data/evaluate_full_node_bench.jl

@@ -0,0 +1,200 @@
+using CSV
+using DataFrames
+using Plots
+using StatsPlots
+using LaTeXStrings
+
+if (length(ARGS) < 1)
+    println("Please use with \"input_file.csv\"")
+end
+
+processes = [
+    "QED Process: 'ke->ke'",
+    "QED Process: 'ke->kke'",
+    "QED Process: 'ke->kkke'",
+    "QED Process: 'ke->kkkke'",
+    "QED Process: 'ke->kkkkke'",
+]
+
+function proc_to_n(str::AbstractString)
+    parts = split(str, "'")
+    parts = split(parts[2], "->")
+    k_count = count(c -> c == 'k', parts[2])
+    return k_count
+end
+
+function beautify_title(str::AbstractString)
+    parts = split(str, "'")
+
+    preprefix = parts[1]
+    infix = parts[2]
+    sufsuffix = parts[3]
+
+    parts = split(infix, "->")
+
+    prefix = parts[1]
+    suffix = parts[2]
+
+    k_count = count(c -> c == 'k', suffix)
+    B_count = count(c -> c == 'B', suffix)
+
+    if k_count == 1 || B_count == 1
+        new_suffix = suffix
+    elseif k_count >= 1
+        new_suffix = replace(suffix, r"k+" => "k^$k_count")
+    elseif B_count >= 1
+        new_suffix = replace(suffix, r"B+" => "B^$B_count")
+    end
+
+    return "QED Compton Scattering Process " * L"%$prefix \rightarrow %$new_suffix" * sufsuffix
+end
+
+input_file = ARGS[1]
+df = CSV.read(input_file, DataFrame)
+n_inputs = df[:, "n_inputs"][1]
+gpus = df.gpu_devices[1]
+cpus = df.cpu_threads[1]
+
+power = Int(round(log2(n_inputs)))
+
+chunk_sizes = ["1024", "4096", "16384", "65536", "262144", "1048576"]
+
+best_times = Vector{Float64}()
+best_times_std = Vector{Float64}()
+
+# plotting with threads as x axis
+for process_name in processes
+    df_filt = filter(:process_name => x -> x == process_name, df)
+
+    df_filt.cpu_ratio = df_filt.cpu_chunks ./ (df_filt.cpu_chunks .+ df_filt.gpu_chunks) .* 100.0
+    df_filt.gpu_ratio = df_filt.gpu_chunks ./ (df_filt.cpu_chunks .+ df_filt.gpu_chunks) .* 100.0
+
+    push!(best_times, minimum(df_filt.time))
+
+    bar(chunk_sizes, df_filt.cpu_ratio, label = "workload completed by \$$(cpus)\$ CPU threads (%)")
+    bar!(
+        chunk_sizes,
+        [100, 100, 100, 100, 100, 100],
+        label = "workload completed by $(gpus) GPUs (%)",
+        fillto = df_filt.cpu_ratio,
+    )
+
+    plot!(
+        title = "$(beautify_title(process_name))\nComputing $(n_inputs) (\$2^{$(power)}\$) Matrix Elements",
+        yscale = :linear,
+        #xticks = [1024 4096 16384 65536 262144 1048576],
+        ylim = (0, 105),
+        legend = :outerbottom,
+        legendcolumns = 1,
+        legend_font_pointsize = 10,
+        size = (800, 600),
+        ylabel = "contribution (%)",
+        xlabel = "chunk size (#)",
+    )
+
+    savefig("full_node_chunk_size_$(process_name)_ratio.pdf")
+
+
+    scatter(
+        chunk_sizes,
+        df_filt.rate,
+        label = "total execution rate (\$s^{-1}\$)",
+        title = "$(beautify_title(process_name))\nComputing $(n_inputs) (\$2^{$(power)}\$) Matrix Elements",
+        yscale = :log10,
+        #xticks = [1024 4096 16384 65536 262144 1048576],
+        legend = :outerbottom,
+        legendcolumns = 1,
+        legend_font_pointsize = 10,
+        size = (800, 600),
+        ylabel = "rate (\$s^{-1}\$)",
+        xlabel = "chunk size (#)",
+        markersize = 7,
+    )
+
+    savefig("full_node_chunk_size_$(process_name)_rate.pdf")
