Poster Type: Research Posters
Author: Zenchang Sun (Cornell University), Arman Margarian (Cornell University), Jefferson Zhou (Cornell University), Giulia Guidi (Cornell University)
Abstract: Sparse matrix-vector multiplication (SpMV) is bounded by data movement rather than arithmetic, since every nonzero fetches from an unpredictable address. We present a general tile-wise SpMV kernel for the Cerebras WSE-3, whose ~893,000 processing elements (PE) hold the working set in on-chip SRAM. The kernel partitions the matrix with a deterministic host-side balancing algorithm and sums per-PE partials eastward along each grid row. Sweeping to the full wafer, the local compute scales linearly and a two-parameter model predicts it exactly across 17 configurations; the eastward reduction, parallel along one axis only, is the sole limit on total scaling, and a standalone chain-latency model underpredicts it by up to 26% once it overlaps compute. On a biobank-scale genotype matrix of 5.28 billion nonzeros the kernel sustains 64,700 GFLOP/s; on a matched smaller instance it completes in 59.6 μs against 5.46 ms for four A100 GPUs.
Best Poster Finalist (BP): no
Poster: PDF
Poster Summary: PDF