Optimizing Sparse Linear Algebra for Large-Scale Graph Analytics

Emerging data-intensive applications attempt to process and provide insight into vast amounts of online data. A new class of linear algebra algorithms can efficiently execute sparse matrix-matrix and matrix-vector multiplications on large-scale, shared memory multiprocessor systems, enabling analyst...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Computer (Long Beach, Calif.) Ročník 48; číslo 8; s. 26 - 34
Hlavní autoři: Buono, Daniele, Gunnels, John A., Xinyu Que, Checconi, Fabio, Petrini, Fabrizio, Tai-Ching Tuan, Long, Chris
Médium: Journal Article
Jazyk:angličtina
Vydáno: New York IEEE 01.08.2015
The Institute of Electrical and Electronics Engineers, Inc. (IEEE)
Témata:
Algebra
Algorithms
Computer memory
Concurrent programming
Data analysis
Data-intensive applications
graph analytics
irregular applications
Memory management
Multiprocessing
Multiprocessors
shared memory multiprocessor systems
Social networks
software
Software engineering
sparse linear algebra
Sparse matrices
Sparsity
system performance
ISSN:0018-9162, 1558-0814
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Emerging data-intensive applications attempt to process and provide insight into vast amounts of online data. A new class of linear algebra algorithms can efficiently execute sparse matrix-matrix and matrix-vector multiplications on large-scale, shared memory multiprocessor systems, enabling analysts to more easily discern meaningful data relationships, such as those in social networks.
Bibliografie:SourceType-Scholarly Journals-1
ObjectType-Feature-1
content type line 14
ISSN:0018-9162
1558-0814
DOI:10.1109/MC.2015.228