+
+
+    scatter(
+        chunk_sizes,
+        df_filt.time,
+        yerror = df_filt.std,
+        label = "total execution time (s)",
+        title = "$(beautify_title(process_name))\nComputing $(n_inputs) (\$2^{$(power)}\$) Matrix Elements",
+        #xticks = [1024 4096 16384 65536 262144 1048576],
+        ylim = (0, maximum(df_filt.time) * 1.05),
+        legend = :outerbottom,
+        legendcolumns = 1,
+        legend_font_pointsize = 10,
+        size = (800, 600),
+        ylabel = "time (s)",
+        xlabel = "chunk size (#)",
+        markersize = 7,
+    )
+
+    savefig("full_node_chunk_size_$(process_name)_time.pdf")
+
+end
+
+
+# plotting with process size as x axis
+A100_rates = [2.530045276927587e9, 1.16972304616864e9, 2.0002725972692013e8, 3.495722925446318e7, 4.792187095617111e6]
+CPU_32threads_rates =
+    [3.2691139045711152e7, 1.1578342663759507e7, 3.1670680975577887e6, 731037.7069429948, 115001.5594731802]
+
+theory_rates = (A100_rates .+ CPU_32threads_rates) .* 4
+
+scatter(
+    proc_to_n.(processes),
+    best_times,
+    label = "full node best achieved time (s)",
+    title = "QED N-Photon Compton Scattering\nComputing $(n_inputs) (\$2^{$(power)}\$) Matrix Elements",
+    ylim = (0, maximum(best_times) * 1.05),
+    legend = :outerbottom,
+    legendcolumns = 1,
+    legend_font_pointsize = 10,
+    size = (800, 600),
+    ylabel = "time (s)",
+    xlabel = "process size (#)",
+    markersize = 7,
+)
+
+savefig("full_node_process_best_time.pdf")
+
+
+scatter(
+    proc_to_n.(processes),
+    (n_inputs ./ best_times),
+    label = "full node best achieved rate (\$s^{-1}\$)",
+    title = "QED N-Photon Compton Scattering\nComputing $(n_inputs) (\$2^{$(power)}\$) Matrix Elements",
+    ylim = (0, maximum(n_inputs ./ best_times) * 1.05),
+    legend = :outerbottom,
+    legendcolumns = 1,
+    legend_font_pointsize = 10,
+    size = (800, 600),
+    ylabel = "rate (\$s^{-1}\$)",
+    xlabel = "process size (#)",
+    markersize = 7,
+)
+
+savefig("full_node_process_best_rate.pdf")
+
+
+
+scatter(
+    proc_to_n.(processes),
+    [(n_inputs ./ best_times) theory_rates],
+    label = ["full node best achieved rate (\$s^{-1}\$)" "theoretical rate from previous benchmarks (\$s^{-1}\$)"],
+    title = "QED N-Photon Compton Scattering\nComputing $(n_inputs) (\$2^{$(power)}\$) Matrix Elements",
+    #ylim = (0, max(maximum(n_inputs ./ best_times), maximum(theory_rates)) * 1.05),
+    yscale = :log10,
+    legend = :outerbottom,
+    legendcolumns = 1,
+    legend_font_pointsize = 10,
+    size = (800, 600),
+    ylabel = "rate (\$s^{-1}\$)",
+    xlabel = "process size (#)",
+    markersize = 7,
+)
+
+savefig("full_node_process_best_rate_plus_theory.pdf")
+

data/full_node_bench_hemera_32t_4a100_371467c.csv

@@ -0,0 +1,31 @@
+process_name,cpu_threads,gpu_devices,n_inputs,chunk_size,time,std,rate,cpu_chunks,gpu_chunks,memory_est
+QED Process: 'ke->ke',124,4,67108864,1024,3.498023189,0.3076721764605935,1.9184796776371513e7,45294.0,20242.0,1.447298688e10
+QED Process: 'ke->ke',124,4,67108864,4096,2.401494263,0.17013104933519752,2.7944628073425464e7,8872.5,7511.5,1.340607224e10
+QED Process: 'ke->ke',124,4,67108864,16384,2.040338948,0.11818602853354224,3.2891037082726903e7,1942.0,2154.0,1.2850894448e10
