User Tools

Site Tools


Differences

This shows you the differences between two versions of the page.

Link to this comparison view

Both sides previous revision Previous revision
ffnamespace:performance [2014/08/31 02:51]
aldinuc
ffnamespace:performance [2014/08/31 02:52] (current)
aldinuc
Line 1: Line 1:
 ===== Applications and Performances =====  ===== Applications and Performances ===== 
-==== NGS tools (Bowtie2, BWA) - 2014 ====+==== [2014] ​NGS tools (Bowtie2, BWA) ====
 Bowtie2.0.6,​ Bowtie-2.2.1,​ and BWA compared in performance against their porting onto the FastFlow library. Tested on    Bowtie2.0.6,​ Bowtie-2.2.1,​ and BWA compared in performance against their porting onto the FastFlow library. Tested on   
      * Intel 4-socket 8-core Nehalem (64 HT) @2.0GHz, 72MB L3, 64 GB mem, Linux x86_64      * Intel 4-socket 8-core Nehalem (64 HT) @2.0GHz, 72MB L3, 64 GB mem, Linux x86_64
Line 12: Line 12:
 |{{:​ffnamespace:​bowtie2-speedup.png?​300|}}|{{:​ffnamespace:​bowtie-bwa-maxspeedup.png?​300|}}| |{{:​ffnamespace:​bowtie2-speedup.png?​300|}}|{{:​ffnamespace:​bowtie-bwa-maxspeedup.png?​300|}}|
  
-==== Yadt-ff (parallel C4.5)  - 2012 ====+==== [2012] ​Yadt-ff (parallel C4.5) ====
 The well-known C4.5 statistical classifier is a double hard algorithm. First of all, because data-miners simply would not like to spend time on a yet another brand new parallel version :-) Many past experiences demonstrated that tiny improvements of the sequential algorithm could bring much more performance than a robust investment on parallelization. This clearly does not absolutely mean that parallelization is useless, but, at least in our understanding,​ that a low-effort and conservative parallelization is the only fairly welcome parallelization in the data-mining community. Unfortunately that kind of parallelization,​ i.e. loop and recursion parallelization,​ is technically complex because independent tasks generated in this way may exhibit several non nice proprieties,​ including a huge range of variability in the task size that in turn may induce both severe synchronization overheads and non-trivial load balancing problems that limit the speedup. The well-known C4.5 statistical classifier is a double hard algorithm. First of all, because data-miners simply would not like to spend time on a yet another brand new parallel version :-) Many past experiences demonstrated that tiny improvements of the sequential algorithm could bring much more performance than a robust investment on parallelization. This clearly does not absolutely mean that parallelization is useless, but, at least in our understanding,​ that a low-effort and conservative parallelization is the only fairly welcome parallelization in the data-mining community. Unfortunately that kind of parallelization,​ i.e. loop and recursion parallelization,​ is technically complex because independent tasks generated in this way may exhibit several non nice proprieties,​ including a huge range of variability in the task size that in turn may induce both severe synchronization overheads and non-trivial load balancing problems that limit the speedup.
  
ffnamespace/performance.txt ยท Last modified: 2014/08/31 02:52 by aldinuc