How long does it take to copy 300GB of data?

EnderW

[H]F Junkie
Joined
Sep 25, 2003
Messages
11,249
I'm copying over everything from my 300GB Seagate to my 4 x WD2500JD RAID5 array.
I've got a LSI MegaRAID SATA 150-4, which has less than stellar writing performance.
The Seagate has been making some whining sounds which makes me a little nervous, so I want to get everything off of it and run some tests.

I'm about to goto bed, so hopefully it will be done in the morning.
 
At a pace of 700MB/min, which is a snail like 12MB/sec, it will just over seven hours. If the Seagate drive can read at a more realistic 40MB/sec, you're looking at just over two hours to dump the entire disk.
 
It's going slow alright. 8 hours later and it's only halfway done.
 
I almost got the LSI. Luckily, the fine folks here talked me out of it and I went with the 3Ware 9500S-8, which is a true hardware Raid-5 card.

Costly ($490/NewEgg) but should be worth it.

-Larry

EnderW said:
I'm tired of the horrible write performance on this LSI card. I just ordered a Broadcom RAIDCore BC4452.

It should be over 3x faster at 4 disk RAID5 write performance.
http://www.tweakers.net/reviews/557/19
 
TechLarry said:
I almost got the LSI. Luckily, the fine folks here talked me out of it and I went with the 3Ware 9500S-8, which is a true hardware Raid-5 card.

Costly ($490/NewEgg) but should be worth it.

-Larry
Why on earth would you buy that overpriced POS?

You could have picked up the Broadcom RAIDCore BC4852, one of the best cards available, for $358.



Read what Tweakers.net had to say about the 9500S-8 in that same comparison:
http://www.tweakers.net/reviews/557/32
The 3ware Escalade 9500S-8 scores very good in the desktop and workstation RAID 10 benchmarks. With the exception of the database and mailserver scenarios, the server performance lags behind the competition. The Escalade 9500S-8 and the Escalade 8506-8 failed to perform in RAID 5. The Escalade 9500S-8 will continue to fall behind in terms of features as long as the promised firmware update is not released by 3ware. At this time, RAID level migration and online capacity expansion still haven't been implemented by 3ware. If all this would be offered for a reasonable price, buying the Escalade 9500S-8 could still be considered. Unfortunately, the high price of around € 600 stands in no relation to the performance and the featureset. We wonder if 3ware isn't on a dead end with its non-commoditized RAID implementation. The performance is disappointing, the featureset is not up to the standards of the competition and the architecture is out-of-date.


Compared to the RAIDCore BC4852, which came in second to a $600 Areca card
The performance of the RAIDCore BC4852 does not reach the high levels set by the Areca ARC-1120. Nevertheless, this hostbased RAID adapter shows very good scalability. The weakness of the RAIDCore BC4852 is its RAID 1 and RAID 10 performance, and its high CPU usage in workloads with a large amount of sequential write operations. In real world situations, the last remark should be no real drawback, since only a few applications will generate write rates of hundreds of MB/s. High performance dual processor systems will have more than sufficient processor power to handle the RAID logic. RAIDCore has succeeded in developing a stable and feature-rich RAID stack with user friendly management tools thanks to the self-tuning proporties of the RAID stack. Based on the good performance in our workload simulations, we can conclude that self-tuning indeed works very well. The price of BC4852 is attractive in comparison to 8-port intelligent RAID implementations like the Areca ARC-1120 and the 3ware Escalade 9500S08. For this reason, the RAIDCore BC4852 and its four port sibling deserve the title 'recommended'.
 
But how well is that same card going to perform in a basic PCI slot, seeing as how you aren't (to my knowledge) going to be able to run it in PCI-X mode?
 
mikelz85 said:
But how well is that same card going to perform in a basic PCI slot, seeing as how you aren't (to my knowledge) going to be able to run it in PCI-X mode?
Damnit, I forgot about that.

Hard to say since I don't know of any benchmarks of the LSI card in a regular PCI slot.

I guess it's unlikely I'll see big gains, according to this review of the RAIDCore card in a PCI slot.
http://www.gamepc.com/labs/view_content.asp?id=raid505&page=9

hmm :(
 
EnderW said:
Why on earth would you buy that overpriced POS?

You could have picked up the Broadcom RAIDCore BC4852, one of the best cards available, for $358.



Read what Tweakers.net had to say about the 9500S-8 in that same comparison:
http://www.tweakers.net/reviews/557/32



Compared to the RAIDCore BC4852, which came in second to a $600 Areca card

1) The Broadcom card is not a true hardware RAID card due to the fact that it uses Software XOR and eats up CPU cycles.
2) The Tweakers.net review is old (March) and strangely enough the "promised firmware" they ask for from 3ware came out in January and addressed all the issues they had. On top of that I ran my own tests (which is what matters) and the card performs pretty fucking nice for being in a 32 bit PCI slot.
 
Ender said:
1) The Broadcom card is not a true hardware RAID card due to the fact that it uses Software XOR and eats up CPU cycles.
2) The Tweakers.net review is old (March) and strangely enough the "promised firmware" they ask for from 3ware came out in January and addressed all the issues they had. On top of that I ran my own tests (which is what matters) and the card performs pretty fucking nice for being in a 32 bit PCI slot.

1) Yes, I have noticed it has higher CPU utilization than most of the other cards, but I've got a X2 running at 2.6GHz, so I think I should be fine.
2) I'd like to see some tests with the new firmware. Although, it would have to be big gains for me to change my position, since the price of the 3ware card is getting close to the Areca.

Nice name ;)
 
EnderW said:
1) Yes, I have noticed it has higher CPU utilization than most of the other cards, but I've got a X2 running at 2.6GHz, so I think I should be fine.
2) I'd like to see some tests with the new firmware. Although, it would have to be big gains for me to change my position, since the price of the 3ware card is getting close to the Areca.

Nice name ;)

Well if you have that kind of CPU then utilization doesnt really come into play but my server has a 1.4T-bird which is rather dismal by todays standards.

As for my results... The numbers are not going to be close to what the tweakers.net review was but they are fast enough for my needs. The thing is, if you're building a file server chances are you're not going to buy 10000RPM 74gig raptors, yes they are fast but I along with most others want space. This is why I dont put a lot of value on their review.

For example I have 5 250gig 7200RPM drives with 1 as a hot spare:

8 74gig raptors in RAID5 gives you ~500gigs,
8 250gig drives gives you ~1.6TB.

As for cost... Those raptors run close to $200, x8 is ~$1600, all that for 500GB. For $1600 you could get 6 400GB drives and have some $ left over giving you 2TB's. Or get 9 300gig drives giving you a bit over 2TB, a spare to store in a closet + left over cash.

These prices might be a bit off as I got them off Froogle in a rush but you get the idea.

Having done consulting work for small businesses they usually wont go with RAID5, more often then not they do RAID1 using large drives and its plenty fast for them. As you move up to medium or large business they go RAID 5 SCSI due to proven reliability and speed. WD needs a big manufacturor like HP/IBM/Dell to push their drives, still they havent been around long enough convince CIO's.

I just came back from auditing a casino which involved a datacenter walkthrough.... When you have the $$$ and uptime is #1 you go SCSI, I was drooling over the 100+ TB's they had scattered throughout, on top of that they created an identical datacenter a mile away in case something happened, linked by fiber.

I guess what I'm trying to say is dont base your entire view of RAID5 controllers on 1 review, especially one that doesnt reflect real world situations. Listen to the people that actually have hands on experience with them as their results will be most similar to yours.

P.S. Ender is my real name.
 
Back
Top