一句话：提出首个细粒度PIM设计，同时加速顶点中心push式图应用的update与apply阶段。
PIM利用DRAM内部带宽与并行性带来显著性能收益，但图应用的不规则访存使其难以适配PIM。现有细粒度PIM方案通常只加速顶点中心push式图计算的update阶段，导致apply阶段成为瓶颈。CoGraf提出首个完全加速update与apply两阶段的FGPIM设计：在update阶段，采用可多粒度合并的tuple-based LLC聚合图更新，并设计multi-DRAM column FGPIM指令以匹配FGPIM的行级并行；在apply阶段，提出带谓词的bank-parallel FGPIM指令，使条件更新也能被加速。评估显示，在感兴趣区域相对朴素FGPIM平均加速1.8×/3×，相对HBM2/DDR4最佳非PIM基线加速4.4×/9.8×，DRAM能耗降低67%/86%和88%/94%。结果表明，同时加速update与apply的完整方案对图应用至关重要。
Processing-in-Memory (PIM) delivers enormous performance by taking advantage of internal DRAM bandwidth and parallelism. However, graph applications are difficult to adapt to PIM due to their irregular access patterns. We present the first Fine-Grained PIM (FGPIM) design that fully accelerates vertex-centric push-based graph applications by accelerating both their update (computing vertex updates) and apply (summing up the updates) phases. For the update phase, we design a tuple-based LLC that can coalesce at different granularities to group graph updates together and propose multi-DRAM column processing FGPIM instructions to match the cache coalescing to the row-level parallelism of the FGPIM. With this acceleration, the apply phase becomes the bottleneck, and we propose bank-parallel FGPIM instructions with predicates to allow FGPIM to accelerate the conditional updates as well. We achieve an average speedup in the region of interest of 1.8×/3× compared to naive FGPIM and 4.4×/9.8× compared to state-of-the-art non-PIM baseline (HBM2/DDR4), and DRAM energy reduction of 67%/86% and 88%/94%. These results show the importance of providing a complete solution that accelerates both the update and apply phases.
@inproceedings{YenimolN0B26,
title = {CoGraf: Fully Accelerating Graph Applications with Fine-Grained PIM},
author = {Ali Semi Yenimol and Anirban Nag and Chang Hyun Park and David Black-Schaffer},
booktitle = {ASPLOS},
year = {2026},
doi = {10.1145/3779212.3790142}
}