Skip to content

Commit

Permalink
emptydir results for a100 and v100 nodes
Browse files Browse the repository at this point in the history
  • Loading branch information
knikolla committed Nov 12, 2024
1 parent 7a8f6c0 commit bbbf9ce
Show file tree
Hide file tree
Showing 2 changed files with 260 additions and 0 deletions.
130 changes: 130 additions & 0 deletions results/ocp_a100ssd_unet3d_1000_a100_20241112/summary.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,130 @@
{
"start": "2024-11-12T17:15:12.470362",
"num_accelerators": 1,
"num_hosts": 1,
"hostname": "mlperf-storage-job-a100-nogpu-k6zf9",
"metric": {
"train_au_percentage": [
33.50913131343001,
33.45152654562511,
33.71637729910411,
33.615818542127265,
33.969711027890995
],
"train_au_mean_percentage": 33.6525129456355,
"train_au_meet_expectation": "fail",
"train_au_stdev_percentage": 0.1827531579603483,
"train_throughput_samples_per_second": [
3.5971160947687384,
3.603389443383753,
3.628881848004376,
3.63031642862578,
3.651156203327297
],
"train_throughput_mean_samples_per_second": 3.6221720036219884,
"train_throughput_stdev_samples_per_second": 0.019657164802690953,
"train_io_mean_MB_per_second": 506.41316457271745,
"train_io_stdev_MB_per_second": 2.748253540777197
},
"num_files_train": 1000,
"num_files_eval": 0,
"num_samples_per_file": 1,
"host_cpu_count": [
128
],
"host_processor_name": "x86_64",
"potential_caching": [
1
],
"host_cpuinfo": {
"vendor_id": "GenuineIntel",
"cpu family": "6",
"model": "106",
"model name": "Intel(R) Xeon(R) Platinum 8358 CPU @ 2.60GHz",
"stepping": "6",
"microcode": "0xd0003b9",
"cpu MHz": "2043.222",
"cache size": "49152 KB",
"physical id": "1",
"siblings": "64",
"core id": "31",
"cpu cores": "32",
"apicid": "191",
"initial apicid": "191",
"fpu": "yes",
"fpu_exception": "yes",
"cpuid level": "27",
"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 pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm rdt_a avx512f avx512dq rdseed adx smap avx512ifma clflushopt clwb intel_pt avx512cd sha_ni avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local split_lock_detect wbnoinvd dtherm ida arat pln pts hwp_epp avx512vbmi umip pku ospke avx512_vbmi2 gfni vaes vpclmulqdq avx512_vnni avx512_bitalg tme avx512_vpopcntdq la57 rdpid fsrm md_clear pconfig flush_l1d arch_capabilities",
"vmx flags": "vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple shadow_vmcs pml ept_mode_based_exec tsc_scaling",
"bugs": "spectre_v1 spectre_v2 spec_store_bypass swapgs mmio_stale_data eibrs_pbrsb gds",
"bogomips": "5217.03",
"clflush size": "64",
"cache_alignment": "64",
"address sizes": "46 bits physical, 57 bits virtual",
"power management": ""
},
"host_meminfo": {
"MemTotal": "1056470016 kB",
"MemFree": "924607404 kB",
"MemAvailable": "1009560332 kB",
"Buffers": "1038460 kB",
"Cached": "84482968 kB",
"SwapCached": "0 kB",
"Active": "12892928 kB",
"Inactive": "103708760 kB",
