| System Information | ||||
| Affiliation: | IBM T.J. Watson Research Center | URL: | http://www.ibm.com | |
| Location: | USA, New York, Yorktown Heights | System Use: | Research | |
| System Manufacturer: | IBM | System Name: | Blue Gene/L | |
| Interconnect Manufacturer: | IBM | Interconnect Type: | Blue Gene Custom Interconnect | |
| Operating System: | BLRTS DRV 248 | MPI: | MPICH 1.1 | |
| MPI Wtick: | 0.000006 | BLAS: | ESSL 4.2.2 beta | |
| Language: | C | Compiler: | blrts_xlc 7.0 | |
| Compiler Flags: | -O3 -qarch=440d -qtune=440; FFT, STREAMS and link: -O5 -qarch=440d -qtune=440 | Processor Type: | IBM PowerPC 440 | |
| Processor Speed: | 0.7 GHz | Total Processors: | 40960 | |
| Processors Entered: | 32768 | Processors determined: | 16384 | |
| Cores per chip: | 2 | HPL Processes: | 16384 | |
| MPI Processes: | 16384 | Threads Entered: | 1 | |
| Threads determined: | 1 | FLOPs per cycle: | 4 | |
| Theoretical peak: | 91.7504 TFlop/s | Total memory: | GiB | |
| FFT library: | ||||
| Explain Optimizations: | ||||
| HPL: Custom Algorithm (mostly to take advantage of collectives). Special Assembly kernels and data format (replace factorization routines ... uses HPL's backsolve). FFT: Use of intrinsics (pseudo ASM instructions) to help the compiler. STREAMS: Assembly kernels MPIRandomAccess: Custom Algorithm (same as used on LLNL machine ... we will be following up on that front). | ||||
| HPL | ||||
| HPL: | 67.1174 Tflop/s | HPL time: | 4831.22 | |
| HPL eps: | 1.11022e-16 | HPL Rnorm1: | 0.0000000581115 | |
| HPL Anorm1: | 197102 | HPL AnormI: | 197165 | |
| HPL Xnorm1: | 370679 | HPL XnormI: | 3.1902 | |
| HPL N: | 786431 | HPL NB: | 96 | |
| HPL NProw: | 32 | HPL NPcol: | 512 | |
| HPL depth: | 12 | HPL NBdiv: | 6 | |
| HPL NBmin: | 6 | HPL CPfact: | L | |
| HPL CRfact: | R | HPL CPtop: | 1 | |
| HPL order: | R | |||
| HPL dMach EPS: | 1.110223e-16 | HPL sMach EPS: | 0.00000005960464 | |
| HPL dMach sfMin: | 0 | HPL sMach sfMin: | 1.175494e-38 | |
| HPL dMach Base: | 2 | HPL sMach Base: | 2 | |
| HPL dMach Prec: | 2.220446e-16 | HPL sMach Prec: | 0.0000001192093 | |
| HPL dMach mLen: | 53 | HPL sMach mLen: | 24 | |
| HPL dMach Rnd: | 1 | HPL sMach Rnd: | 1 | |
| HPL dMach eMin: | -1021 | HPL sMach eMin: | -125 | |
| HPL dMach rMin: | 0 | HPL sMach rMin: | 1.175494e-38 | |
| HPL dMach eMax: | 1024 | HPL sMach eMax: | 128 | |
| HPL dMach rMax: | 1.797693e308 | HPL sMach rMax: | 3.402823e38 | |
| dweps: | 1.110223e-16 | sweps: | 0.00000005960464 | |
| PTRANS | ||||
| PTRANS: | 137.238 GB/s | PTRANS time: | 9.01313 seconds | |
| PTRANS residual: | 0 | PTRANS N: | 393215 | |
