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/

Zendy Plus

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

Zendy Tools

Zendy Open

Graph Convolutional Networks (GCNs) have drawn tremendous attention in thepast three years. Compared with other deep learning modalities,high-performance hardware acceleration of GCNs is as critical but even morechallenging. The hurdles arise from the poor data locality and redundantcomputation due to the large size, high sparsity, and irregular non-zerodistribution of real-world graphs.  In this paper we propose a novel hardware accelerator for GCN inference,called I-GCN, that significantly improves data locality and reduces unnecessarycomputation. The mechanism is a new online graph restructuring algorithm werefer to as islandization. The proposed algorithm finds clusters of nodes withstrong internal but weak external connections. The islandization process yieldstwo major benefits. First, by processing islands rather than individual nodes,there is better on-chip data reuse and fewer off-chip memory accesses. Second,there is less redundant computation as aggregation for common/shared neighborsin an island can be reused. The parallel search, identification, and leverageof graph islands are all handled purely in hardware at runtime working in anincremental pipeline. This is done without any preprocessing of the graph dataor adjustment of the GCN model structure.  Experimental results show that I-GCN can significantly reduce off-chipaccesses and prune 38% of aggregation operations, leading to performancespeedups over CPUs, GPUs, the prior art GCN accelerators of 5549x, 403x, and5.7x on average, respectively.

I-GCN: A Graph Convolutional Network Accelerator with Runtime Locality Enhancement through Islandization