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

Graph neural networks (GNNs) start to gain momentum after showing significantperformance improvement in a variety of domains including molecular science,recommendation, and transportation. Turning such performance improvement ofGNNs into practical applications relies on effective and efficient execution,especially for inference. However, neither CPU nor GPU can meet these needs ifconsidering both performance and energy efficiency. That's because acceleratingGNNs is challenging due to their excessive memory usage and arbitraryinterleaving of diverse operations. Besides, the semantics gap between thehigh-level GNN programming model and efficient hardware makes it difficult inaccelerating general-domain GNNs.  To address the challenge, we propose Zipper, an efficient yet generalacceleration system for GNNs. The keys to Zipper include a graph-nativeintermediate representation (IR) and the associated compiler. By capturing GNNprimitive operations and representing with GNN IR, Zipper is able to fit GNNsemantics into hardware structure for efficient execution. The IR also enablesGNN-specific optimizations including sparse graph tiling and redundantoperation elimination. We further present an hardware architecture designconsisting of dedicated blocks for different primitive operations, along with arun-time scheduler to map a IR program to the hardware blocks. Our evaluationshows that Zipper achieves 93.6x speedup and 147x energy reduction over IntelXeon CPU, and 1.56x speedup and 4.85x energy reduction over NVIDIA V100 GPU onaverages.

ZIPPER: Exploiting Tile- and Operator-level Parallelism for General and Scalable Graph Neural Network Acceleration