[Pw_forum] The low usage of CUPs

[Giovanni Cantele]

wangqj1 wrote:
> Dear pwscf users
> I use 16 CPUs to run a job,but the usage of CPUs is very slow ,it as like:
> Tasks: 179 total, 1 running, 178 sleeping, 0 stopped, 0 zombie
> Cpu0 : 0.0%us, 4.8%sy, 0.0%ni, 90.3%id, 0.3%wa, 0.0%hi, 4.5%si, 0.0%st
> Cpu1 : 0.0%us, 1.0%sy, 0.0%ni, 95.8%id, 3.3%wa, 0.0%hi, 0.0%si, 0.0%st
> Cpu2 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
> Cpu3 : 0.7%us, 0.0%sy, 0.0%ni, 99.0%id, 0.3%wa, 0.0%hi, 0.0%si, 0.0%st
> Cpu4 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
> Cpu5 : 0.0%us, 0.3%sy, 0.0%ni, 99.7%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
> Cpu6 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
> Cpu7 : 0.0%us, 0.0%sy, 0.0%ni,100.0%id, 0.0%wa, 0.0%hi, 0.0%si, 0.0%st
> Mem: 8048812k total, 7995856k used, 52956k free, 283692k buffers
> Swap: 4192956k total, 124k used, 4192832k free, 7492420k cached
> PID USER PR NI VIRT RES SHR S %CPU %MEM TIME+ COMMAND
> 4522 root 15 -5 0 0 0 S 5 0.0 17:59.16 nfsd
> 2012 root 15 -5 0 0 0 S 1 0.0 6:29.01 kjournald
> 4404 root 20 0 104m 17m 5044 S 0 0.2 55:07.93 X
> 4519 root 15 -5 0 0 0 S 0 0.0 18:05.58 nfsd
> 4521 root 15 -5 0 0 0 S 0 0.0 16:55.33 nfsd
> 5023 gdm 20 0 235m 31m 11m S 0 0.4 13:39.99 gdm-simple-gree
> 1 root 20 0 1064 408 348 S 0 0.0 0:02.90 init
> 2 root 15 -5 0 0 0 S 0 0.0 0:00.02 kthreadd
> 3 root RT -5 0 0 0 S 0 0.0 0:00.00 migration/0
> 4 root 15 -5 0 0 0 S 0 0.0 0:00.22 ksoftirqd/0
>
> The ifort ,MKL,and mpi I used is :
> INTFC=/opt/intel/Compiler/11.0/081
> INTMKL=/opt/intel/mkl/10.1.1.019
> /opt/mpich2/bin/mpd
> My machine model is as following :
> processor : 0
> vendor_id : GenuineIntel
> cpu family : 6
> model : 23
> model name : Intel(R) Xeon(R) CPU E5410 @ 2.33GHz
> stepping : 10
> cpu MHz : 2327.489
> cache size : 6144 KB
> physical id : 0
> siblings : 4
> core id : 0
> cpu cores : 4
> apicid : 0
> initial apicid : 0
> fpu : yes
> fpu_exception : yes
> cpuid level : 13
> wp : yes
> flags : fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov
> pat pse36 clflush dts acpi mmx fxsr sse sse2 ss ht tm pbe syscall nx 
> lm constant
> _tsc arch_perfmon pebs bts rep_good nopl pni monitor ds_cpl vmx est 
> tm2 ssse3 cx
> 16 xtpr dca sse4_1 lahf_lm
> bogomips : 4654.97
> ...............
> The pwscf version is espresso-4.0.3 .
> So , I want to know why the usage is so slow ?
> How can I solve this problem ? Anybody who meet the same problem ?
> Your kind help will be appreciated !
> Best regards
> Q.J.Wang
> XiangTan University
>
Some possibilities are:

1) 
http://www.quantum-espresso.org/wiki/index.php/Frequently_Asked_Questions#Why_is_my_parallel_job_running_in_such_a_lousy_way.3F
2) you are running a job over a very slow network (interconnecting the 
different nodes)
3) you are trying to force a node to write on a disk which is local to 
other nodes and connected to it
through a very-low performance network
4) your job requests much more resources than you have, e.g. RAM memory 
(causing a node to swap). From
your data it seems you are using all the 8Gb mem, actually the same test 
should be done on every node.


How are distributed the 16 CPU? Over more than one node?

Giovanni

-- 



Dr. Giovanni Cantele
Coherentia CNR-INFM and Dipartimento di Scienze Fisiche
Universita' di Napoli "Federico II"
Complesso Universitario di Monte S. Angelo - Ed. 6
Via Cintia, I-80126, Napoli, Italy
Phone: +39 081 676910
Fax:   +39 081 676346
E-mail: giovanni.cantele at cnr.it
        giovanni.cantele at na.infn.it
Web: http://people.na.infn.it/~cantele
Research Group: http://www.nanomat.unina.it
Skype contact: giocan74