English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Crossbar-based PIM DNN accelerators can provide massively parallel in-situoperations. A specifically designed compiler is important to achieve highperformance for a wide variety of DNN workloads. However, some key compilationissues such as parallelism considerations, weight replication selection, andarray mapping methods have not been solved. In this work, we propose PIMCOMP -a universal compilation framework for NVM crossbar-based PIM DNN accelerators.PIMCOMP is built on an abstract PIM accelerator architecture, which iscompatible with the widely used Crossbar/IMA/Tile/Chip hierarchy. On thisbasis, we propose four general compilation stages for crossbar-based PIMaccelerators: node partitioning, weight replicating, core mapping, and dataflowscheduling. We design two compilation modes with different inter-layer pipelinegranularities to support high-throughput and low-latency application scenarios,respectively. Our experimental results show that PIMCMOP yields improvements of1.6$\times$ and 2.4$\times$ in throughput and latency, respectively, relativeto PUMA.

PIMCOMP: A Universal Compilation Framework for Crossbar-based PIM DNN Accelerators