Need help with my Highpoint RAID

[Gargoyle]

I've had a Highpoint RocketRAID 100 for awhile now, and I've been too lazy to switch from Windows software striping (which was set up before I bought the card) to using the card until now.

So I uninstalled the software striped drive and updated the BIOS and driver for the highpoint card.

I went into the BIOS utility and created an array. But, uh, I don't know how to make it show up in Windows XP. I went into the Disk Manager and intialized the drive, but the only thing it would let me do is convert it into a dynamic disk (which I did, and can reverse if need be).

Do I need to go into DOS and format it or something?

Side question, one that's probably been asked a million times, what's the consensus on cluster size? This time I made it 32k, but I can still change it.

While this is probably irrelevant, the two drives are WD400BB's (ata-100, 2mb cache, 40gb). There is a WD800JB on the controller that isn't on the array that will be moved to the onboard IDE later. Right now I'm setting up the array so I can benchmark it compared to the WD800JB, and decide which drive I will put Windows on when I get rid of my current Windows HD, an archaic ATA 66 thing.

Thanks.

[TheLostSwede]

I would choose 32k/32k for general use. At least start with that and bench.

I assume you have the right drivers installed for the card and that it shows up in xp ok. Since you created an array (i guess Raid-0) of 2 drives, it will show up in XP as one big drive which you will have to create a partition and format it as usual. You dont have to make it a Dynamic at all.

[EMT]

I think you can also format in Disk Manager.

After benching quite a bit I chose 16/16 for my WB1200JB's on onboard HPT. 32/32 was almost as good though.

[Mt_Goat]

Try right clicking "My Computer">Manage>Storage (in right column)>Disk Management. If it doesn't ask you to initialize a new disk found see if there is one that is about 80GB that does not have your known partitions on it. Also make sure your device is listed properly in the "Device Manager". As for stripe/cluster size it will depend on what you will be using it for. My WD's like 32/8 and 32/16.

[Gargoyle]

Thanks guys. For some reason the option to format it didn't show up until I restarted. And it took me awhile to realize that "Create New Volume" (New Partition in the Basic Disk case) meant format. I was looking so hard for the word "format" that I didn't even notice. ;D

[Spinner]

Yeah it runs a wizard usually. A lot of my mates get stuck with that, one just the other day actually. Glad you got it sorted anyhow.

[Gargoyle]

OMG-For-Freakin'-Real!!!

I figured my ATTO would inrease a little. After all, consumer level RAID is pretty much software raid, and I didn't think it would be much better than my drives set up with Windows striping.

My reads went up. A lot. Almost 200%. I'm freakin' thrilled!. All this time my PoserRaid™ was only about as fast as a single drive was.

I'm gonna go install some stuff now :D

[EMT]

Sweet. Have you played with latency on the controller to push the write speeds up some?

[Gargoyle]

I believe it's at 192 latency now. I lost powerstrip when I wiped the drives, so after I ditch the old drive and reinstall windows I'll mess with it some more. When I used the Windows striping, 192 was the best.

Maybe you all can help me figure out what to do with my new array:

My plan is to install Windows and my programs on the raid drives and put all my music, movies, and documents on the 80gb drive. I know that if one of my drives fail that I'll loose Windows, but that's not really a big deal.
But do I need to fast read times for Windows, or does that only matter on startup? And should the swap file be on the array with the programs or on the other drive?

The other drive is the aforementioned WD800JB. Here's it's atto on the slave of IDE1 on the HPT controller. It *should* be about the same once I move it to the mainboard VIA IDE.

[Mt_Goat]

Here's how I like to set mine up and it means I hardly ever have to defrag and my system stays at 97% or better. First I createa a 10-12GB partition for windows, apps and most progs. I make the next partition at least 25GB and put "My documents", "Temporary Internet Files" and games or non-essential progs on it. I create a 3-4GB partition at the beginning of my spare drive and put a Windows managed swap file on it. I like it to be this size so it can be fast and not have any problems defraging if needed, too small and you get errors. I like to make the next partition on my spare drive my backup of my system or at least the stuff I don't want to loose off it and do regular backups.

