1. Are you able to narrow this behaviour down to a particular test by removing one of them each test run (or by running only the disk tests or all the others without the disk test)? Is it always happening after a similar amount of time?
2. More tests does not mean faster tests, by running so many disk tests at once there is likely to be a bottleneck somewhere in the system slowing it all down. The time to complete a disk test cycle depends on the test file size you are using, the disk read/write speeds and the available system resources. Running them in conjunction with the CPU (and other) tests will also slow it down more.
For example if you're using 2TB hard drives with a 15% test file size (~300GB test file) then with a read speed of 100MB/s and write of 70 MB/s you should see a cycle complete in around 2 hours (not including time spent verifying). However every additional disk test you run will add more load to the system, and once it becomes bottlenecked then every test started after this point will only slow it down more.
3. What duty cycle levels do you have set for the tests you are running? Is your CPU being fully utilised during the test? It may be if you are running the same combination of tests as above (with 96 disk test) the system may not be doing much except constantly servicing read/write requests to the hard disks.
Announcement
Collapse
No announcement yet.
Linux BurnIn 3.1 HDD Tests Problem
Collapse
X
-
Linux BurnIn 3.1 HDD Tests Problem
Hi Admin,
I tried to run 96 pieces of HDD with Linux BurnIn3.1, and found below problems:
1. BurnIn program will auto quit sometimes when run CPU, Memory, Serial port, 2D video and 96 pieces of HDD.
2. After running 48 hours, only 1 test cycle are performed on HDD tests.
3. The power comsuption have not any changes (about 800W) when running whole system stress with Linux BurnIn V3.1, but I can find there are big changes (about 1400W) with IOMeter program.
Please kindly advise, thanks in advance.Tags: None
Leave a comment: