Cascade: a Collaborative Algorithm for Scalable and Efficient Neighborhood Allgather

Neighborhood collectives are a critical feature of MPI, enabling efficient communication in applications with sparse communication patterns. This research proposes Cascade, a new algorithm for neighborhood allgather collective that organizes computing nodes along multiple paths based on their distan...

Celý popis

Podrobná bibliografia

Vydané v:	Proceedings / IEEE International Conference on Cluster Computing s. 1 - 13
Hlavní autori:	Sharifian, Hamed, Sojoodi, Amirhossein, Afsahi, Ahmad
Médium:	Konferenčný príspevok..
Jazyk:	English
Vydavateľské údaje:	IEEE 02.09.2025
Predmet:	Allgather Clustering algorithms Collaboration Computational modeling Costs Kernel MPI Neighborhood Collectives Performance gain Performance Modeling Prediction algorithms Predictive models Scalability Sparse matrices
ISSN:	2168-9253
On-line prístup:	Získať plný text
Tagy:	Pridať tag Žiadne tagy, Buďte prvý, kto otaguje tento záznam!