Hi,
Should I stop crunching this on disks that are mechanical ie. hard drives and keep letting it run on ssd's?
I want to crunch as much as possible but also want my CPU to be computing as much as can be at a given time to help the project.
Also getting quite a few errors on this particular project.

Thanks adriverhoef - that fixed the formatting... sorry for the delay in thanking you - suffer from the "You don't have permission to access /forums/wcg/addpostprocess on this server." problem...

wolfman1360 - don't seem to have the 'slow-down' other report when running all mip1 - even on the i7 with 8 CPUs, so don't know that this will help. Most of my machines have a single SATA drive with platters. The machine here most likely to slowdown would be one i7 - but it doesn't.

This Intel i7 has conventional spinning disk drives for the wcg 'work' files, WD 'Red' NAS drives. However, the partition wcg uses is a RAID1 with two drives... this has the benefit of sustaining two data transfers simultaneously.

[root@danda ~]# hdparm -t /dev/md2 # one transfer

/dev/md2:
Timing buffered disk reads: 408 MB in 3.00 seconds = 136.00 MB/sec


[root@danda ~]# hdparm -t /dev/md2 & hdparm -t /dev/md2 # 2 transfers
[2] 26647

/dev/md2:

/dev/md2:
Timing buffered disk reads: Timing buffered disk reads:
404 MB in 3.00 seconds = 134.63 MB/sec
414 MB in 3.00 seconds = 137.78 MB/sec
[1] Done hdparm -t /dev/md2
[2]+ Done hdparm -t /dev/md2

[root@danda ~]# cat /proc/mdstat | grep -A 3 'md2'
md2 : active raid1 sdd2[1] sdc2[0]
10485696 blocks [2/2] [UU]
bitmap: 2/160 pages [8KB], 32KB chunk

As you can see - two reads at once - no slowdown... maybe that helps.

Thanks adriverhoef
Was a bad decision when the original html was created - and forgotten about. Am about 80% the way through a conversion from .html to .php with variables and realized that anomaly had never been fixed, added to the todo list.

Since you noticed - pushed it to the top and now completed - completely renamed both...

My Microbiome jobs seem to use CPU time very sporadically. Typically, I have 8-10 jobs running at one time but the Progress bar seems all over the board. It appears all the jobs don't really run at the same time. Hard to really put into words, but has anybody else noticed similar results with this project?

My impression is that the work units have increased in size recently. Mine have gone from 1.5 hours to about 4.5 hours, starting week or so ago. Other than that, they seem to be processing normally. Given that we are doing many different organisms in the human biome, I would expect that the work units will vary quite a bit over time.
Cheers