Skip to content

Instantly share code, notes, and snippets.

@ekinakyurek
Last active February 15, 2019 23:31
Show Gist options
  • Star 0 You must be signed in to star a gist
  • Fork 0 You must be signed in to fork a gist
  • Save ekinakyurek/46d5cad530630cadb0313b78e700577d to your computer and use it in GitHub Desktop.
Save ekinakyurek/46d5cad530630cadb0313b78e700577d to your computer and use it in GitHub Desktop.
LogKnetArraysxCuarrays
[ec2-user@ip-172-31-14-239 ~]$ sh benchmark.sh
++ mktemp -d
+ cd /tmp/tmp.WbX6402Bko
+ git clone https://github.com/ekinakyurek/Mac-Network .
Cloning into '.'...
remote: Enumerating objects: 68, done.
remote: Counting objects: 100% (68/68), done.
remote: Compressing objects: 100% (52/52), done.
remote: Total 328 (delta 30), reused 47 (delta 15), pack-reused 260
Receiving objects: 100% (328/328), 29.89 MiB | 50.42 MiB/s, done.
Resolving deltas: 100% (168/168), done.
+ sed -i 1d benchmark.jl
++ mktemp -d
+ export JULIA_DEPOT_PATH=/tmp/tmp.cBhTWWfPOz
+ JULIA_DEPOT_PATH=/tmp/tmp.cBhTWWfPOz
+ julia --project -e 'using Pkg; Pkg.instantiate(); pkg"free Knet"; pkg"add Knet#cuarrays CuArrays#master CUDAnative#master CUDAdrv#master"'
Cloning default registries into `/tmp/tmp.cBhTWWfPOz`
Cloning registry from "https://github.com/JuliaRegistries/General.git"
Added registry `General` to `/tmp/tmp.cBhTWWfPOz/registries/General`
Cloning git-repo `https://github.com/denizyuret/Knet.jl.git`
Cloning git-repo `https://github.com/ekinakyurek/KnetLayers.jl.git`
Installed AxisAlgorithms ──────────── v0.3.0
Installed PaddedViews ─────────────── v0.4.2
Installed NaNMath ─────────────────── v0.3.2
Installed PlotThemes ──────────────── v0.3.0
Installed CoordinateTransformations ─ v0.5.0
Installed ImageCore ───────────────── v0.7.4
Installed Colors ──────────────────── v0.9.5
Installed SortingAlgorithms ───────── v0.3.1
Installed ImageDistances ──────────── v0.1.1
Installed MappedArrays ────────────── v0.2.1
Installed ProgressMeter ───────────── v0.9.0
Installed IdentityRanges ──────────── v0.3.0
Installed SpecialFunctions ────────── v0.7.2
Installed TranscodingStreams ──────── v0.8.1
Installed Interpolations ──────────── v0.11.2
Installed RangeArrays ─────────────── v0.3.1
Installed BinaryProvider ──────────── v0.5.3
Installed VersionParsing ──────────── v1.1.3
Installed TiledIteration ──────────── v0.2.3
Installed Graphics ────────────────── v0.4.0
Installed ImageShow ───────────────── v0.1.2
Installed Images ──────────────────── v0.17.2
Installed FileIO ──────────────────── v1.0.5
Installed ImageMorphology ─────────── v0.1.1
Installed JSON ────────────────────── v0.20.0
Installed Ratios ──────────────────── v0.3.0
Installed Compat ──────────────────── v1.5.1
Installed Plots ───────────────────── v0.23.0
Installed ColorVectorSpace ────────── v0.6.2
Installed DataStructures ──────────── v0.15.0
Installed PlotUtils ───────────────── v0.5.5
Installed Reexport ────────────────── v0.2.0
Installed StaticArrays ────────────── v0.10.2
Installed BinDeps ─────────────────── v0.8.10
Installed ImageAxes ───────────────── v0.5.0
Installed Conda ───────────────────── v1.2.0
Installed Showoff ─────────────────── v0.2.1
Installed MacroTools ──────────────── v0.4.4
Installed FFTViews ────────────────── v0.2.0
Installed CustomUnitRanges ────────── v0.2.0
Installed RecipesBase ─────────────── v0.6.0
Installed CUDAapi ─────────────────── v0.5.4
Installed SIUnits ─────────────────── v0.1.0
Installed OrderedCollections ──────── v1.0.2
Installed URIParser ───────────────── v0.4.0