[Gargoyle]

Well crap. I've just got my Windows installed on the array and this is the new atto.

Any ideas?

[Mt_Goat]

After installing Windows, did you defrag? Also try un-applying the latency and re-applying it and I would go as high as it takes. You must keep an array defragged and I like to keep mine at 97% or better.

[Gargoyle]

You're right, it was badly fragmented (19%). But after defragging, my results actually got worse! What's going on?

My writes are higher in this screenshot than the last one because the latency was increased. But how can I get my reads back up to 70 like they were before Windows was on the drive? Did they bench so high because they were empty?

[Mt_Goat]

It still doesn't look right. Look how low the shorter reads and writes are. Try putting the latency at 248 (HPT controllers like very high latecies). My HPT 1520 likes 200. Check your cables also. What do you have the HD jumpers set on?

[Gargoyle]

I'll go check the jumpers. This one is with the latency at 228.

It totally killed my midrange, but helped the smaller range. Weird.

[Gargoyle]

Correction - 224 latency, not 228.

This is the atto after I changed the jumpers. They were both set on Master, I had forgotten that WD's like the jumpers off if they're single drives.

Anyway, the controller card recognizes them quicker but it didn't seem to affect benchmarks.

[Mt_Goat]

Try 248 and see what that does!

Do you have anything else on on the controller? Are you certain both cables are Ultra ATA cables?

After looking at your WD800JB ATTO I noticed that it looks low for one of those. If it is defragged and running right it should top out right around 50K. Your array with the drives and controller you are using should be a minimum of 75K (usually better)

This is what I am getting with mine after loading windows and tweaking.
.
.
.
.

[Gargoyle]

Well, hehe, um... apparently turning off Folding@Home makes a difference :banghead:

Still not as good as before I put Windows on it, but at least the midrange isn't really, really bad any more!

Now if I could only get those 70mb/s reads back...

The cables are indeed ATA 100. They came with the card, those nice teflon coated ones.

[Mt_Goat]

Your HPT array looks like one of my WD1000JB's benched solo. This brings me to the next step. I would recomend breaking the array and leaving the drives as they are on the controller and formatting a small 2GB partition (default cluster size) at the beginning of each one. Then bench each and compare them to see if one is much different than the other. You may have one drive not performing and it is bringing the other down when you put them in a RAID array. I would not mess with the latency of the drive that is running by itself. When you do any benching you must close everything possible.

[Gargoyle]

Yeah, maybe one is slow. I've had one for almost a year before I got the other one. I really don't want to format again though (considering my Windows is on it and I just started to get most of my programs back on). I'll try that one day though.

Thanks for all your help mtgoat! :thumbsup:

[MediaMan]

The infamous Tex beat it into me over a year. 16k/16k is the best way to go.

[Mt_Goat]

MediaMan said
The infamous Tex beat it into me over a year. 16k/16k is the best way to go.

MM,

That may have been true as general rule years ago but the newer WD's, especially the JB's actually do better with other configs. I worked with Tex on some of this while some others did it idepenently so it really does wash out. My WD1000JB's like 32/8 (on HPT 1520) far better than anything else while my WD600JB's like 32/16 (on onboard SI) better. I have swapped the 600JB's to the HPT controller to see if that was the factor but it wasn't, so it must have more to do with the density of the drives. Then I have another machine with 2 Maxtor 740's on a HPT 372 and they like the old 16/16 best by a long shot. So my conclusion is; It is always best to test!

I think Mac and Equito are the other two WD RAID'ers here.

[Gargoyle]

Ok, I thought I knew this but I just realized I don't :buck:.

