![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
I am having unusually bad performance when writing to a softRAID5 array on my server, but more than enough speed if I transfer the same files locally on the computer to the same softRAID5 array. I don't get it, thoughts?
Been working this on another forum and these are the test results, looking for some fresh eyes. Please don't just say soft RAID sucks in write performance, yes it isn't that good but the array test results say it more than meets my needs. Just for some reason it doesn't do that over a network.
Systems are in my signature.
Iperf results Says to me my network is OK
FastCopy test results. All tests are using the same files which are a bluray rip in file mode, this is the type of files the server is going to be working with. In general I see the same performance in using the regular windows copy past through explorer but fastcopy gives me a nice record of the results.
Pushing from client to server ICH9R mirror using fastcopy I can get this: Good
Pushing from client to server softRAID5 using fastcopy I get this: Very bad, and this is what I want to do!
Server pulling from client to softRAID5 using fastcopy. I don't get why pulling files is better than pushing them?
From server ICH9R mirror to server softRAID5 using fastcopy I get this. This is what makes me think the array is capable of what I want.
IO meter results, graph courtesy of Stereodude, raw CSV attached. (well I guess we aren't allowing attachments these days)
Been working this on another forum and these are the test results, looking for some fresh eyes. Please don't just say soft RAID sucks in write performance, yes it isn't that good but the array test results say it more than meets my needs. Just for some reason it doesn't do that over a network.
Systems are in my signature.
Iperf results Says to me my network is OK
Code:
0.0-360.0 sec 33.7 GBytes 803 Mbits/sec (or 98MB/s)
Code:
0.0-10.0 sec 942 MBytes 790 Mbits/sec (or 96MB/s)FastCopy test results. All tests are using the same files which are a bluray rip in file mode, this is the type of files the server is going to be working with. In general I see the same performance in using the regular windows copy past through explorer but fastcopy gives me a nice record of the results.
Pushing from client to server ICH9R mirror using fastcopy I can get this: Good
Code:
TotalRead = 35006.1 MB
TotalWrite = 35006.1 MB
TotalFiles = 898 (37)
TotalTime= 438.68 sec
TransRate= [b]79.80 MB/s[/b]
FileRate = 2.05 files/sPushing from client to server softRAID5 using fastcopy I get this: Very bad, and this is what I want to do!
Code:
TotalRead = 35006.1 MB
TotalWrite = 35006.1 MB
TotalFiles = 898 (37)
TotalTime= 1902.95 sec
TransRate= [b]18.40 MB/s[/b]
FileRate = 0.47 files/sServer pulling from client to softRAID5 using fastcopy. I don't get why pulling files is better than pushing them?
Code:
TotalRead = 35006.1 MB
TotalWrite = 35006.1 MB
TotalFiles = 898 (37)
TotalTime= 644.98 sec
TransRate= [b]54.27 MB/s[/b]
FileRate = 1.39 files/sFrom server ICH9R mirror to server softRAID5 using fastcopy I get this. This is what makes me think the array is capable of what I want.
Code:
TotalRead = 35006.1 MB
TotalWrite = 35006.1 MB
TotalFiles = 898 (37)
TotalTime= 362.14 sec
TransRate= [b]96.66 MB/s[/b]
FileRate = 2.48 files/sIO meter results, graph courtesy of Stereodude, raw CSV attached. (well I guess we aren't allowing attachments these days)