Installed ImageMagick ─────────────── v0.7.1
Installed OffsetArrays ────────────── v0.10.0
Installed IndirectArrays ──────────── v0.5.0
Installed ColorTypes ──────────────── v0.7.5
Installed StatsBase ───────────────── v0.27.0
Installed Requires ────────────────── v0.5.2
Installed TimerOutputs ────────────── v0.4.0
Installed WoodburyMatrices ────────── v0.4.1
Installed TexExtensions ───────────── v0.1.0
Installed ComputationalResources ──── v0.3.0
Installed AxisArrays ──────────────── v0.3.0
Installed SimpleTraits ────────────── v0.8.0
Installed ImageFiltering ──────────── v0.5.3
Installed ImageTransformations ────── v0.7.2
Installed IterTools ───────────────── v1.1.1
Installed AutoGrad ────────────────── v1.1.3
Installed Distances ───────────────── v0.7.4
Installed Contour ─────────────────── v0.5.1
Installed CodecZlib ───────────────── v0.5.1
Installed FFTW ────────────────────── v0.2.4
Installed ImageMetadata ───────────── v0.6.1
Installed AbstractFFTs ────────────── v0.3.2
Installed IntervalSets ────────────── v0.3.1
Installed GR ──────────────────────── v0.37.0
Installed CatIndices ──────────────── v0.2.0
Installed Missings ────────────────── v0.4.0
Installed JLD2 ────────────────────── v0.1.2
Installed FixedPointNumbers ───────── v0.5.3
Installed Crayons ─────────────────── v1.0.0
Installed Measures ────────────────── v0.3.0
Installed Rotations ───────────────── v0.10.0
Building SpecialFunctions → `/tmp/tmp.cBhTWWfPOz/packages/SpecialFunctions/fvheQ/deps/build.log`
Building GR ──────────────→ `/tmp/tmp.cBhTWWfPOz/packages/GR/shnUy/deps/build.log`
Building Plots ───────────→ `/tmp/tmp.cBhTWWfPOz/packages/Plots/UQI78/deps/build.log`
Building Conda ───────────→ `/tmp/tmp.cBhTWWfPOz/packages/Conda/CpuvI/deps/build.log`
Building FFTW ────────────→ `/tmp/tmp.cBhTWWfPOz/packages/FFTW/p7sLQ/deps/build.log`
Building ImageMagick ─────→ `/tmp/tmp.cBhTWWfPOz/packages/ImageMagick/d5KBL/deps/build.log`
Building CodecZlib ───────→ `/tmp/tmp.cBhTWWfPOz/packages/CodecZlib/DAjXH/deps/build.log`
Building Knet ────────────→ `/tmp/tmp.cBhTWWfPOz/packages/Knet/T1oum/deps/build.log`
Resolving package versions...
Installed Rotations ─ v0.11.1
Installed GR ──────── v0.38.1
Updating `/tmp/tmp.WbX6402Bko/Project.toml`
[1902f260] ~ Knet v1.2.0+ #56de50a (https://github.com/denizyuret/Knet.jl.git) ⚲ ⇒ v1.2.0+ #56de50a (https://github.com/denizyuret/Knet.jl.git)
[80bfaf46] ~ KnetLayers v0.1.0+ #abea616 (https://github.com/ekinakyurek/KnetLayers.jl.git) ⚲ ⇒ v0.1.0 #abea616 (https://github.com/ekinakyurek/KnetLayers.jl.git) ⚲
Updating `/tmp/tmp.WbX6402Bko/Manifest.toml`
[28b8d3ca] ↑ GR v0.37.0 ⇒ v0.38.1
[1902f260] ~ Knet v1.2.0+ #56de50a (https://github.com/denizyuret/Knet.jl.git) ⚲ ⇒ v1.2.0+ #56de50a (https://github.com/denizyuret/Knet.jl.git)
[80bfaf46] ~ KnetLayers v0.1.0+ #abea616 (https://github.com/ekinakyurek/KnetLayers.jl.git) ⚲ ⇒ v0.1.0 #abea616 (https://github.com/ekinakyurek/KnetLayers.jl.git) ⚲
[6038ab10] ↑ Rotations v0.10.0 ⇒ v0.11.1
Building GR → `/tmp/tmp.cBhTWWfPOz/packages/GR/IVBgs/deps/build.log`
Updating git-repo `https://github.com/denizyuret/Knet.jl.git`
Cloning git-repo `https://github.com/JuliaGPU/CuArrays.jl.git`
Updating git-repo `https://github.com/JuliaGPU/CuArrays.jl.git`
Cloning git-repo `https://github.com/JuliaGPU/CUDAnative.jl.git`
Updating git-repo `https://github.com/JuliaGPU/CUDAnative.jl.git`
Cloning git-repo `https://github.com/JuliaGPU/CUDAdrv.jl.git`
Updating git-repo `https://github.com/JuliaGPU/CUDAdrv.jl.git`
Resolving package versions...
