Benchmarks results of several sequential and parallel implementations of abstract interpretation (on analyze of Scheme programs)
![[Scala-Par-AM]](../_png/Scala-Par-AM--128x128-t.png)
![[Scala-Par-AM QR Code]](../_svg/Scala-Par-AM--QR-Code.svg)
-
Implementation
(in Scala
with Akka):
Scala-Par-AM
(HTML online documentation from the source code)
- Scala source directory: src/main/scala/
- results/ directory that contains all results in raw TSV format and others
- Scheme-examples/ directory
-
Final document of the master thesis:
Parallélisation d’un interpréteur abstrait, au travers du modèle acteur — Application à l’interpréteur abstrait Scala-AM PDF (3.3 Mio) (August 2019) -
Bitbucket repositories with all sources and data:
- An efficient and parallel abstract interpreter in Scala
- An efficient and parallel abstract interpreter in Scala — Preparatory Work —
- Scala-Par-AM: parallel adaptation of Scala-AM
- Scala-AM: Fork of Scala-AM: A Framework for Static Analysis of Dynamic Languages
Results in fancy HTML tables
-
Computed on personal SuperWOPR – Dell Precision T3500 – Intel Xeon CPU W3530 @ 2.8GHz – 4 cores (× 2 by Hyper-Threading)
- Resumed benchmarks of Sergey/jfp/primtest.scm all times
- Resumed benchmarks of Sergey/kcfa/solovay-strassen.scm all times
-
Computed on Bertha (from Software Languages Lab of VUB) – Intel Xeon CPU E5-2637 v3 @ 3.5GHz – 2 processors × 4 cores × 2 by Hyper-Threading
- Resumed benchmarks of AlgoDat1/qsort.scm all times
- Resumed benchmarks of Larceny/Gabriel/takl.scm all times
- Resumed benchmarks of OPi/gen5fibonacci_recur.scm all times
- Resumed benchmarks of OPi/linear_10000.scm all times
- Resumed benchmarks of OPi/linear_5000_gen4fibonacci_recur.scm all times (not in the above List of example Scheme programs)
- Resumed benchmarks of OPi/tree/tree_16.scm all times (not in the above List of example Scheme programs)
- Resumed benchmarks of Parallel_AAM_Andersen_2013/factorial_05.scm with 35 repetitions all times
- Resumed benchmarks of Parallel_AAM_Andersen_2013/hailstone_05.scm with 35 repetitions all times
- Resumed benchmarks of Sergey/jfp/primtest.scm all times
- Resumed benchmarks of Sergey/jfp/primtest.scm with 35 repetitions all times
- Resumed benchmarks of Sergey/kcfa/kcfa-worst-case-20.scm all times
- Resumed benchmarks of Sergey/kcfa/kcfa-worst-case-32.scm all times
- Resumed benchmarks of Sergey/kcfa/kcfa-worst-case-40.scm all times
- Resumed benchmarks of Sergey/kcfa/solovay-strassen.scm all times
-
Computed on Serenity (from Software Languages Lab of VUB) –AMD Opteron 6376 @ 2.3GHz – 4 processors × 16 cores
- Resumed benchmarks of AlgoDat1/qsort.scm all times
- Resumed benchmarks of Larceny/Gabriel/takl.scm all times
- Resumed benchmarks of OPi/gen5fibonacci_recur.scm all times
- Resumed benchmarks of OPi/linear_10000.scm all times
- Resumed benchmarks of Sergey/jfp/primtest.scm all times
- Resumed benchmarks of Sergey/kcfa/kcfa-worst-case-20.scm all times
- Resumed benchmarks of Sergey/kcfa/kcfa-worst-case-32.scm all times
- Resumed benchmarks of Sergey/kcfa/kcfa-worst-case-40.scm all times
- Resumed benchmarks of Sergey/kcfa/solovay-strassen.scm all times
-
Computed on (one node of) Hydra cluster (from ULB and VUB) – Intel Xeon Gold 6148 CPU @ 2.4GHz – 2 processors × 20 cores (not necessary all used)
- (Old) Testing of all example Scheme programs with SeqAAMLS
- (Old) List of example Scheme programs that correctly finished in less than 10 minutes with SeqAAMLS
- Resumed benchmarks of different structures for the worklist all times
- Resumed benchmarks of AlgoDat1/qsort.scm all times
- Resumed benchmarks of Parallel_AAM_Andersen_2013/factorial_05.scm with 35 repetitions all times
- Resumed benchmarks of Sergey/jfp/primtest.scm all times
- Resumed benchmarks of Sergey/jfp/primtest.scm with 35 repetitions all times
- Resumed benchmarks of Sergey/kcfa/solovay-strassen.scm all times