System Information | ||||
Affiliation: | Army High Performance Computing Research Center (AHPCRC) | URL: | ||
Location: | Minneapolis Minnesota | System Use: | Government | |
System Manufacturer: | Cray Inc. | System Name: | X1 | |
Interconnect Manufacturer: | Cray | Interconnect Type: | Cray modified 2D torus | |
Operating System: | Unicos/MP 2.4 | MPI: | MPT.2.3.0.3 | |
MPI Wtick: | BLAS: | Cray libsci 5.2.0.0 | ||
Language: | C | Compiler: | ||
Compiler Flags: | Processor Type: | Cray X1 MSP | ||
Processor Speed: | 0.8 GHz | Total Processors: | 128 | |
Processors Entered: | 124 | Processors determined: | 124 | |
Cores per chip: | HPL Processes: | 124 | ||
MPI Processes: | 124 | Threads Entered: | 1 | |
Threads determined: | 1 | FLOPs per cycle: | 16 | |
Theoretical peak: | 1.5874 TFlop/s | Total memory: | GiB | |
FFT library: | ||||
Explain Optimizations: | ||||
STREAM: Aligned the data to cache line boundaries and added no_cache_alloc directives. Single cpu RandomAccess: Change vector length to 1024 and added concurrent directive. MPI RandomAccess: Changed distribution so all processors have equal number of elements except last cpu, this eliminated the need for a if test. Implemented "Extra Buckets" to vectorize. |
HPL | ||||
HPL: | 1.18203 Tflop/s | HPL time: | 4217.86 | |
HPL eps: | 1.110223e-16 | HPL Rnorm1: | ||
HPL Anorm1: | HPL AnormI: | |||
HPL Xnorm1: | HPL XnormI: | |||
HPL N: | 195555 | HPL NB: | 112 | |
HPL NProw: | 4 | HPL NPcol: | 31 | |
HPL depth: | 1 | HPL NBdiv: | 2 | |
HPL NBmin: | 4 | HPL CPfact: | R | |
HPL CRfact: | R | HPL CPtop: | 1 | |
HPL order: | R | |||
HPL dMach EPS: | HPL sMach EPS: | |||
HPL dMach sfMin: | HPL sMach sfMin: | |||
HPL dMach Base: | HPL sMach Base: | |||
HPL dMach Prec: | HPL sMach Prec: | |||
HPL dMach mLen: | HPL sMach mLen: | |||
HPL dMach Rnd: | HPL sMach Rnd: | |||
HPL dMach eMin: | HPL sMach eMin: | |||
HPL dMach rMin: | HPL sMach rMin: | |||
HPL dMach eMax: | HPL sMach eMax: | |||
HPL dMach rMax: | HPL sMach rMax: | |||
dweps: | sweps: |
PTRANS | ||||
PTRANS: | 39.3826 GB/s | PTRANS time: | 1.94 seconds | |
PTRANS residual: | 0 | PTRANS N: | 97777 | |
PTRANS NB: | 121 | PTRANS NProw: | 4 | |
PTRANS NPcol: | 31 |
STREAM | ||||
S-STREAM Copy: | 20.4881 GB/s | S-STREAM Scale: | 21.0179 GB/s | |
S-STREAM Add: | 23.8419 GB/s | S-STREAM Triad: | 24.0247 GB/s | |
EP-STREAM Copy: | 19.5068 GB/s | EP-STREAM Scale: | 19.3419 GB/s | |
EP-STREAM Add: | 21.1848 GB/s | EP-STREAM Triad: | 21.7521 GB/s | |
STREAM Vector Size: | 102800416 | STREAM Threads: | 1 |
RandomAccess | ||||
S-RandomAccess: | 0.208216 Gup/s | EP-RandomAccess: | 0.208677 Gup/s | |
G-RandomAccess: | Gup/s | G-RandomAccess N: | ||
G-RandomAccess time: | 90.268261 seconds | G-RandomAccess Check Time: | seconds | |
G-RandomAccess Errors: | G-RandomAccess Errors Fraction: | |||
G-RandomAccess TimeBound: | G-RandomAccess ExeUpdates: | |||
RandomAccess N: |
FFT | ||||
S-FFT: | GFlop/s | EP-FFT: | GFlop/s | |
MPIFFT: | GFlop/s | MPIFFT N: | ||
MPIFFT Max Error: | MPIFFT time0: | seconds | ||
MPIFFT time1: | seconds | MPIFFT time2: | seconds | |
MPIFFT time3: | seconds | MPIFFT time4: | seconds | |
MPIFFT time5: | seconds | MPIFFT time6: | seconds | |
FFTEnblk: | FFTEnp: | |||
FFTEl2size: |
DGEMM | ||||
S-DGEMM: | GFlop/s | EP-DGEMM: | GFlop/s | |
DGEMM N: |
RandomRing Latency/Bandwidth | ||||
RandomRing Latency: | 20.8481 usec | RandomRing Bandwidth: | 0.803881 GB/s |
NaturalRing Latency/Bandwidth | ||||
NaturalRing Latency: | 18.089 usec | NaturalRing Bandwidth: | 4.11851 GB/s |
PingPong Latency/Bandwidth | ||||
Maximum PingPong Latency: | 9.69063 usec | Maximum PingPong Bandwidth: | 9.103323 GB/s | |
Minimum PingPong Latency: | 8.109 usec | Minimum PingPong Bandwidth: | 4.99775 GB/s | |
Average PingPong Latency: | 8.981 usec | Average PingPong Bandwidth: | 8.502434 GB/s |
Size of Data Types | ||||
char: | byte | short: | bytes | |
int: | bytes | long: | bytes | |
void ptr: | bytes | float: | bytes | |
double: | bytes | size t: | bytes | |
s64Int: | bytes | u64Int: | bytes |
OpenMP | ||||
M OpenMP: | OpenMP Num Threads: | |||
OpenMP Num Procs: | OpenMP Max Threads: |
Memory | ||||
MemProc: | MemSpec: | |||
MemVal: |
CPS | ||||
CPS_HPCC_FFT_235: | CPS_HPCC_FFTW_ESTIMATE: | |||
CPS_HPCC_MEMALLCTR: | CPS_HPL_USE_GETPROCESSTIMES: | |||
CPS_RA_SANDIA_NOPT: | CPS_RA_SANDIA_OPT2: |