I have had two clients ask me about the Drobo Elite, an 8 bay BeyondRAID SAN solution from Data Robotic Inc. I’m already a fan of the 4 bay Drobo for USB 2.0 and FireWire 800 equipped machines, so I started reading up on the Drobo Elite. My first interest was in the disk speeds. I wanted to know what kind of data access speeds workstations could expect when connecting to the Drobo Elite via iSCSI. To my surprise, no benchmarks were available online. So when I had the chance to evaluate the Elite first hand, I ran some tests and put it through the paces.
My network centers around a 24 port gigabit D-Link switch. But since this is an unmanaged (consumer quality) switch, I wanted to make sure the fairly low end switch would not hinder test results. To remove the switch from the equation, I simply attached a Macintosh to each of the Drobo Elite’s gigabit network ports directly via the Cat-5e network cables included with the Drobo. In the end, the tests I ran via a connection through the D-Link switch were directly comparable to the direct connection to the Drobo, but the following benchmarks were conducted via direct connection.
One Mac was a first generation quad core 2.66GHz Mac Pro. The other connection was to a 2.53GHz Unibody MacBook Pro. Both have onboard gigabit networking. Both were running OS X 10.6.3. Both of their network settings were configured manually to 1000baseT, full-duplex, and with a Jumbo MTU of 9000. Though directly connected, the machines were assigned static IP addresses on the same subnet as the Drobo. Since some auto negotiating switches sometimes flip out in odd situations while trying to re-negotiate a connection in the fly, I wanted to make sure a similar situation would not sacrifice speed in the set situation.
Similarly, the Drobo Elite’s iSCSI settings were assigned static IP addresses on the same subnet and the MTU was set to 9000.
Though the Drobo has 8 drive bays, for the purpose of testing i used 3 Hitachi Deskstar 2TB SATA drives. The drives were 7200 RPM with 32MB buffers. The Drobo was set up with 2 iSCSI targets each with a 1TB max size. These targets were attached via iSCSI to each computer simultaneously while the tests were run.
Each drive was tested locally in order to establish the baseline performance of each computers internal drives. These numbers also gave me something to compare the overall performance of the Drobo to. Ideally, the tests over iSCSI would yield scores comparable to a computers internal drive mechanism.
MacBook Pro Internal Boot Drive:
Mac Pro internal 2TB Non-Boot Drive:
The remainder of the tests were run from the MacBook Pro using SpeedTools Utilities Pro from Intech. Large transfers were moving either to or from the Drobo Elite via the Mac Pro. While the massive transfer was under way, I wanted to find out what another machine on the LAN could expect for performance. For the sake of testing, 1 set of files contained many large media files of about 10GB each. Another test was entirely very small web design html, gif, and jpg files (500,000+).
MacBook Pro Test Results: Large File Write
(MacBook Pro running bench test while massive large file transfer is being written to the Drobo from the Mac Pro)
MacBook Pro Test Results: Large File Read
(MacBook Pro running bench test while massive large file transfer is being read from the Drobo to the Mac Pro)
MacBook Pro Test Results: Small File Write
(MacBook Pro running bench test while massive small file transfer is being written to the Drobo from the Mac Pro)
MacBook Pro Test Results: Small File Read
(MacBook Pro running bench test while massive small file transfer is being read from the Drobo to the Mac Pro)
Without having any idea what to expect from the Drobo Elite in terms of read/write performance, I was impressed. My 4 bay Drobo isn’t even close to the fastest external USB 2.0 drive in my collection. But it is the largest. And it has redundancy which is uncommon in external drives. I was worried that the Elite would suffer the same sort of performance hit. That proved not the be the case. While it didn’t test out quite as fast as an internal SATA drive on the Mac Pro, it compared favorably to the internal drive of the MacBook Pro (5400 rpm 500GB).
Additionally, the speeds returned by SpeedTools showed less of a performance hit than I expected when a massive file transfer was already under way between the Mac Pro and the Drobo Elite. Ideally I would have had a couple of additional computers moving data to and from the Drobo while I ran the tests, but I didn’t have any other gigabit Mac’s handy and didn’t want to throw Windows into the mix and muddy the waters.
There is one more consideration when using an iSCSI solution. All of the data is moving across the network port of the computer. This puts a considerable demand on the processor. On the Core 2 Duo MacBook Pro, while doing a sustained file transfer, I was seeing about 25% constant usage on each of the two processor cores. On the quad core Mac Pro, I was seeing about 10% processor usage. This is directly inline with the processor usage I see when I do a large file transfer to a conventional network file server. USB 2.0 suffers a simpler processor related performance hit while FireWire transfers offload the overhead to the FireWire controller leaving the processor virtually unaffected.
Needs differ and results will vary depending on the number of networked users, the congestion of the network, the speeds of the drives installed in the Drobo, and the type of data being moved around. It is my hope these benchmarks offer some kind of idea what sort of speeds the Drobo Elite can offer.