Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom

Improve type stability and inference for bilinear assemble #61

Closed
wants to merge 1 commit into from
Closed

Improve type stability and inference for bilinear assemble #61

wants to merge 1 commit into from

Conversation

ghislainb
Copy link
Member

No description provided.

@github-actions GitHub Actions
Copy link

Benchmark result

Judge result

Benchmark Report for /home/runner/work/Bcube.jl/Bcube.jl

Job Properties

  • Time of benchmarks:
    • Target: 16 Mar 2024 - 22:05
    • Baseline: 16 Mar 2024 - 22:09
  • Package commits:
    • Target: c73e35
    • Baseline: 67cc67
  • Julia commits:
    • Target: 17cfb8
    • Baseline: 17cfb8
  • Julia command flags:
    • Target: None
    • Baseline: None
  • Environment variables:
    • Target: None
    • Baseline: None

Results

A ratio greater than 1.0 denotes a possible regression (marked with ❌), while a ratio less
than 1.0 denotes a possible improvement (marked with ✅). Only significant results - results
that indicate possible regressions or improvements - are shown below (thus, an empty table means that all
benchmark results remained invariant between builds).

ID time ratio memory ratio
["assemble", "basic bilinear", "mass matrix"] 0.21 (5%) ✅ 0.10 (1%) ✅
["assemble", "basic bilinear", "stiffness matrix"] 0.35 (5%) ✅ 0.04 (1%) ✅
["assemble", "poisson DG", "AffineFESystem"] 0.73 (5%) ✅ 0.53 (1%) ✅
["assemble", "poisson DG", "a_Γ(u, v)"] 0.73 (5%) ✅ 0.60 (1%) ✅
["assemble", "poisson DG", "a_Γb(u, v)"] 0.56 (5%) ✅ 0.56 (1%) ✅
["assemble", "poisson DG", "a_Ω(u, v)"] 0.74 (5%) ✅ 0.02 (1%) ✅
["assemble", "poisson DG", "l_Γb(v)"] 0.95 (5%) ✅ 1.00 (1%)
["covo", "TriQuad", "explicit_step"] 0.93 (5%) ✅ 1.00 (1%)
["entities", "nnodes"] 1.22 (5%) ❌ 1.00 (1%)

Benchmark Group List

Here's a list of all the benchmark groups executed by this job:

  • ["assemble", "basic bilinear"]
  • ["assemble", "poisson DG"]
  • ["covo", "Quad"]
  • ["covo", "TriQuad"]
  • ["entities"]
  • ["mesh"]

Julia versioninfo

Target

Julia Version 1.8.5
Commit 17cfb8e65ea (2023-01-08 06:45 UTC)
Platform Info:
  OS: Linux (x86_64-linux-gnu)
      Ubuntu 22.04.4 LTS
  uname: Linux 6.5.0-1016-azure #16~22.04.1-Ubuntu SMP Fri Feb 16 15:42:02 UTC 2024 x86_64 x86_64
  CPU: AMD EPYC 7763 64-Core Processor: 
              speed         user         nice          sys         idle          irq
       #1  3243 MHz       2250 s          0 s        241 s       3951 s          0 s
       #2  2445 MHz       2653 s          0 s        285 s       3501 s          0 s
       #3  3208 MHz       1551 s          0 s        203 s       4695 s          0 s
       #4  3243 MHz       1407 s          0 s        188 s       4847 s          0 s
  Memory: 15.606498718261719 GB (11944.40234375 MB free)
  Uptime: 647.98 sec
  Load Avg:  1.01  1.13  0.75
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, znver3)
  Threads: 1 on 4 virtual cores

