Re: tar x cpu bound

[Edgar Fuß <ef%math.uni-bonn.de@localhost>]

> You said nothing about the system you are running this on.  How many cores?
Opteron 4226, six cores.

> What kind of disks?
RAIDframe level 5 on five Samsung PM1643a 1,92TB SAS SSDs on a SAS9207-8i 
(mfii).
Varying SpSU on the RAID (that's why I'm doing this in the first place).

> Are they using MSI/MSIX to distribute interrupts to multiple CPUs?
Oops, no. Neither on -8 nor on -10.

> What file system and what mount options?
FFS with quotas (newfs -O 2 -q user -q group), varying block sizes (see above).

> Perhaps you should run your tests to extract to a tmpfs, to remove vagaries
> from hardware from the experiment and to establish a baseline.
Well, the original intent is to test various SpSU/fsbsize combinations on the 
test system before updating the NFS server with the SSDs.

EF> The second question is what the single number without a backtrace means.
CB> That is the number of time that the backtrace was sampled.
An empty backtrace?