Which side of the slash is cluster and which side is stripe again? (as in 16/16 and 32/8).
And I'm not sure exactly how to set that up. I had an option for cluster size and an option for allocation unit size (one on the card bios, and one in windows, but i forget which was where), and I set both to 32 in this case. But I didn't see an option for stripe size.

[Mt_Goat]

stripe/cluster

Edit

Sorry I missed the second part.

The option for stripe size is in the HPT bios when you set up the array and the cluster is in the format partition.

[Mt_Goat]

Gargoyle

YGPM

[Spinner]

mtgoat said
That may have been true as general rule years ago but the newer WD's, especially the JB's actually do better with other configs. I worked with Tex on some of this while some others did it idepenently so it really does wash out. My WD1000JB's like 32/8 (on HPT 1520) far better than anything else while my WD600JB's like 32/16 (on onboard SI) better. I have swapped the 600JB's to the HPT controller to see if that was the factor but it wasn't, so it must have more to do with the density of the drives. Then I have another machine with 2 Maxtor 940's on a HPT 372 and they like the old 16/16 best by a long shot. So my conclusion is; It is always best to test!

I think Mac and Equito are the other two WD RAID'ers here.

What about me? My WD1200JB's are royally offended you didn't mention them!

On a more serious note though, yea, I absolutely agree. 16/16 was the defacto a couple of years ago, but now it's not so black and white. However... My WDJB's, on a HPT 372, have always ran best on 16/16, though 32/32, 32/16, 16/8, all worked almost as well. I haven't properly tested my Raptors yet but I get the feeling the 16/16 I'm running them on at the moment, (Silicon Image controller) isn't quite the best for them. So like Mtgoat said, test test test, but 16/16 is always a good place to start.

[Mt_Goat]

Spinner,

Sorry I left ya out. ;)

[EQuito]

mtgoat said
I think Mac and Equito are the other two WD RAID'ers here.
I had Mac's WD's with me for a couple of days and I run some benches but nothing serious though.
I do have a system for repairs here with HPT374 / 2 x WD800JB's RAID0 NTFS 16/16
Here's something interesting, check this out:

[EQuito]

Now I move the same array to my SATA ports:

[EQuito]

and here are my "lousy" 2MB cache Maxtors on the same SATA ports. What's going on here? :confused:

[EQuito]

so..., does the HPT374 sucks big time? JB's like SATA better? or do I have a pair of magic Maxtor drives? :scratch:

[Tex]

You were also testing and empty array before and thus testing the extreme outside 32mb of the array and thats not representative of what you get further into the disk also expect a 5k to 10k loss there alone with smaller drives raided.

When you tested your software raid also... was it on a disk with a good bit of data or empty drives?

Tex

[EQuito]

Hi Tex, thanks for stepping in.
I sure can use some advise not because of the scores (I can care less) but to understand the differences.
All tests were run with an empty array and the configuration is the same for all (NTFS 16/16)
I've tried converting to dynamic when using HPT374 and also used your latency guide but the scores don't change.
They only get better when I move the WD array to SATA.
Next, I'm going to try my Maxtors on the HPT374 to see how they react to the change.

ps: I haven't use software raid at all

[Mt_Goat]

Equito,

First, I have been trying to tell people that the SATA controllers are just faster. Point, I had thw WD SE's on a HPT RocketRaid 133 (HPT-372 controller core) and the combo did well. Then I got a RocketRaid 1520 SATA card (HPT-372 controller core) and it did about 12 - 14% better with the same drives. Since tweaking on my S.I. onboard controller it pretty well keeps up with my 1520 now.

Was the SATA you used an onboard or another PCI card? What other variables could be possible between the two? Were these raw or tweaked?

[EQuito]

First of all, let me apologize to Gargoyle for hijacking his thread... :o

Larry, to answer your Q's:
I'm using the onboard SATA controller, no other variables that I can think of and no tweaks at all.

Since tweaking on my S.I. onboard controller it pretty well keeps up with my 1520 now.
How do you tweak the Si controller?