Baseline

Julia Version 1.8.5
Commit 17cfb8e65ea (2023-01-08 06:45 UTC)
Platform Info:
  OS: Linux (x86_64-linux-gnu)
      Ubuntu 22.04.4 LTS
  uname: Linux 6.5.0-1016-azure #16~22.04.1-Ubuntu SMP Fri Feb 16 15:42:02 UTC 2024 x86_64 x86_64
  CPU: AMD EPYC 7763 64-Core Processor: 
              speed         user         nice          sys         idle          irq
       #1  3244 MHz       2852 s          0 s        316 s       5758 s          0 s
       #2  3268 MHz       3051 s          0 s        354 s       5517 s          0 s
       #3  2906 MHz       1997 s          0 s        247 s       6688 s          0 s
       #4  2597 MHz       2258 s          0 s        284 s       6384 s          0 s
  Memory: 15.606498718261719 GB (12240.140625 MB free)
  Uptime: 896.73 sec
  Load Avg:  1.02  1.06  0.83
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, znver3)
  Threads: 1 on 4 virtual cores

Target result

Benchmark Report for /home/runner/work/Bcube.jl/Bcube.jl

Job Properties

  • Time of benchmark: 16 Mar 2024 - 22:5
  • Package commit: c73e35
  • Julia commit: 17cfb8
  • Julia command flags: None
  • Environment variables: None

Results

Below is a table of this job's results, obtained by running the benchmarks.
The values listed in the ID column have the structure [parent_group, child_group, ..., key], and can be used to
index into the BaseBenchmarks suite to retrieve the corresponding benchmarks.
The percentages accompanying time and memory values in the below table are noise tolerances. The "true"
time/memory value for a given benchmark is expected to fall within this percentage of the reported value.
An empty cell means that the value was zero.

ID time GC time memory allocations
["assemble", "basic bilinear", "mass matrix"] 14.299 ms (5%) 13.82 MiB (1%) 117
["assemble", "basic bilinear", "stiffness matrix"] 74.180 ms (5%) 13.82 MiB (1%) 117
["assemble", "poisson DG", "AffineFESystem"] 52.156 ms (5%) 86.88 MiB (1%) 20157
["assemble", "poisson DG", "a_Γ(u, v)"] 36.117 ms (5%) 71.48 MiB (1%) 15409
["assemble", "poisson DG", "a_Γb(u, v)"] 7.580 ms (5%) 13.85 MiB (1%) 3715
["assemble", "poisson DG", "a_Ω(u, v)"] 7.639 ms (5%) 349.81 KiB (1%) 69
["assemble", "poisson DG", "l_Γb(v)"] 90.138 μs (5%) 192.66 KiB (1%) 1077
["assemble", "poisson DG", "l_Ω(v)"] 20.067 μs (5%) 22.61 KiB (1%) 21
["covo", "Quad", "explicit_step"] 100.261 ms (5%) 69.28 MiB (1%) 3474
["covo", "Quad", "integral_surface"] 48.595 ms (5%)
["covo", "Quad", "integral_volume"] 22.544 ms (5%)
["covo", "TriQuad", "explicit_step"] 354.442 ms (5%) 364.68 MiB (1%) 1634486
["covo", "TriQuad", "integral_surface"] 233.000 ms (5%) 210.06 MiB (1%) 1246116
["covo", "TriQuad", "integral_volume"] 57.942 ms (5%) 48.39 MiB (1%) 379526
["entities", "edges2nodes"] 1.552 ns (5%)
["entities", "nedges"] 1.552 ns (5%)
["entities", "nnodes"] 3.396 ns (5%)
["entities", "nodes"] 1.552 ns (5%)
["mesh", "todo"] 1.552 ns (5%)

Benchmark Group List

Here's a list of all the benchmark groups executed by this job:

  • ["assemble", "basic bilinear"]
  • ["assemble", "poisson DG"]
  • ["covo", "Quad"]
  • ["covo", "TriQuad"]
  • ["entities"]
  • ["mesh"]

