PackSELL packs delta-encoded indices and values into single words with tunable bit allocation, delivering up to 1.63x faster FP16 SpMV and FP32-accurate performance exceeding FP16 cuSPARSE while reducing memory traffic.
In: Proceedings of the International Conference for High Performance Computing, Networking, Storage and Analysis, SC ’23
3 Pith papers cite this work, alongside 18 external citations. Polarity classification is still indexing.
3
Pith papers citing it
18
external citations · external index
citation-role summary
background 2
citation-polarity summary
years
2026 3roles
background 2polarities
background 2representative citing papers
PHOENIX recovers permanent node failures in LLM training via hot-swapping of spares using zero-overhead per-step in-memory optimizer-state replication, finishing recovery in under 40 s on up to 512 GPUs.
Review chapter summarizing advances in parallel sparse direct solvers along communication reduction and data-sparse compression axes.
citing papers explorer
-
PackSELL: A Sparse Matrix Format for Precision-Agnostic High-Performance SpMV
PackSELL packs delta-encoded indices and values into single words with tunable bit allocation, delivering up to 1.63x faster FP16 SpMV and FP32-accurate performance exceeding FP16 cuSPARSE while reducing memory traffic.
-
PHOENIX: Resilient LLM Training with Hot-Swapping via Zero-Overhead Checkpoint
PHOENIX recovers permanent node failures in LLM training via hot-swapping of spares using zero-overhead per-step in-memory optimizer-state replication, finishing recovery in under 40 s on up to 512 GPUs.
-
Parallel Sparse and Data-Sparse Factorization-based Linear Solvers
Review chapter summarizing advances in parallel sparse direct solvers along communication reduction and data-sparse compression axes.