Skip to content

Latest commit

 

History

History
23 lines (16 loc) · 1.4 KB

README.md

File metadata and controls

23 lines (16 loc) · 1.4 KB

Optimal C Value

This folder contains our results for sweeping across multiple C values to find the best one for a machine. We ran these on the two eecs583 machines made available to students (server A and B)

Subdirectory Content

Each subdirectory contains the autogenerated csv that was used to create the plots labeled optimal_c_const_value.csv. Furthermore, each subdirectory contains plots described in the following sections.

TEST_c_const_value.png

TEST - This prefix denotes which of the following benchmark this plot represents

  • g500 = graph500
  • hj2 = hashjoin-ph-2
  • hj8 = hashjoin-ph-8
  • cg = nas-cg
  • rand = randacc

The X-Axis represents the hard coded C value, and the Y-Axis represents the execution time of the benchmark. Note, that there are no units for the time component, since each benchmark outputs execution time in different units. These graphs are meant to show relative performance for a single benchmark.

machine_X_c_const_value.png

X - denotes which machine this represents (A = server A, B = server B)

This graph combines all the idividual graphs generated into a single graph where each line represents a different benchmark. Furthermore, the data points for each benchmark are normalized to what the C=32 performance is for that specific benchmark (i.e. each row entry is divided by the C=32 entry in that row).

Note, the Y-Axis now represents the normalized performance.