Julia versioninfo

Julia Version 1.8.5
Commit 17cfb8e65ea (2023-01-08 06:45 UTC)
Platform Info:
  OS: Linux (x86_64-linux-gnu)
      Ubuntu 22.04.4 LTS
  uname: Linux 6.5.0-1016-azure #16~22.04.1-Ubuntu SMP Fri Feb 16 15:42:02 UTC 2024 x86_64 x86_64
  CPU: AMD EPYC 7763 64-Core Processor: 
              speed         user         nice          sys         idle          irq
       #1  3243 MHz       2250 s          0 s        241 s       3951 s          0 s
       #2  2445 MHz       2653 s          0 s        285 s       3501 s          0 s
       #3  3208 MHz       1551 s          0 s        203 s       4695 s          0 s
       #4  3243 MHz       1407 s          0 s        188 s       4847 s          0 s
  Memory: 15.606498718261719 GB (11944.40234375 MB free)
  Uptime: 647.98 sec
  Load Avg:  1.01  1.13  0.75
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, znver3)
  Threads: 1 on 4 virtual cores

Baseline result

Benchmark Report for /home/runner/work/Bcube.jl/Bcube.jl

Job Properties

  • Time of benchmark: 16 Mar 2024 - 22:9
  • Package commit: 67cc67
  • Julia commit: 17cfb8
  • Julia command flags: None
  • Environment variables: None

Results

Below is a table of this job's results, obtained by running the benchmarks.
The values listed in the ID column have the structure [parent_group, child_group, ..., key], and can be used to
index into the BaseBenchmarks suite to retrieve the corresponding benchmarks.
The percentages accompanying time and memory values in the below table are noise tolerances. The "true"
time/memory value for a given benchmark is expected to fall within this percentage of the reported value.
An empty cell means that the value was zero.

ID time GC time memory allocations
["assemble", "basic bilinear", "mass matrix"] 67.413 ms (5%) 140.16 MiB (1%) 245117
["assemble", "basic bilinear", "stiffness matrix"] 213.405 ms (5%) 31.356 ms 329.22 MiB (1%) 1255117
["assemble", "poisson DG", "AffineFESystem"] 71.915 ms (5%) 162.46 MiB (1%) 90509
["assemble", "poisson DG", "a_Γ(u, v)"] 49.665 ms (5%) 119.74 MiB (1%) 34009
["assemble", "poisson DG", "a_Γb(u, v)"] 13.420 ms (5%) 24.94 MiB (1%) 7459
["assemble", "poisson DG", "a_Ω(u, v)"] 10.256 ms (5%) 16.57 MiB (1%) 48101
["assemble", "poisson DG", "l_Γb(v)"] 95.327 μs (5%) 192.66 KiB (1%) 1077
["assemble", "poisson DG", "l_Ω(v)"] 20.257 μs (5%) 22.61 KiB (1%) 21
["covo", "Quad", "explicit_step"] 100.810 ms (5%) 69.28 MiB (1%) 3474
["covo", "Quad", "integral_surface"] 51.141 ms (5%)
["covo", "Quad", "integral_volume"] 22.354 ms (5%)
["covo", "TriQuad", "explicit_step"] 379.737 ms (5%) 33.730 ms 364.68 MiB (1%) 1634488
["covo", "TriQuad", "integral_surface"] 232.537 ms (5%) 210.06 MiB (1%) 1246116
["covo", "TriQuad", "integral_volume"] 57.036 ms (5%) 48.39 MiB (1%) 379526
["entities", "edges2nodes"] 1.552 ns (5%)
["entities", "nedges"] 1.552 ns (5%)
["entities", "nnodes"] 2.785 ns (5%)
["entities", "nodes"] 1.552 ns (5%)
["mesh", "todo"] 1.552 ns (5%)

Benchmark Group List

Here's a list of all the benchmark groups executed by this job:

  • ["assemble", "basic bilinear"]
  • ["assemble", "poisson DG"]
  • ["covo", "Quad"]
  • ["covo", "TriQuad"]
  • ["entities"]
  • ["mesh"]

