You are here: Foswiki>ABI Web>ThesesHome>ThesisMCRazer>ThesisMCRazerWeeklyReports>ThesisMCRazerParallelOptions (14 Feb 2010, mriese)Edit Attach

Page ThesisMCRazerParallelOptions

Option Details and Options Pro Contra TBB?
       
Parallelize by Genome
RazerS-Piktogram-Parallel-by-Genome-sm.png Filter and verifiere copy per genome   needs a lot of memory  
    prob. not cache local as it needs to switch between genomes  
Parallelize by Reads
RazerS-Piktogram-Parallel-by-Reads-sm.png coupled filter-verifier-pairs no spin lock for verifier    
each pair over a subset of the reads filter is parallel    
a) pairs operate independent from each other no regulation necessary one might run away (cache)  
b) pairs operate in the same window little regulation, simple find() waiting times  
c) Distance between pairs is restricted   waiting times, regulation for more than two pairs get complex  
distance between filter and verifier fixed ?      
Pipeline
RazerS-Piktogram-Parallel-Hybrid-sm.png One filter over all reads, verification parallelized   spin locks for verifier necessary, probably bad use of capacity in repeats  
a) windows simple find() filters are not parallel (problem with high number of cores)  
b) tokens   complex find() TBB
c) distance between filter and verifier   complex find()  
Edit  | Attach | Print version | History: r4 < r3 < r2 < r1 | Backlinks | View wiki text | Edit wiki text | More topic actions
Topic revision: r3 - 14 Feb 2010, mriese
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback