Adaptec 4805SAS RAID 5 Review and Benchmarks

[SnoMunke]

Yes...controlller goes wacky when I connect SATA drives to it...I really think it is my HDDs....

 

[Googer]

Originally posted by: SnoMunke
Yes...controlller goes wacky when I connect SATA drives to it...I really think it is my HDDs....

What question are you answering yes to? Did you try to format the drives using the controller bios?

 

[sethfluid]

hi,

i was hoping you could tell me how things went.

we have a box like that and i would HIGHLY urge you not to do what we did if you wish to also use the onboard raid. we still have yet to get it working, in windows or linux.

the areca is solid. the second class on board raid i was hoping to use is still a dream.

good luck,

 

[Googer]

Originally posted by: sethfluid
hi,

i was hoping you could tell me how things went.

we have a box like that and i would HIGHLY urge you not to do what we did if you wish to also use the onboard raid. we still have yet to get it working, in windows or linux.

the areca is solid. the second class on board raid i was hoping to use is still a dream.

good luck,

Author Icon
sethfluid
Junior Member

Posts: 1
Joined: 02/24/2006

Welcome to the forums!

 

[imported_Gibbon]

Well, I've got my system up and running if anyone's interested.

Asus A8N32-SLI
Opteron 170 @ 2.6GHz (10 x 260MHz)
2GB of OCZ Platinum EB PC4000 (2 x 1GB) @ 260MHz, 3-3-2-8
XFX 7800GTX in the first PCIe x16 slot
Areca 1220 8-channel SATA-II RAID controller (PCIe x8 card) in the second PCIe x16 slot
M-Audio Revolution 7.1 sound card in PCI slot 2
FusionHDTV card in PCI slot 3

HDDs are set up as:
2 x 120GB Seagate 7200.8 SATA drives on the NF4 controller (SATA 1 and 2) - boot drive and temp drive
8 x 300GB Seagate 7200.8 SATA drives on the Areca controller in RAID 5 - data drive (one volume)

and there's a Pioneer DVR-110D DVD+/-RW on the Primary IDE controller.
It's running WinXP SP2 32-bit.

Everything seems to be stable and happy 🙂
Happy to provide whatever info anyone might be interested in.

Cheers,
Mat.

 

[Googer]

Originally posted by: Gibbon
Well, I've got my system up and running if anyone's interested.

Asus A8N32-SLI
Opteron 170
2GB of OCZ Platinum EB PC4000 (2 x 1GB)
XFX 7800GTX in the first PCIe x16 slot
Areca 1220 8-channel SATA-II RAID controller (PCIe x8 card) in the second PCIe x16 slot
M-Audio Revolution 7.1 sound card in PCI slot 2
FusionHDTV card in PCI slot 3

HDDs are set up as:
2 x 120GB Seagate 7200.8 SATA drives on the NF4 contoller (SATA 1 and 2) - boot drive and temp drive
8 x 300GB Seagate 7200.8 SATA drives on the Areca controller in RAID 5 - data drive (one volume)

and there's a Pioneer DVR-110D DVD+/-RW on the Primary IDE controller.
It's running WinXP SP2 32-bit.

Everything seems to be stable and happy 🙂
Happy to provide whatever info anyone might be interested in.

Cheers,
Mat.

Can you give us some photgraphic benchmark results compairing the onboard nFORCE to the external PCI-e X8 controller? I would love to see the differance.

 

[imported_Gibbon]

Well I can, but it'd be comparing a single drive against 8 drives in RAID 5 .. I'm not sure that would be a very useful comparison? Happy to do it though, what benchmarking program would be appropriate?

Also, forgot to mention, the onboard silicon image controller is disabled.

 

[Googer]

Originally posted by: Gibbon
Well I can, but it'd be comparing a single drive against 8 drives in RAID 5 .. not sure that would be a very useful comparison?

