Alpha software: FOSSbench scores have not yet been tested to be 100% accurate.
FOSSbench

Individual benchmark run

POWER10 (architected), altivec supported

IBM,9105-22A · tested Sep 16, 2026, 12:48 PM · uploaded by @l_28

In Review · not yet included in the processor average

C score443,781ASM score: 414,935

System information

Processor
POWER10 (architected), altivec supported
Computer model
IBM,9105-22A
Architecture
PowerPC 64-bit big-endian
Physical cores reported by run
16
Hardware threads reported by run
128
L1 cache
80 KB
L2 cache
80 KB
L3 cache
80 KB
Memory
63,023 MB
Operating system
Red Hat Enterprise Linux 10.2 (Coughlan)
Kernel
Linux 6.12.0-211.7.3.el10_2.ppc64le
Compiler
Clang 21.1.8 (Red Hat, Inc. 21.1.8-1.el10)
FOSSbench version
0.4.2
Total test time
1 min 19 sec

Score summary

C multi-core443,781
C single-core9,704
ASM multi-core414,935
ASM single-core9,736

Test environment

Background monitoring
Available
Average background CPU use
0%
Peak background CPU use
0.1%
Available memory before test
58,904 MB
Running processes
1,689
Activity sample duration
10 seconds

Individual workload results

WorkloadRunMeasured performanceRaw throughputScoreElapsedIterationsThreadsChecksum
Native Integer Mathcommon 32-bit multiply/add/rotate mix10% of overall scoreMulti-core398,968.15 Mops/s398,968.15 million work units/sec1,292,8330.49 seconds959128219898061312
Single-core4,159.08 Mops/s4,159.08 million work units/sec13,4770.5 seconds1,2891934523089
Wide Integer Math64-bit PowerPC integer ALU10% of overall scoreMulti-core362,954.11 Mops/s362,954.11 million work units/sec1,176,1310.5 seconds5941288508060607444462464
Single-core3,429.08 Mops/s3,429.08 million work units/sec11,1120.52 seconds7371327040404707099736
Floating Point MathPowerPC scalar double-precision floating point10% of overall scoreMulti-core60,699.77 Mops/s60,699.77 million work units/sec360,8790.5 seconds1191281389379145371492096
Single-core697.2 Mops/s697.2 million work units/sec4,1450.46 seconds16114622540543000602686
Prime Numberssieve of Eratosthenes to 2M10% of overall scoreMulti-core18,857.8 Mcand/s18,857.8 million work units/sec232,2390.53 seconds39128743473536
Single-core924.88 Mcand/s924.88 million work units/sec11,3900.51 seconds236135148188
Extended InstructionsAltiVec: 128-bit integer vectors (PowerPC 970)10% of overall scoreMulti-core558,126.13 Mops/s558,126.13 million work units/sec848,7320.48 seconds649128361170782464
Single-core6,214.94 Mops/s6,214.94 million work units/sec9,4510.52 seconds1,00913527819322
CompressionLZ77 match finder, 4 MiB corpus10% of overall scoreMulti-core33,665.7 MB/s33,665.7 million work units/sec499,4910.51 seconds321286881189888
Single-core529.5 MB/s529.5 million work units/sec7,8560.52 seconds661110878548
EncryptionChaCha20, 20 rounds, 1 MiB10% of overall scoreMulti-core44,326.59 MB/s44,326.59 million work units/sec1,091,7880.5 seconds164128138988955776
Single-core503.14 MB/s503.14 million work units/sec12,3930.52 seconds24913631195130
Physics512-body direct-sum gravity10% of overall scoreMulti-core7,234.29 Mpair/s7,234.29 million work units/sec114,6480.51 seconds1111283084528844328101504
Single-core200.08 Mpair/s200.08 million work units/sec3,1710.52 seconds39814635494768022377952
Sortingheapsort, 1M uint3210% of overall scoreMulti-core5,060.9 Mkey-cmp/s5,060.9 million work units/sec139,4190.53 seconds112810637378391587277440
Single-core217.45 Mkey-cmp/s217.45 million work units/sec5,9900.58 seconds610
Memory Latencydependent-load pointer chase, 16 MiB0% of overall scoreMulti-core241.33 ns/access530.39 million work units/sec67,1380.72 seconds31281121112992
Single-core32.54 ns/access30.73 million work units/sec3,8900.55 seconds1713261560
Memory BandwidthSTREAM triad, 12 MiB working set per thread10% of overall scoreMulti-core100,559.48 MB/s100,559.48 million work units/sec314,2480.51 seconds32128136432615424
Single-core19,173.65 MB/s19,173.65 million work units/sec59,9180.57 seconds86411060120064