Now, check this out: the Maxtors (2MB cache) on the HPT374, again NTFS 16/16 empty array. Notice that they scored better than the JB's (8MB cache), specially during 'reads' but the pattern is the same when it comes to 'writes' under HPT... :confused:

[Mt_Goat]

All I meant on the S.I. tweaking was how to config everything. Nailing down the latency and the stripe/cluster.

Have you tried adjusting the latency on the HPT just to see what it does? I don't think the controller is bad since you get good reads but you may need a new bios and or drivers. I would work with the latency first though and remember the HPT's really like the high side. My S.I. likes 96, my 1530 likes 200 and my RR133 likes 248 just like the onboard HPT's did.

[EQuito]

I must be doing something wrong then cause I've tried a bunch of different latency values but none of them make any significant difference. I'm using latest bios and drivers btw.

Thanks for the tips. Cheers!

[Mt_Goat]

As one lase hoorah, have you tried the HPT in different slots or even another machine? I am just as puzzled as you are. After you said that adjusting the latency had nil affect I am even more puzzled and am now wondering if something in the card has broken down.

[EQuito]

I'm sorry, I should've mentioned that is not a PCI card, is the HPT374 onboard controller of the Abit AT7.
No sweat dude, it ain't mine and the customer is happy with the current performance.
I was just trying to understand why the same array behaves so differently under these two controllers and why these cheap Maxtors are faster than JB's.

[Tex]

EQuito said
Hi Tex, thanks for stepping in.
I sure can use some advise not because of the scores (I can care less) but to understand the differences.
All tests were run with an empty array and the configuration is the same for all (NTFS 16/16)
I've tried converting to dynamic when using HPT374 and also used your latency guide but the scores don't change.
They only get better when I move the WD array to SATA.
Next, I'm going to try my Maxtors on the HPT374 to see how they react to the change.

ps: I haven't use software raid at all
And I hadnt seen your posts when I replied. It was meant for the original guy.

When you say you get no differance adjusting the latency? What bios and drivers are you running? From the looks of the atto its sharing a IRQ or needs the latency increased big time for the writes to come up. This is just guessing from the ATTO though.

Tex

[seversphere]

What about two Maxtor 80GB SATA model 6Y080M0 on the NF7-S in raid 0. stripe/cluster suggestions?

[Gargoyle]

Ok, midterms are over and I finally got around to working with my drives again. :) For the last 2 weeks I've been running Windows practically crippled, and the only program installed was MS Word!

Anyway, I've got my Windows running on the 80gb WD 8mg cache. I've just formatted the 2x40's and ran an atto on the array. Doesn't the sudden jump from about 30mb/s read to 70/mb seem a little strange?

[Mt_Goat]

Now it looks like time to try latencies from 248 on down as the high end yields better performance.

[Gargoyle]

Tried it at 192 and 248 latency and got the same results. Here's the 248 screenie. Gained a few mb/s in writes from default :D

I'm still a little suspicious of the sudden jump of read speed, and a little irritated that my 32k read is so low... but after I get some things installed we'll see how it feels :)

[Mt_Goat]

What board is this in and what slot is the card in? Check your IRQ's and see if it is sharing with anything.

[Gargoyle]

In an Epox 8kha+, 4th pci slot from the top. How do I check IRQs?

[Mt_Goat]

the easiest way is to go to;

Start > Run > "msinfo32.exe", then expand "Hardware Resources" and click on 'IRQs"

Then look to see for your controller and see if anything else has the same IRQ assigned.

[Gargoyle]

It seems I do have a conflict. How can I change IRQs?

/edit: well maybe the USB doesn't count. Under Device Manager it says "No Conflicts."

[Mt_Goat]

Try moveing it up one slot to 3rd from the top. I'm pretty sure that PCI slot won't have anything else. You are trying to make the HPT the only thing on an IRQ.