"Active(anon)": "154500 kB",
"Inactive(anon)": "31194160 kB",
"Active(file)": "12738428 kB",
"Inactive(file)": "72514600 kB",
"Unevictable": "549060 kB",
"Mlocked": "545988 kB",
"SwapTotal": "0 kB",
"SwapFree": "0 kB",
"Zswap": "0 kB",
"Zswapped": "0 kB",
"Dirty": "2216 kB",
"Writeback": "0 kB",
"AnonPages": "30913604 kB",
"Mapped": "2215076 kB",
"Shmem": "233732 kB",
"KReclaimable": "6382736 kB",
"Slab": "9681220 kB",
"SReclaimable": "6382736 kB",
"SUnreclaim": "3298484 kB",
"KernelStack": "101824 kB",
"PageTables": "118148 kB",
"NFS_Unstable": "0 kB",
"Bounce": "0 kB",
"WritebackTmp": "0 kB",
"CommitLimit": "528235008 kB",
"Committed_AS": "74475180 kB",
"VmallocTotal": "13743895347199 kB",
"VmallocUsed": "743076 kB",
"VmallocChunk": "0 kB",
"Percpu": "399872 kB",
"HardwareCorrupted": "0 kB",
"AnonHugePages": "26419200 kB",
"ShmemHugePages": "0 kB",
"ShmemPmdMapped": "0 kB",
"FileHugePages": "0 kB",
"FilePmdMapped": "0 kB",
"CmaTotal": "0 kB",
"CmaFree": "0 kB",
"HugePages_Total": "0",
"HugePages_Free": "0",
"HugePages_Rsvd": "0",
"HugePages_Surp": "0",
"Hugepagesize": "2048 kB",
"Hugetlb": "0 kB",
"DirectMap4k": "12478760 kB",
"DirectMap2M": "387721216 kB",
"DirectMap1G": "675282944 kB"
},
"host_memory_GB": [
1007.5283203125
],
"data_size_per_host_GB": 136.532474309206,
"epochs": 5,
"end": "2024-11-12T17:38:08.085114"
}
130 changes: 130 additions & 0 deletions results/ocp_v100ssd_unet3d_1000_a100_20241112/summary.json
Original file line number Diff line number Diff line change
@@ -0,0 +1,130 @@
{
"start": "2024-11-12T16:04:10.261455",
"num_accelerators": 1,
"num_hosts": 1,
"hostname": "mlperf-storage-job-v100-qvj8f",
"metric": {
"train_au_percentage": [
46.99785217967899,
47.2780389861644,
47.372152964926094,
47.44031344355234,
48.17845300658986
],
"train_au_mean_percentage": 47.45336211618233,
"train_au_meet_expectation": "fail",
"train_au_stdev_percentage": 0.39261928012508746,
"train_throughput_samples_per_second": [
5.050948204863154,
5.11519439152014,
5.136124306379946,
5.136554756987737,
5.192166059948317
],
"train_throughput_mean_samples_per_second": 5.126197543939859,
"train_throughput_stdev_samples_per_second": 0.04547788467743791,
"train_io_mean_MB_per_second": 716.689852899209,
"train_io_stdev_MB_per_second": 6.358229116271948
},
"num_files_train": 1000,
"num_files_eval": 0,
"num_samples_per_file": 1,
"host_cpu_count": [
80
],
"host_processor_name": "x86_64",
"potential_caching": [
1
],
"host_cpuinfo": {
"vendor_id": "GenuineIntel",
"cpu family": "6",
"model": "85",
"model name": "Intel(R) Xeon(R) Gold 6248 CPU @ 2.50GHz",
"stepping": "7",
"microcode": "0x5003604",
"cpu MHz": "2595.232",
"cache size": "28160 KB",
"physical id": "1",
"siblings": "40",
"core id": "26",
"cpu cores": "20",
"apicid": "117",
"initial apicid": "117",
"fpu": "yes",
"fpu_exception": "yes",
"cpuid level": "22",
