a games freak would.
But actually, I'd be interested to hear about (non-NT) RAID in terms of:
the difference between RAID 0 and RAID 1 benchmark wise;
the response to the user.
I suspect it's not that great, as in I and most people won't notice the difference but a games freak would.
15 August 2002, 11:03 PM
#40
Scooby Regular
Join Date: Nov 2001
Location: Leeds - It was 562.4bhp@28psi on Optimax, How much closer to 600 with race fuel and a bigger turbo?
Posts: 15,239
Likes: 0
Received 1 Like
on
1 Post
Gimme a compaq Smart 5300 anyday.. (still 1k for a raid card)
bastrd preferred supplier prices..
David
bastrd preferred supplier prices..
David
15 August 2002, 11:06 PM
#42
Scooby Regular
iTrader: (7)
Join Date: Jun 2001
Location: Stalking Kate Beckinsale
Posts: 4,265
Likes: 0
Received 0 Likes
on
0 Posts
Miles, at the mo, all that I can give you is a result from my current setup:
Epox 4g4a+ Mobo (Highpoint 372 ATA133 RAID controller embedded)
Intel P4 2.2G Northwood Core
1024 MB PC2700@PC2100 currently
Twin IBM 120GXP Series 7200RPM ATA100 Drives(Striped, 128k clusters)
Twin Maxtor 740 Series 7200RPM ATA133 Drives(Striped, 128k clusters)
Formatted with NTFS. Win XP Pro OS
Cheers,
Nick
[Edited by Mr Footlong - 8/15/2002 11:07:04 PM]
Epox 4g4a+ Mobo (Highpoint 372 ATA133 RAID controller embedded)
Intel P4 2.2G Northwood Core
1024 MB PC2700@PC2100 currently
Twin IBM 120GXP Series 7200RPM ATA100 Drives(Striped, 128k clusters)
Twin Maxtor 740 Series 7200RPM ATA133 Drives(Striped, 128k clusters)
Formatted with NTFS. Win XP Pro OS
Cheers,
Nick
[Edited by Mr Footlong - 8/15/2002 11:07:04 PM]
16 August 2002, 07:57 AM
#45
Scooby Regular
Thread Starter
Join Date: Feb 2002
Location: Lurkin Somewhere
Posts: 7,951
Likes: 0
Received 0 Likes
on
0 Posts
Whats the point of putting **** like that??
Sorry im not a hard core geek like the rest of you i do appologies.
I also for got at 19yrs old you know every technology out there and could do it all.
My deeps apologies
Sorry im not a hard core geek like the rest of you i do appologies.
I also for got at 19yrs old you know every technology out there and could do it all.
My deeps apologies
16 August 2002, 11:06 AM
#47
Scooby Regular
Join Date: Nov 2001
Location: Leeds - It was 562.4bhp@28psi on Optimax, How much closer to 600 with race fuel and a bigger turbo?
Posts: 15,239
Likes: 0
Received 1 Like
on
1 Post
Cool Performance Site
[Edited by David_Wallis - 8/16/2002 11:06:52 AM]
[Edited by David_Wallis - 8/16/2002 11:07:24 AM]
[Edited by David_Wallis - 8/16/2002 11:06:52 AM]
[Edited by David_Wallis - 8/16/2002 11:07:24 AM]
19 August 2002, 10:08 AM
#51
Scooby Regular
Join Date: Mar 2002
Posts: 337
Likes: 0
Received 0 Likes
on
0 Posts
I am still confused. What do you do that means that 50mb/sec is noticably quicker than 40mb/sec and worth the hassle / expense / loss of reliability? Particularly considering the drop in access speed you get with RAID0. You do lots of video editing at home? Meterology? Geophysical anaylsis? Thermonuclear simulation?
I can't think of one thing that would make me put two inherently less reliable drives (IDE - 3yr warranty vs 5yr for SCSI) in a stripe set.
Please enlighten me as to what I am missing...
I can't think of one thing that would make me put two inherently less reliable drives (IDE - 3yr warranty vs 5yr for SCSI) in a stripe set.
Please enlighten me as to what I am missing...
