Engineering a Distributed-Memory Triangle Counting Algorithm

Counting triangles in a graph and incident to each vertex is a fundamental and frequently considered task of graph analysis. We consider how to efficiently do this for huge graphs using massively parallel distributed-memory machines. Unsurprisingly, the main issue is to reduce communication between...

Celý popis

Uloženo v:
Podrobná bibliografie
Vydáno v:Proceedings - IEEE International Parallel and Distributed Processing Symposium s. 702 - 712
Hlavní autoři: Sanders, Peter, Uhl, Tim Niklas
Médium: Konferenční příspěvek
Jazyk:angličtina
Vydáno: IEEE 01.05.2023
Témata:
clustering coefficient
Distributed processing
distributed-memory algorithm
graph analysis
Memory management
MPI
Program processors
Routing
Task analysis
triangle counting
ISSN:1530-2075
On-line přístup:Získat plný text
Tagy: Přidat tag
Žádné tagy, Buďte první, kdo vytvoří štítek k tomuto záznamu!
Popis
Shrnutí:Counting triangles in a graph and incident to each vertex is a fundamental and frequently considered task of graph analysis. We consider how to efficiently do this for huge graphs using massively parallel distributed-memory machines. Unsurprisingly, the main issue is to reduce communication between processors. We achieve this by counting locally whenever possible and reducing the amount of information that needs to be sent in order to handle (possible) nonlocal triangles. We also achieve linear memory requirements despite superlinear communication volume by introducing a new asynchronous sparse-all-to-all operation. Furthermore, we dramatically reduce startup overheads by allowing this communication to use indirect routing. Our algorithms scale (at least) up to 32 768 cores and are up to 18 times faster than the previous state of the art.
ISSN:1530-2075
DOI:10.1109/IPDPS54959.2023.00076