November 15, 201510 yr Ok, So here is my build, My711 https://lime-technology.com/forum/index.php?topic=21470.0 I'm using LSI MegaRAID card with an Intel RAID Twenty-four port Expander Card RES2SV240. disks 1-4 are connected via one port directly to the LSI MegaRAID card. disks 5-8 are connected via one port directly to the Intel expander card. disks 9-12 are connected via one port directly to the Intel expander card. disks 13-16 are connected via one port directly to the Intel expander card. disks 17-20 are connected via one port directly to the Intel expander card. disks 21-24 are connected via one port directly to the Intel expander card. and the two cards are connected via a port on each. When I transfer large amount of data from disk 3 to disk 10 for example, can someone explain the data transmission? What about if I transfer data between two disks that are using the same port connection (e.g. disk18 to disk19)? Will it be faster? I need to identify the bottlenecks in the above tasks.
November 18, 201510 yr It's all read from the source, into system memory and then written to target (and parity which means more reads to memory). Hence the recommendation for ECC.
Archived
This topic is now archived and is closed to further replies.