+QED Process: 'ke->ke',124,4,67108864,65536,1.9094240825,0.0703634619424293,3.514612841382763e7,431.0,593.0,1.2549270576e10
+QED Process: 'ke->ke',124,4,67108864,262144,1.663800835,0.09683669072918406,4.033467383131828e7,124.0,132.0,1.3304598192e10
+QED Process: 'ke->ke',124,4,67108864,1048576,4.363178006,0.15362201468327347,1.5380730263059545e7,60.0,4.0,1.635789408e10
+QED Process: 'ke->kke',124,4,67108864,1024,3.595792702,0.22982494201906578,1.8663162635230247e7,38403.0,27133.0,1.6004835648e10
+QED Process: 'ke->kke',124,4,67108864,4096,2.759823507,0.1517042378902089,2.431636074907887e7,7680.5,8703.5,1.521632832e10
+QED Process: 'ke->kke',124,4,67108864,16384,2.438895872,0.2675005760041791,2.7516084130712736e7,1666.0,2430.0,1.4017263776e10
+QED Process: 'ke->kke',124,4,67108864,65536,2.102141753,0.19062540340211018,3.19240431356391e7,372.0,652.0,1.4542648832e10
+QED Process: 'ke->kke',124,4,67108864,262144,1.963817117,0.270899103650203,3.417266476550423e7,124.0,132.0,1.5452095984e10
+QED Process: 'ke->kke',124,4,67108864,1048576,4.678252536,0.24347387468900175,1.4344857077206746e7,60.0,4.0,1.885768088e10
+QED Process: 'ke->kkke',124,4,67108864,1024,3.922357603,0.5369159040482943,1.7109318117417965e7,25347.0,40189.0,1.672316376e10
+QED Process: 'ke->kkke',124,4,67108864,4096,2.363901747,0.2856164398652175,2.8389024241454653e7,3829.0,12555.0,1.5615308512e10
+QED Process: 'ke->kkke',124,4,67108864,16384,1.893031068,0.15072532448179204,3.545048210481879e7,744.0,3352.0,1.5101105104e10
+QED Process: 'ke->kkke',124,4,67108864,65536,2.137146225,0.12041389223719733,3.140115693300303e7,248.0,776.0,1.5780112288e10
+QED Process: 'ke->kkke',124,4,67108864,262144,2.776542017,0.01817299764175009,2.4169943616596095e7,124.0,132.0,1.7599575872e10
+QED Process: 'ke->kkke',124,4,67108864,1048576,9.735021382,0.14312958275274476,6.89355075522319e6,60.0,4.0,2.1005184928e10
+QED Process: 'ke->kkkke',124,4,67108864,1024,13.24530958,1.1412505488619353,5.066613475107616e6,24940.0,40596.0,1.8619863968e10
+QED Process: 'ke->kkkke',124,4,67108864,4096,6.654388343,2.053901855132858,1.0084903456317563e7,3588.5,12795.5,1.7528517888e10
+QED Process: 'ke->kkkke',124,4,67108864,16384,4.255322572,0.4138751997857821,1.5770570353837797e7,496.0,3600.0,1.7020825088e10
+QED Process: 'ke->kkkke',124,4,67108864,65536,3.756418218,2.7132996417968327,1.78651204699274e7,124.0,900.0,1.701754576e10
+QED Process: 'ke->kkkke',124,4,67108864,262144,12.8016601195,0.06866851952937757,5.24220010323326e6,124.0,132.0,1.974705816e10
+QED Process: 'ke->kkkke',124,4,67108864,1048576,45.664600753,0.4542000947311946,1.4696036512613369e6,62.0,2.0,2.3504986032e10
+QED Process: 'ke->kkkkke',124,4,67108864,1024,56.517159165,NaN,1.1874068865364918e6,24819.0,40717.0,2.1017071136e10
+QED Process: 'ke->kkkkke',124,4,67108864,4096,30.239194471,8.720258576297342,2.2192675821559583e6,2980.0,13404.0,1.940682472e10
+QED Process: 'ke->kkkkke',124,4,67108864,16384,12.192125241,2.800688574554207,5.5042794158908855e6,275.0,3821.0,1.8713809728e10
+QED Process: 'ke->kkkkke',124,4,67108864,65536,19.262260871,0.18994391422350254,3.483955723029103e6,124.0,900.0,1.9165067936e10
+QED Process: 'ke->kkkkke',124,4,67108864,262144,74.888312523,NaN,896119.3240853072,124.0,132.0,2.1894531872e10
+QED Process: 'ke->kkkkke',124,4,67108864,1048576,269.327736071,NaN,249171.7525235084,61.5,2.5,2.5652469296e10