Julia versioninfo

Julia Version 1.8.5
Commit 17cfb8e65ea (2023-01-08 06:45 UTC)
Platform Info:
  OS: Linux (x86_64-linux-gnu)
      Ubuntu 22.04.4 LTS
  uname: Linux 6.5.0-1016-azure #16~22.04.1-Ubuntu SMP Fri Feb 16 15:42:02 UTC 2024 x86_64 x86_64
  CPU: AMD EPYC 7763 64-Core Processor: 
              speed         user         nice          sys         idle          irq
       #1  3244 MHz       2852 s          0 s        316 s       5758 s          0 s
       #2  3268 MHz       3051 s          0 s        354 s       5517 s          0 s
       #3  2906 MHz       1997 s          0 s        247 s       6688 s          0 s
       #4  2597 MHz       2258 s          0 s        284 s       6384 s          0 s
  Memory: 15.606498718261719 GB (12240.140625 MB free)
  Uptime: 896.73 sec
  Load Avg:  1.02  1.06  0.83
  WORD_SIZE: 64
  LIBM: libopenlibm
  LLVM: libLLVM-13.0.1 (ORCJIT, znver3)
  Threads: 1 on 4 virtual cores

Runtime information

Runtime Info
BLAS #threads 2
BLAS.vendor() openblas64
Sys.CPU_THREADS 4

lscpu output:

Architecture:                       x86_64
CPU op-mode(s):                     32-bit, 64-bit
Address sizes:                      48 bits physical, 48 bits virtual
Byte Order:                         Little Endian
CPU(s):                             4
On-line CPU(s) list:                0-3
Vendor ID:                          AuthenticAMD
Model name:                         AMD EPYC 7763 64-Core Processor
CPU family:                         25
Model:                              1
Thread(s) per core:                 2
Core(s) per socket:                 2
Socket(s):                          1
Stepping:                           1
BogoMIPS:                           4890.86
Flags:                              fpu vme de pse tsc msr pae mce cx8 apic sep mtrr pge mca cmov pat pse36 clflush mmx fxsr sse sse2 ht syscall nx mmxext fxsr_opt pdpe1gb rdtscp lm constant_tsc rep_good nopl tsc_reliable nonstop_tsc cpuid extd_apicid aperfmperf pni pclmulqdq ssse3 fma cx16 pcid sse4_1 sse4_2 movbe popcnt aes xsave avx f16c rdrand hypervisor lahf_lm cmp_legacy svm cr8_legacy abm sse4a misalignsse 3dnowprefetch osvw topoext invpcid_single vmmcall fsgsbase bmi1 avx2 smep bmi2 erms invpcid rdseed adx smap clflushopt clwb sha_ni xsaveopt xsavec xgetbv1 xsaves clzero xsaveerptr rdpru arat npt nrip_save tsc_scale vmcb_clean flushbyasid decodeassists pausefilter pfthreshold v_vmsave_vmload umip vaes vpclmulqdq rdpid fsrm
Virtualization:                     AMD-V
Hypervisor vendor:                  Microsoft
Virtualization type:                full
L1d cache:                          64 KiB (2 instances)
L1i cache:                          64 KiB (2 instances)
L2 cache:                           1 MiB (2 instances)
L3 cache:                           32 MiB (1 instance)
NUMA node(s):                       1
NUMA node0 CPU(s):                  0-3
Vulnerability Gather data sampling: Not affected
Vulnerability Itlb multihit:        Not affected
Vulnerability L1tf:                 Not affected
Vulnerability Mds:                  Not affected
Vulnerability Meltdown:             Not affected
Vulnerability Mmio stale data:      Not affected
Vulnerability Retbleed:             Not affected
Vulnerability Spec rstack overflow: Vulnerable: Safe RET, no microcode
Vulnerability Spec store bypass:    Vulnerable
Vulnerability Spectre v1:           Mitigation; usercopy/swapgs barriers and __user pointer sanitization
Vulnerability Spectre v2:           Mitigation; Retpolines, STIBP disabled, RSB filling, PBRSB-eIBRS Not affected
Vulnerability Srbds:                Not affected
Vulnerability Tsx async abort:      Not affected
Cpu Property Value
Brand AMD EPYC 7763 64-Core Processor
Vendor :AMD
Architecture :Unknown
Model Family: 0xaf, Model: 0x01, Stepping: 0x01, Type: 0x00
Cores 16 physical cores, 16 logical cores (on executing CPU)
No Hyperthreading hardware capability detected
Clock Frequencies Not supported by CPU
Data Cache Level 1:3 : (32, 512, 32768) kbytes
64 byte cache line size
Address Size 48 bits virtual, 48 bits physical
SIMD 256 bit = 32 byte max. SIMD vector size
Time Stamp Counter TSC is accessible via rdtsc
TSC runs at constant rate (invariant from clock frequency)
Perf. Monitoring Performance Monitoring Counters (PMC) are not supported
Hypervisor Yes, Microsoft

