Sean and Liviu,

I have been testing the snapshot you sent me with the modified distributed
code. As I said, I wanted the utilization of the slaves to be better, and
that is why you suggested I give the new code a try. However, I am seeing a
very sparse utilization of the slaves in this version. They sit idle for the
majority of the time! I don't know what is going on, but I think the serial
verision is faster than this. I have slaves sitting idle for 20 seconds,
then working for 1 second, and waiting again for another 20... I am using 19
slaves. Can you look into this?
My master and slave settings are below.


eval.masterproblem.debug-info = false
eval.masterproblem.max-jobs-per-slave = 20
evalthreads = 20

seed.0 = time
seed.1 = time
seed.2 = time
seed.3 = time
seed.4 = time
seed.5 = time
seed.6 = time
seed.7 = time
seed.8 = time
seed.9 = time
seed.10 = time
seed.11 = time
seed.12 = time
seed.13 = time
seed.14 = time
seed.15 = time
seed.16 = time
seed.17 = time
seed.18 = time
seed.18 = time
seed.19 = time
seed.20 = time =
eval.master.port = 15000
eval.masterproblem = ec.eval.MasterProblem

eval.i-am-slave = true
eval.return-inds = false
stat.file = dele.stat