Installed DiffResults ────────── v0.0.4
Installed ForwardDiff ────────── v0.10.2
Installed CommonSubexpressions ─ v0.2.0
Installed DiffRules ──────────── v0.0.9
Installed NNlib ──────────────── v0.4.3
Installed Adapt ──────────────── v0.4.2
Installed FillArrays ─────────── v0.4.0
Installed GPUArrays ──────────── v0.6.1
Installed LLVM ───────────────── v1.0.0
Updating `/tmp/tmp.WbX6402Bko/Project.toml`
[c5f51814] + CUDAdrv v1.0.1+ #master (https://github.com/JuliaGPU/CUDAdrv.jl.git)
[be33ccc6] + CUDAnative v1.0.1+ #master (https://github.com/JuliaGPU/CUDAnative.jl.git)
[3a865a2d] + CuArrays v0.10.0 #master (https://github.com/JuliaGPU/CuArrays.jl.git)
[1902f260] ~ Knet v1.2.0+ #56de50a (https://github.com/denizyuret/Knet.jl.git) ⇒ v1.2.0+ #cuarrays (https://github.com/denizyuret/Knet.jl.git)
Updating `/tmp/tmp.WbX6402Bko/Manifest.toml`
[79e6a3ab] + Adapt v0.4.2
[c5f51814] + CUDAdrv v1.0.1+ #master (https://github.com/JuliaGPU/CUDAdrv.jl.git)
[be33ccc6] + CUDAnative v1.0.1+ #master (https://github.com/JuliaGPU/CUDAnative.jl.git)
[bbf7d656] + CommonSubexpressions v0.2.0
[3a865a2d] + CuArrays v0.10.0 #master (https://github.com/JuliaGPU/CuArrays.jl.git)
[163ba53b] + DiffResults v0.0.4
[b552c78f] + DiffRules v0.0.9
[1a297f60] + FillArrays v0.4.0
[f6369f11] + ForwardDiff v0.10.2
[0c68f7d7] + GPUArrays v0.6.1
[1902f260] ~ Knet v1.2.0+ #56de50a (https://github.com/denizyuret/Knet.jl.git) ⇒ v1.2.0+ #cuarrays (https://github.com/denizyuret/Knet.jl.git)
[929cbde3] + LLVM v1.0.0
[872c559c] + NNlib v0.4.3
Building LLVM ──────→ `/tmp/tmp.cBhTWWfPOz/packages/LLVM/tPWXv/deps/build.log`
Building CUDAdrv ───→ `/tmp/tmp.cBhTWWfPOz/packages/CUDAdrv/2JOOG/deps/build.log`
Building CUDAnative → `/tmp/tmp.cBhTWWfPOz/packages/CUDAnative/QvP9a/deps/build.log`
Building CuArrays ──→ `/tmp/tmp.cBhTWWfPOz/packages/CuArrays/SkEdG/deps/build.log`
Building Knet ──────→ `/tmp/tmp.cBhTWWfPOz/packages/Knet/LMOxK/deps/build.log`
+ julia --project -L benchmark.jl -e 'for i in 1:3 @show @time benchmark(M,o;N=100) end'
arrtype = KnetArray{Float32,N} where N
[ Info: Building the CUDAnative run-time library for your sm_37 device, this might take a while...