Single drive compairisons only. Why not disconnect all the drives from the Areca controller and plug a single drive in to it benchmark and then plug that same drive in to the motherboard and re-benchmark. Submit the results to this forum.

 

[imported_Gibbon]

Originally posted by: Googer

Originally posted by: Gibbon
Well I can, but it'd be comparing a single drive against 8 drives in RAID 5 .. not sure that would be a very useful comparison?

Single drive compairisons only. Why not disconnect all the drives from the Areca controller and plug a single drive in to it benchmark and then plug that same drive in to the motherboard and re-benchmark. Submit the results to this forum.

I don't think there's much point doing single drive benchmarks on an 8-channel RAID controller ... nobody in their right mind would run one drive on it 🙂

I'd be surprised if it was much different to the NF4 controller with one drive anyway. The drive itself would be the bottleneck. I'll do it sometime if I feel bored though 🙂

 

[Googer]

Originally posted by: Gibbon

Originally posted by: Googer

Originally posted by: Gibbon
Well I can, but it'd be comparing a single drive against 8 drives in RAID 5 .. not sure that would be a very useful comparison?

Single drive compairisons only. Why not disconnect all the drives from the Areca controller and plug a single drive in to it benchmark and then plug that same drive in to the motherboard and re-benchmark. Submit the results to this forum.

I don't think there's much point doing single drive benchmarks on an 8-channel RAID controller ... nobody in their right mind would run one drive on it 🙂

I'd be surprised if it was much different to the NF4 controller with one drive anyway. The drive itself would be the bottleneck. I'll do it sometime if I feel bored though 🙂

Single drive performance is an indication of how well a controller can add to or take away from the performance of a drive. Some drives will double their performance with a simple change in the controller. You would be surprised at how many people buy a raid card just to run a single drive. I am thinking of doing it, but I want to see a few benchmarks of various controllers first.

 

[imported_Gibbon]

Well, I'd argue that single drive performance on a RAID controller is deceptive, as they're not in any way designed to run one drive. Besides, I'm one of those people who thinks benchmarks in non-real-world configurations are pretty much pointless. I'm certainly never going to run a single drive on this controller, and I'd suggest you'd have to have some issues to buy an Areca card for one drive, given their price 🙂

Like I said, I'll do it sometime if I'm bored, but I'd rather spend my time providing something more useful 🙂

Want to wager that it'd make ah heck all difference? 🙂

 

[Googer]

Originally posted by: Gibbon
Well, I'd argue that single drive performance on a RAID controller is deceptive, as they're not in any way designed to run one drive. besides, I'm one of those people who thinks benchmarks in non-real-world configurations are pretty much pointless. I'm certainly never going to run a single drive on this controller, and I'd suggest you'd have to have some issues to buy an Areca card for one drive, given their price 🙂

Like I said, I'll do it sometime if I'm bored, but I'd rather spend my time providing something more useful 🙂

Thanks, keep me posted.

 

[Googer]

Originally posted by: Gibbon

Want to wager that it'd make ah heck all difference? 🙂

What do you mean?

 

[imported_Gibbon]

Hehe, it just means do you want to bet that it won't make much difference? 🙂

 

[SnoMunke]

Hey, spin this sh!t off on a new thread...Gibbon, you got some good info...don't bury it in my thread...get your own thread and get the attention you deserve.

BTW, I still can't get a stable RAID 5 array going on my Adaptec 4805SAS controller...either it is bad controller or it doesn't like my HDDs...have to call Adaptec Monday and see what they recommend.

 

[Googer]

Originally posted by: Gibbon
Hehe, it just means do you want to bet that it won't make much difference? 🙂

It will depend on your controller, I do no know what it can do. With some controllers there is a performance loss and others like a Silicon Image 3132 it is almost double the number of IO's over nFORCE. Since I am not familiar with Areca I will bet you that there will be a differancce for better or for worse but I will not get any money involved in it. Winner gets bragging rights.