@github-actions GitHub Actions
Copy link

Benchmark Results

main e4c0a49... main/e4c0a496d3f365...
assemble/basic bilinear/mass matrix 0.0894 ± 0.024 s 14.4 ± 0.42 ms 6.22
assemble/basic bilinear/stiffness matrix 0.217 ± 0.0043 s 0.0735 ± 0.00071 s 2.95
assemble/poisson DG/AffineFESystem 0.0818 ± 0.012 s 0.0534 ± 0.013 s 1.53
assemble/poisson DG/a_Γ(u, v) 0.0471 ± 0.012 s 0.0366 ± 0.0013 s 1.29
assemble/poisson DG/a_Γb(u, v) 13.1 ± 0.47 ms 7.82 ± 0.19 ms 1.67
assemble/poisson DG/a_Ω(u, v) 13.2 ± 2.8 ms 7.64 ± 0.011 ms 1.73
assemble/poisson DG/l_Γb(v) 0.166 ± 0.0054 ms 0.163 ± 0.0067 ms 1.02
assemble/poisson DG/l_Ω(v) 22.8 ± 2.6 μs 22.7 ± 2.5 μs 1.01
covo/Quad/explicit_step 0.103 ± 0.0047 s 0.1 ± 0.0018 s 1.03
covo/Quad/integral_surface 0.0507 ± 0.0008 s 0.0501 ± 0.00018 s 1.01
covo/Quad/integral_volume 22 ± 0.044 ms 22.3 ± 0.033 ms 0.988
covo/TriQuad/explicit_step 0.375 ± 0.017 s 0.39 ± 0.0061 s 0.962
covo/TriQuad/integral_surface 0.271 ± 0.024 s 0.288 ± 0.046 s 0.94
covo/TriQuad/integral_volume 0.0694 ± 0.0033 s 0.0703 ± 0.0033 s 0.987
entities/edges2nodes 1.56 ± 0.01 ns 1.55 ± 0.01 ns 1.01
entities/nedges 1.56 ± 0.01 ns 1.56 ± 0.01 ns 1
entities/nnodes 2.79 ± 0 ns 3.11 ± 0.01 ns 0.9
entities/nodes 1.56 ± 0.01 ns 1.55 ± 0.01 ns 1.01
mesh/todo 1.56 ± 0.01 ns 1.55 ± 0.01 ns 1.01
time_to_load 14.5 ± 0.071 s 14.4 ± 0.022 s 1

Benchmark Plots

A plot of the benchmark results have been uploaded as an artifact to the workflow run for this PR.
Go to "Actions"->"Benchmark a pull request"->[the most recent run]->"Artifacts" (at the bottom).

@ghislainb ghislainb closed this Apr 3, 2024
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers
No reviews
Assignees
No one assigned
Labels
None yet
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
1 participant
@ghislainb