Ignore:
Timestamp:
Apr 21, 2022, 9:02:09 AM (2 years ago)
Author:
m3zulfiq <m3zulfiq@…>
Branches:
ADT, ast-experimental, master, pthread-emulation, qualifiedEnum
Children:
4c530a5
Parents:
5b84a321
Message:

completed performance/evaluations chapter

File:
1 edited

Legend:

Unmodified
Added
Removed
  • doc/theses/mubeen_zulfiqar_MMath/benchmarks.tex

    r5b84a321 r3c79ea9  
    5656-threadA :  sets number of alloc threads (producers) for mem benchmark\\
    5757-consumeS:  sets production and conumption round size\\
    58 -threadF :  sets number of free threads (consumers) for mem benchmark
     58-threadF :  sets number of free threads (consumers) for each producer for mem benchmark
    5959
    6060Object allocation size can be changed using the knobs:\\
     
    7070 It uses following chains and measures allocator runtime speed against them:
    7171\begin{itemize}
    72 \item malloc 0
    73 \item free NULL
    7472\item malloc
    7573\item realloc
     
    10199each chain.
    102100
    103 Number of worker threads can be adjust using a command-line argument -threadN.
     101Following knobs can be adjusted to tune memory usage.\\
     102-maxS    :  sets max object size\\
     103-minS    :  sets min object size\\
     104-stepS   :  sets object size increment\\
     105-distroS :  sets object size distribution\\
     106-objN    :  sets number of objects per thread\\
     107-threadN :  sets number of worker threads\\
    104108
    105109\section{Churn Benchmark} Churn benchmark measures the overall runtime speed of an allocator in a multi-threaded
Note: See TracChangeset for help on using the changeset viewer.