19 August 2002, 10:33 AM
#52
Scooby Regular
iTrader: (7)
Join Date: Jun 2001
Location: Stalking Kate Beckinsale
Posts: 4,265
Likes: 0
Received 0 Likes
on
0 Posts
What you are missing is freedom of choice mate 
Loads of today's motherboards come with these controllers onboard.
There really isn't much hassle involved in striping them.
I have mentioned that it is personal user risk as to whether they want to risk striping their data across 2 drives and that the risk of complete data loss is doubled.
I do actually use my drives a lot for compressing/uncompressing/ recompiling very large files. Again, this is not why I did it, I did it becuse I wanted to.
XP boots faster, ie loads faster, Outlook loads faster, games load faster.
Yes it is noticeably quicker than running IDE singly or running 15krpm SCSI. I have both at home, so I feel able to comment.
Disc access time has not been an issue at all for me, I certainly have not noticed any slowdown whatsoever. This is not being used in a server environment, so tons of of continous calls are not being made to the disk.
I have in no way told people that they should or must use it and I have been luck enough to get 2 of the fastest current IDE drives available, so results will vary.
I use it and I like it and that is all that matters....
Cheers,
Nick
Loads of today's motherboards come with these controllers onboard.
There really isn't much hassle involved in striping them.
I have mentioned that it is personal user risk as to whether they want to risk striping their data across 2 drives and that the risk of complete data loss is doubled.
I do actually use my drives a lot for compressing/uncompressing/ recompiling very large files. Again, this is not why I did it, I did it becuse I wanted to.
XP boots faster, ie loads faster, Outlook loads faster, games load faster.
Yes it is noticeably quicker than running IDE singly or running 15krpm SCSI. I have both at home, so I feel able to comment.
Disc access time has not been an issue at all for me, I certainly have not noticed any slowdown whatsoever. This is not being used in a server environment, so tons of of continous calls are not being made to the disk.
I have in no way told people that they should or must use it and I have been luck enough to get 2 of the fastest current IDE drives available, so results will vary.
I use it and I like it and that is all that matters....
Cheers,
Nick
19 August 2002, 12:57 PM
#54
Scooby Regular
Thread Starter
Join Date: Feb 2002
Location: Lurkin Somewhere
Posts: 7,951
Likes: 0
Received 0 Likes
on
0 Posts
the difference is really noticeable, moving 15gb files around was done in seconds virus scan 4mins to complete where it used to be 15-20mins
Si
Si
19 August 2002, 01:00 PM
#55
Scooby Regular
iTrader: (7)
Join Date: Jun 2001
Location: Stalking Kate Beckinsale
Posts: 4,265
Likes: 0
Received 0 Likes
on
0 Posts
Si, by striping 2 drives together, the absolutely max increase in data throughput is 200%. It is rarely close to this either.
Sorry, but I think that you must have a setting changed on your AV program. Didn't you do a fressh install onto this array?
If so, then chances are that you havent got nearly as much data on there as before either.....
Nick.
Sorry, but I think that you must have a setting changed on your AV program. Didn't you do a fressh install onto this array?
If so, then chances are that you havent got nearly as much data on there as before either.....
Nick.
19 August 2002, 01:15 PM
#56
Scooby Regular
Join Date: Mar 2002
Posts: 337
Likes: 0
Received 0 Likes
on
0 Posts
Impossible to copy 15G in "seconds". With PCI bus running at 133Mb/sec it will still take 2 minutes. Suggest you buy a new watch.
EDIT: OK. Fair enough - you did say "move". If you are moving files on the same disk there is not necessarily any concept of shifting data, just modifying the file system. Anyways, as Mr Footlong says... don't compare your old system to your new one, many more things than your hardware have changed.
[Edited by MrDeference - 8/19/2002 1:19:08 PM]
EDIT: OK. Fair enough - you did say "move". If you are moving files on the same disk there is not necessarily any concept of shifting data, just modifying the file system. Anyways, as Mr Footlong says... don't compare your old system to your new one, many more things than your hardware have changed.
[Edited by MrDeference - 8/19/2002 1:19:08 PM]
Thread
Thread Starter
Forum
Replies
Last Post