Method

Parallel for loop

• Custom multi-threading code to create parallel operations.
• Same number of threads launched, different number of jobs submitted

Multi-threaded kernel functions

• Kernels run on any number of threads
• Each stress the CPU in slightly different ways

Timing code

• Timing and plotting done with a quick python script.
• Varies number of threads from 1-8
• Plot speedup vs number of threads per kernel

Benchmark results

Results

Floating point addition

Division bound task

Memory bound task

Why add is different

Multiple kernel benchmarks

Division and addition

Addition and memory

Division and memory

More complex kernels

N-body forces calculation

N-body and addition

N-body and memory

Building and running IACA

Building

• mkdir build && cd build
• cmake -DCMAKE_BUILD_TYPE=Release ..
• make

Running intel architecture code analyser

• Uncomment IACA_BEGIN and IACA_END from the kernel code
• cmake –build build
• iaca/iaca -arch HSW build/libkernels.a