Parallel String Graph Construction and Transitive Reduction for De Novo Genome Assembly

Guidi, Giulia; Selvitopi, Oğuz; Ellis, Marquita; Oliker, Leonid; Yelick, Katherine; Buluç, Aydın

doi:10.1109/ipdps49936.2021.00060

Cited by 7 publications

(2 citation statements)

References 32 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…CombBLAS is currently the primary backend of several distributed-memory applications such as the graph clustering code HipMCL [38], protein homology search code PASTIS [9], and read-to-read genome aligner di-BELLA.2D [10]. Several other applications such as the sparse direct solver SuperLU [44], the structured dense matrix solver STRUMPACK [45], and the sketching library libskylark [46] also utilize CombBLAS for certain tasks.…”

Section: Discussionmentioning

confidence: 99%

“…The primary one is the capability to use arbitrary functions and semirings in its operations. The ability to pass userdefined functions and semirings allow CombBLAS to be extremely flexible in terms of implementing a wide range of combinatorial algorithms, including those arising in graph computations [7], [8] as well as other areas such as computational biology [9], [10]. CombBLAS also takes advantage of C++ templates to a high degree, allowing the user to implement previously unspecified operations by just changing the template type of matrix/vector values and the semiring definition.…”

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

Combinatorial BLAS 2.0: Scaling combinatorial algorithms on distributed-memory systems

Azad¹,

Selvitopi²,

Hussain³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Combinatorial algorithms such as those that arise in graph analysis, modeling of discrete systems, bioinformatics, and chemistry, are often hard to parallelize. The Combinatorial BLAS library implements key computational primitives for rapid development of combinatorial algorithms in distributed-memory systems. During the decade since its first introduction, the Combinatorial BLAS library has evolved and expanded significantly. This paper details many of the key technical features of Combinatorial BLAS version 2.0, such as communication avoidance, hierarchical parallelism via in-node multithreading, accelerator support via GPU kernels, generalized semiring support, implementations of key data structures and functions, and scalable distributed I/O operations for human-readable files. Our paper also presents several rules of thumb for choosing the right data structures and functions in Combinatorial BLAS 2.0, under various common application scenarios.

show abstract