78.341572 seconds (18.58 M allocations: 10.364 GiB, 1.42% gc time)
#= none:1 =# @time(benchmark(M, o; N=100)) = nothing
78.549632 seconds (18.58 M allocations: 10.364 GiB, 1.41% gc time)
#= none:1 =# @time(benchmark(M, o; N=100)) = nothing
78.663026 seconds (18.58 M allocations: 10.364 GiB, 1.42% gc time)
#= none:1 =# @time(benchmark(M, o; N=100)) = nothing
++ mktemp -d
+ export JULIA_DEPOT_PATH=/tmp/tmp.stkhbrUgaE
+ JULIA_DEPOT_PATH=/tmp/tmp.stkhbrUgaE
+ julia --project -e 'using Pkg; Pkg.instantiate(); pkg"free Knet";pkg"add Knet#master"'
Cloning default registries into `/tmp/tmp.stkhbrUgaE`
Cloning registry from "https://github.com/JuliaRegistries/General.git"
Added registry `General` to `/tmp/tmp.stkhbrUgaE/registries/General`
Cloning git-repo `https://github.com/denizyuret/Knet.jl.git`
Cloning git-repo `https://github.com/JuliaGPU/CUDAnative.jl.git`
Cloning git-repo `https://github.com/JuliaGPU/CuArrays.jl.git`
Cloning git-repo `https://github.com/ekinakyurek/KnetLayers.jl.git`
Cloning git-repo `https://github.com/JuliaGPU/CUDAdrv.jl.git`
Installed NaNMath ─────────────────── v0.3.2
Installed PaddedViews ─────────────── v0.4.2
Installed PlotThemes ──────────────── v0.3.0
Installed AxisAlgorithms ──────────── v0.3.0
Installed CoordinateTransformations ─ v0.5.0
Installed ImageCore ───────────────── v0.7.4
Installed Colors ──────────────────── v0.9.5
Installed SpecialFunctions ────────── v0.7.2
Installed SortingAlgorithms ───────── v0.3.1
Installed ProgressMeter ───────────── v0.9.0
Installed Interpolations ──────────── v0.11.2
Installed TranscodingStreams ──────── v0.8.1
Installed IdentityRanges ──────────── v0.3.0
Installed ForwardDiff ─────────────── v0.10.2
Installed MappedArrays ────────────── v0.2.1
Installed ImageDistances ──────────── v0.1.1
Installed RangeArrays ─────────────── v0.3.1
Installed ImageShow ───────────────── v0.1.2
Installed Graphics ────────────────── v0.4.0
Installed BinaryProvider ──────────── v0.5.3
Installed FileIO ──────────────────── v1.0.5
Installed Images ──────────────────── v0.17.2
Installed FillArrays ──────────────── v0.4.0
Installed VersionParsing ──────────── v1.1.3
Installed CommonSubexpressions ────── v0.2.0
Installed ImageMorphology ─────────── v0.1.1
Installed TiledIteration ──────────── v0.2.3
Installed JSON ────────────────────── v0.20.0
Installed Adapt ───────────────────── v0.4.2
Installed Ratios ──────────────────── v0.3.0
Installed Plots ───────────────────── v0.23.0
Installed PlotUtils ───────────────── v0.5.5
Installed Compat ──────────────────── v1.5.1
Installed DiffRules ───────────────── v0.0.9
Installed DataStructures ──────────── v0.15.0
Installed ColorVectorSpace ────────── v0.6.2
Installed GPUArrays ───────────────── v0.6.1
Installed BinDeps ─────────────────── v0.8.10
Installed Reexport ────────────────── v0.2.0
Installed StaticArrays ────────────── v0.10.2
Installed Conda ───────────────────── v1.2.0
Installed ImageAxes ───────────────── v0.5.0
Installed MacroTools ──────────────── v0.4.4
Installed FFTViews ────────────────── v0.2.0
Installed Showoff ─────────────────── v0.2.1
Installed CustomUnitRanges ────────── v0.2.0
Installed DiffResults ─────────────── v0.0.4
Installed CUDAapi ─────────────────── v0.5.4
Installed RecipesBase ─────────────── v0.6.0
Installed SIUnits ─────────────────── v0.1.0
Installed OrderedCollections ──────── v1.0.2
Installed ImageMagick ─────────────── v0.7.1
Installed URIParser ───────────────── v0.4.0
Installed OffsetArrays ────────────── v0.10.0
Installed ColorTypes ──────────────── v0.7.5
Installed IndirectArrays ──────────── v0.5.0
Installed StatsBase ───────────────── v0.27.0
Installed Requires ────────────────── v0.5.2
Installed TimerOutputs ────────────── v0.4.0
Installed TexExtensions ───────────── v0.1.0
Installed NNlib ───────────────────── v0.4.3
Installed GR ──────────────────────── v0.38.1
Installed WoodburyMatrices ────────── v0.4.1
Installed AxisArrays ──────────────── v0.3.0
Installed ComputationalResources ──── v0.3.0
Installed SimpleTraits ────────────── v0.8.0
Installed IterTools ───────────────── v1.1.1
Installed ImageFiltering ──────────── v0.5.3
Installed ImageTransformations ────── v0.7.2
Installed AutoGrad ────────────────── v1.1.3
Installed Distances ───────────────── v0.7.4
Installed FFTW ────────────────────── v0.2.4
Installed Contour ─────────────────── v0.5.1
Installed CodecZlib ───────────────── v0.5.1
Installed AbstractFFTs ────────────── v0.3.2
Installed ImageMetadata ───────────── v0.6.1
Installed IntervalSets ────────────── v0.3.1
Installed JLD2 ────────────────────── v0.1.2
Installed Missings ────────────────── v0.4.0
Installed FixedPointNumbers ───────── v0.5.3
Installed LLVM ────────────────────── v1.0.0
Installed CatIndices ──────────────── v0.2.0
Installed Measures ────────────────── v0.3.0
Installed Rotations ───────────────── v0.11.1
Installed Crayons ─────────────────── v1.0.0
Building SpecialFunctions → `/tmp/tmp.stkhbrUgaE/packages/SpecialFunctions/fvheQ/deps/build.log`
Building GR ──────────────→ `/tmp/tmp.stkhbrUgaE/packages/GR/IVBgs/deps/build.log`
Building Plots ───────────→ `/tmp/tmp.stkhbrUgaE/packages/Plots/UQI78/deps/build.log`
Building Conda ───────────→ `/tmp/tmp.stkhbrUgaE/packages/Conda/CpuvI/deps/build.log`
Building FFTW ────────────→ `/tmp/tmp.stkhbrUgaE/packages/FFTW/p7sLQ/deps/build.log`
Building ImageMagick ─────→ `/tmp/tmp.stkhbrUgaE/packages/ImageMagick/d5KBL/deps/build.log`
Building CodecZlib ───────→ `/tmp/tmp.stkhbrUgaE/packages/CodecZlib/DAjXH/deps/build.log`
Building LLVM ────────────→ `/tmp/tmp.stkhbrUgaE/packages/LLVM/tPWXv/deps/build.log`
Building CUDAdrv ─────────→ `/tmp/tmp.stkhbrUgaE/packages/CUDAdrv/2JOOG/deps/build.log`
Building CUDAnative ──────→ `/tmp/tmp.stkhbrUgaE/packages/CUDAnative/QvP9a/deps/build.log`
Building CuArrays ────────→ `/tmp/tmp.stkhbrUgaE/packages/CuArrays/SkEdG/deps/build.log`
Building Knet ────────────→ `/tmp/tmp.stkhbrUgaE/packages/Knet/LMOxK/deps/build.log`
Resolving package versions...
Installed Knet ─ v1.2.0
Updating `/tmp/tmp.WbX6402Bko/Project.toml`
[1902f260] ↓ Knet v1.2.0+ #cuarrays (https://github.com/denizyuret/Knet.jl.git) ⇒ v1.2.0
Updating `/tmp/tmp.WbX6402Bko/Manifest.toml`
[1902f260] ↓ Knet v1.2.0+ #cuarrays (https://github.com/denizyuret/Knet.jl.git) ⇒ v1.2.0
Building Knet → `/tmp/tmp.stkhbrUgaE/packages/Knet/T1oum/deps/build.log`
Updating git-repo `https://github.com/denizyuret/Knet.jl.git`
Resolving package versions...
Updating `/tmp/tmp.WbX6402Bko/Project.toml`
[1902f260] ↑ Knet v1.2.0 ⇒ v1.2.0+ #master (https://github.com/denizyuret/Knet.jl.git)
Updating `/tmp/tmp.WbX6402Bko/Manifest.toml`
[1902f260] ↑ Knet v1.2.0 ⇒ v1.2.0+ #master (https://github.com/denizyuret/Knet.jl.git)
Building Knet → `/tmp/tmp.stkhbrUgaE/packages/Knet/JJd49/deps/build.log`
+ julia --project -L benchmark.jl -e 'for i in 1:3 @show @time benchmark(M,o;N=100) end'
arrtype = KnetArray{Float32,N} where N
68.307643 seconds (11.11 M allocations: 5.232 GiB, 0.75% gc time)
#= none:1 =# @time(benchmark(M, o; N=100)) = nothing
68.499359 seconds (11.12 M allocations: 5.232 GiB, 0.61% gc time)
#= none:1 =# @time(benchmark(M, o; N=100)) = nothing
68.678919 seconds (11.12 M allocations: 5.232 GiB, 0.61% gc time)
#= none:1 =# @time(benchmark(M, o; N=100)) = nothing
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment