Why again is this that impressive? $7,000 for a 150GB RAID? 1GB in 4sec?
So what. I have a 2TB RAID at work connected via fibre to a Mac Pro. I just copied a 2GB folder via Gigabit Ethernet RAID in 10 seconds. I haven't done all of the benchmarks, but you can surely get the kind of performance he's talking about from a standard RAID config - and for ALOT less money.
Sure an Xserve RAID is physically larger than 9 SSD drives, but at $7K for 150GB come on! I can get a 1TB Xserve RAID for $6K, add a fibe card for a couple hundred and I've got a fast big redundant RAID for $800 less than this.
I'll mention that my 4 disk (160gb seagate) array, mentioned in a couple posts up, cost me next to nothing (2 drives were $10 on black friday last year, the other 2 were about $50 each).
You do realize that's physically impossible, right? Gigabit ethernet is only 100mbyte/s at 100% efficiency. The fastest you could be copying those files is 20seconds, and that is _extremely_ unlikely.
Yes I did read the article. That's why I posted. My problem with his entire testing and "comparison" is that he is comparing a SSD RAID 0 to a single WD 150GB drive. He states himself:
"The single drive folder copy performance was identical from a WDRaptor 150 to an Mtron 16GB Pro. All you have to do is add a second Mtron 16GB Pro into the mix and you will cut your combined read/write transfer time almost in half. Add 9 drives and copy/pasting 1GB will take under 4 seconds flat. These drives are incredible people!"
Well, all you have to do is add a second WDRaptor 150 to the mix and the same thing will happen - THAT'S THE POINT OF RAID 0!!!
That one comment invalidates his argument that "These drives are incredible people!" These drives aren't incredible, they perform just like any other drive. SSD has it's advantages, but speed is not the primary one. And if you're looking for a fast RAID, the $/performance ratio is just not there yet.
Actually Gigabit is 1000mbit/sec. So technically 125MB/sec maximum so 16sec. However I was copying from a local SAN volume over Gigabit to another machine with a local RAID, via AppleTalk which uses compression yada yada yada.
Anyway, my point is - even if my test wasn't incredibly scientific and accurate, the throughput of a much cheaper and more enterprise level RAID can easily match the speeds he is getting from this incredibly expensive RAID. (Oh and our RAID is level 5. So the throughput would be even faster if we didn't need the safety of RAID 5)
FYI the Fibre connection is dual 4Gb/sec, and we frequently max that out.
As far as speed, personally 1GB in 4 seconds is much less important to me than large amounts of cheap redundant storage. Thus the reason I'm considering building a simple software raid 5 linux file server. On a related note, a few raid 0 (or 0+1) Cheetah 15k's should get about 80% of the above speed and are likely much less expensive.
The new FiOS HD DVR, arguably the biggest update since Verizon released a DVR, thanks to its external storage support, enhanced multi-room functionality and slick new 16x9 HD user interface.
The most commented posts on Engadget over the past 24 hours.
Now that we've thrown 'em off the trail, use the form below to get in touch with the people at Engadget. Please fill in all of the required fields because they're required.
Why again is this that impressive? $7,000 for a 150GB RAID? 1GB in 4sec?
So what. I have a 2TB RAID at work connected via fibre to a Mac Pro. I just copied a 2GB folder via Gigabit Ethernet RAID in 10 seconds. I haven't done all of the benchmarks, but you can surely get the kind of performance he's talking about from a standard RAID config - and for ALOT less money.
Sure an Xserve RAID is physically larger than 9 SSD drives, but at $7K for 150GB come on! I can get a 1TB Xserve RAID for $6K, add a fibe card for a couple hundred and I've got a fast big redundant RAID for $800 less than this.
I'll mention that my 4 disk (160gb seagate) array, mentioned in a couple posts up, cost me next to nothing (2 drives were $10 on black friday last year, the other 2 were about $50 each).
Did you even click on the read link?
You do realize that's physically impossible, right? Gigabit ethernet is only 100mbyte/s at 100% efficiency. The fastest you could be copying those files is 20seconds, and that is _extremely_ unlikely.
Yes I did read the article. That's why I posted. My problem with his entire testing and "comparison" is that he is comparing a SSD RAID 0 to a single WD 150GB drive. He states himself:
"The single drive folder copy performance was identical from a WDRaptor 150 to an Mtron 16GB Pro. All you have to do is add a second Mtron 16GB Pro into the mix and you will cut your combined read/write transfer time almost in half. Add 9 drives and copy/pasting 1GB will take under 4 seconds flat. These drives are incredible people!"
Well, all you have to do is add a second WDRaptor 150 to the mix and the same thing will happen - THAT'S THE POINT OF RAID 0!!!
That one comment invalidates his argument that "These drives are incredible people!" These drives aren't incredible, they perform just like any other drive. SSD has it's advantages, but speed is not the primary one. And if you're looking for a fast RAID, the $/performance ratio is just not there yet.
Actually Gigabit is 1000mbit/sec. So technically 125MB/sec maximum so 16sec. However I was copying from a local SAN volume over Gigabit to another machine with a local RAID, via AppleTalk which uses compression yada yada yada.
Anyway, my point is - even if my test wasn't incredibly scientific and accurate, the throughput of a much cheaper and more enterprise level RAID can easily match the speeds he is getting from this incredibly expensive RAID. (Oh and our RAID is level 5. So the throughput would be even faster if we didn't need the safety of RAID 5)
FYI the Fibre connection is dual 4Gb/sec, and we frequently max that out.
Erows is right, (2 GB) / (1 (Gb / sec)) = 16 seconds, and that's assuming zero overhead and traffic in the link.
Gotta love google calc: http://www.google.com/search?q=2GB+%2F+1Gb%2Fsec
As far as speed, personally 1GB in 4 seconds is much less important to me than large amounts of cheap redundant storage. Thus the reason I'm considering building a simple software raid 5 linux file server. On a related note, a few raid 0 (or 0+1) Cheetah 15k's should get about 80% of the above speed and are likely much less expensive.