| PTRANS NB: | 96 | PTRANS NProw: | 32 | |
| PTRANS NPcol: | 512 | |||
| STREAM | ||||
| S-STREAM Copy: | 1.64366 GB/s | S-STREAM Scale: | 1.30682 GB/s | |
| S-STREAM Add: | 2.07672 GB/s | S-STREAM Triad: | 2.44088 GB/s | |
| EP-STREAM Copy: | 1.64384 GB/s | EP-STREAM Scale: | 1.30689 GB/s | |
| EP-STREAM Add: | 2.06369 GB/s | EP-STREAM Triad: | 2.44044 GB/s | |
| STREAM Vector Size: | 12582880 | STREAM Threads: | 1 | |
| RandomAccess | ||||
| S-RandomAccess: | 0.00652124 Gup/s | EP-RandomAccess: | 0.00652118 Gup/s | |
| G-RandomAccess: | 17.2911 Gup/s | G-RandomAccess N: | 549755813888 | |
| G-RandomAccess time: | 3.08213 seconds | G-RandomAccess Check Time: | 155.903 seconds | |
| G-RandomAccess Errors: | 0 | G-RandomAccess Errors Fraction: | 0 | |
| G-RandomAccess TimeBound: | 1207.8 | G-RandomAccess ExeUpdates: | 53293432832 | |
| RandomAccess N: | 33554432 | |||
| FFT | ||||
| S-FFT: | 0.208379 GFlop/s | EP-FFT: | 0.209261 GFlop/s | |
| MPIFFT: | 988.181 GFlop/s | MPIFFT N: | 68719476736 | |
| MPIFFT Max Error: | 0.00000000000000202597 | MPIFFT time0: | 0.00000313571 seconds | |
| MPIFFT time1: | 2.61891 seconds | MPIFFT time2: | 1.94716 seconds | |
| MPIFFT time3: | 2.31841 seconds | MPIFFT time4: | 2.62041 seconds | |
| MPIFFT time5: | 2.88623 seconds | MPIFFT time6: | 0.000000558572 seconds | |
| FFTEnblk: | 16 | FFTEnp: | 4 | |
| FFTEl2size: | 1048576 | |||
| DGEMM | ||||
| S-DGEMM: | 2.31594 GFlop/s | EP-DGEMM: | 2.31468 GFlop/s | |
| DGEMM N: | 3071 | |||
| RandomRing Latency/Bandwidth | ||||
| RandomRing Latency: | 5.883 usec | RandomRing Bandwidth: | 0.0218582 GB/s | |
| NaturalRing Latency/Bandwidth | ||||
| NaturalRing Latency: | 3.43614 usec | NaturalRing Bandwidth: | 0.184978 GB/s | |
| PingPong Latency/Bandwidth | ||||
| Maximum PingPong Latency: | 6.74327 usec | Maximum PingPong Bandwidth: | 0.15897 GB/s | |
| Minimum PingPong Latency: | 2.90095 usec | Minimum PingPong Bandwidth: | 0.151557 GB/s | |
| Average PingPong Latency: | 4.69686 usec | Average PingPong Bandwidth: | 0.157671 GB/s | |
| Size of Data Types | ||||
| char: | 1 byte | short: | 2 bytes | |
| int: | 4 bytes | long: | 4 bytes | |
| void ptr: | 4 bytes | float: | 4 bytes | |
| double: | 8 bytes | size t: | 4 bytes | |
| s64Int: | 8 bytes | u64Int: | 8 bytes | |
| OpenMP | ||||
| M OpenMP: | -1 | OpenMP Num Threads: | 0 | |
| OpenMP Num Procs: | 0 | OpenMP Max Threads: | 0 | |
| Memory | ||||
| MemProc: | -1 | MemSpec: | -1 | |
| MemVal: | -1 | |||
| CPS | ||||
| CPS_HPCC_FFT_235: | CPS_HPCC_FFTW_ESTIMATE: | |||
| CPS_HPCC_MEMALLCTR: | CPS_HPL_USE_GETPROCESSTIMES: | |||
| CPS_RA_SANDIA_NOPT: | CPS_RA_SANDIA_OPT2: | |||