"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 pdpe1gb rdtscp lm constant_tsc art arch_perfmon pebs bts rep_good nopl xtopology nonstop_tsc cpuid aperfmperf pni pclmulqdq dtes64 monitor ds_cpl vmx smx est tm2 ssse3 sdbg fma cx16 xtpr pdcm pcid dca sse4_1 sse4_2 x2apic movbe popcnt tsc_deadline_timer aes xsave avx f16c rdrand lahf_lm abm 3dnowprefetch cpuid_fault epb cat_l3 cdp_l3 invpcid_single intel_ppin ssbd mba ibrs ibpb stibp ibrs_enhanced tpr_shadow vnmi flexpriority ept vpid ept_ad fsgsbase tsc_adjust bmi1 avx2 smep bmi2 erms invpcid cqm mpx rdt_a avx512f avx512dq rdseed adx smap clflushopt clwb intel_pt avx512cd avx512bw avx512vl xsaveopt xsavec xgetbv1 xsaves cqm_llc cqm_occup_llc cqm_mbm_total cqm_mbm_local dtherm ida arat pln pts pku ospke avx512_vnni md_clear flush_l1d arch_capabilities",
"vmx flags": "vnmi preemption_timer posted_intr invvpid ept_x_only ept_ad ept_1gb flexpriority apicv tsc_offset vtpr mtf vapic ept vpid unrestricted_guest vapic_reg vid ple shadow_vmcs pml ept_mode_based_exec tsc_scaling",
"bugs": "spectre_v1 spectre_v2 spec_store_bypass swapgs taa itlb_multihit mmio_stale_data retbleed eibrs_pbrsb gds",
"bogomips": "5006.26",
"clflush size": "64",
"cache_alignment": "64",
"address sizes": "46 bits physical, 48 bits virtual",
"power management": ""
},
"host_meminfo": {
"MemTotal": "791175160 kB",
"MemFree": "553388300 kB",
"MemAvailable": "761922088 kB",
"Buffers": "59764 kB",
"Cached": "207624136 kB",
"SwapCached": "0 kB",
"Active": "18274652 kB",
"Inactive": "205697520 kB",
"Active(anon)": "139764 kB",
"Inactive(anon)": "16401608 kB",
"Active(file)": "18134888 kB",
"Inactive(file)": "189295912 kB",
"Unevictable": "415092 kB",
"Mlocked": "412020 kB",
"SwapTotal": "0 kB",
"SwapFree": "0 kB",
"Zswap": "0 kB",
"Zswapped": "0 kB",
"Dirty": "200692 kB",
"Writeback": "0 kB",
"AnonPages": "15945016 kB",
"Mapped": "1793760 kB",
"Shmem": "218272 kB",
"KReclaimable": "6149144 kB",
"Slab": "9775672 kB",
"SReclaimable": "6149144 kB",
"SUnreclaim": "3626528 kB",
"KernelStack": "64256 kB",
"PageTables": "72128 kB",
"NFS_Unstable": "0 kB",
"Bounce": "0 kB",
"WritebackTmp": "0 kB",
"CommitLimit": "395587580 kB",
"Committed_AS": "38635264 kB",
"VmallocTotal": "34359738367 kB",
"VmallocUsed": "620448 kB",
"VmallocChunk": "0 kB",
"Percpu": "156160 kB",
"HardwareCorrupted": "0 kB",
"AnonHugePages": "13164544 kB",
"ShmemHugePages": "0 kB",
"ShmemPmdMapped": "0 kB",
"FileHugePages": "0 kB",
"FilePmdMapped": "0 kB",
"CmaTotal": "0 kB",
"CmaFree": "0 kB",
"HugePages_Total": "0",
"HugePages_Free": "0",
"HugePages_Rsvd": "0",
"HugePages_Surp": "0",
"Hugepagesize": "2048 kB",
"Hugetlb": "0 kB",
"DirectMap4k": "6951936 kB",
"DirectMap2M": "283113472 kB",
"DirectMap1G": "515899392 kB"
},
"host_memory_GB": [
754.5234298706055
],
"data_size_per_host_GB": 136.532474309206,
"epochs": 5,
"end": "2024-11-12T16:20:26.697187"
}

0 comments on commit bbbf9ce

Please sign in to comment.