Laboratory tests: Thunderbolt RAID blows away FireWire 800


The Thunderbolt-ready devices waiting is over. Four months after the beginning of the clap of Thunder in the range MacBook Pro redesigned Apple, peripheral of storage and support from Intel interconnect technology finally arrived. And although these initial products are a significant price tag, the first tests of laboratory Macworld show Thunderbolt offering a great leap forward in terms of performance.


On Tuesday, promise Technologies Announces the availability of the Pegasus R4 and R6. These external RAID systems, is four (Pegasus R4) or six (Pegasus R6) berries of discs, feature two Thunderbolt ports and range in price from $999 for a R4 with four 1TB 7200 hard tours, $ 2,000 for a R6 with six drives of 2TB 7200 RPM. The necessary Thunderbolt cable is not included in each unit, but is available from Apple for $50.


With appropriate maps, Thunderbolt has the ability to support many types of connections of devices (including HDMI, USB, FireWire, Fibre Channel and more). In addition to storage devices, you can connect Mini DisplayPort displays, such as Apple LED Cinema Display ().


Our first tests with the 12 equipped to coup Thunder R6 concluded that it is considerably faster than similar devices, we tested - devices that use FireWire, USB, or both. The R6 is Thunderbolt-only, so we could not isolate the differences in performance of the connection by testing the R6 with a different interface. As a point of reference, we compared the R6 in another array of RAID Promise, $799 SmartStor DS4600 () with four 1TB drives (unfortuantely, we do have access to a table that is similar to the R6). The R6 and DS4600 were tested as RAID 5 devices.


We used two different systems for testing the RAIDs: a 15 inch 2.2 GHz Core i5 MacBook Pro with a drive solid-state 256 GB and 4 GB of RAM; and a 27 inch 2.7 GHz Core i5 iMac with a 1TB hard drive and 4 GB of RAM.

AJA
System
TestAJA
System
Test2GB
File2GB
File2GB
Folder2GB
FolderPhotoshop
CS5Test the results are in Mbits/S (more is better), except for Photoshop CS5, which is in seconds (lower is better). Best results in bold.

When it is connected to the iMac, AJA System Test of the R6 read scores came 566MBps - 6.8 times faster than the DS4600 using FireWire 800. More impressive still was 644MBps AJA System Test of the R6 write the partition, which is 11.5 times faster than FireWire 800 of the DS4600 scores.


Copy a file of 2 GB of internal hard drive of the iMac the raid was 30% faster on the R6. Copy the file of 2 GB of RAID to the internal drive of the iMac has been a little more than two times faster than the R6 as the DS4600 used FireWire 800. Copy of record results were very similar to the results of the copy of the file tests. A Photoshop test low memory with the help of large files and the RAID as a scratch disk showed the R6 to 37% faster than the DS4600.


AJA
System
TestAJA
System
Test2GB
File2GB
File2GB
Folder2GB
FolderPhotoshop
CS5Test the results are in Mbits/S (more is better), except for Photoshop CS5, which is in seconds (lower is better). Best results in bold.

When it is connected to the MacBook Pro (which has a slower than the iMac, but an internal SSD processor faster), we see that the results of the trials of the R6 AJA system are very similar to the results when the R6 is attached to an iMac. In our tests involving the file of 2 GB and 2 GB of files folder, the R6 has benefited in the solid state drive in the MacBook Pro. In all of these tests, R6-MacBook Pro/DSS combination exceeded the R6 attached to an iMac with an internal hard drive.


USB 2.0 and FireWire 400 are included for the sake of comparison. Feel free to snicker at the AJA System Test faster write speed more-than-20-times that the R6 positions on the DS4600 connected via the USB port.


Come back soon for more results, including eSATA, RAID 0 and the reports on the use of love at first sight for both target disk mode and display of the target. As Apple, thunderclap, we will just to start.


[James Galbraith is Macworld Lab Director.]

0 Comments:

Post a Comment