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 Tools annual

Global: Zendy Free Plan

Global: Zendy Tools monthly

Global: Zendy Plus monthly

In GPU graph analytics, the use of external memory such as the host DRAM andsolid-state drives is a cost-effective approach to processing large graphsbeyond the capacity of the GPU onboard memory. This paper studies the use ofCompute Express Link (CXL) memory as alternative external memory for GPU graphprocessing in order to see if this emerging memory expansion technology enablesgraph processing that is as fast as using the host DRAM. Through analysis andevaluation using FPGA prototypes, we show that representative GPU graphtraversal algorithms involving fine-grained random access can tolerate anexternal memory latency of up to a few microseconds introduced by the CXLinterface as well as by the underlying memory devices. This insight indicatesthat microsecond-latency flash memory may be used as CXL memory devices torealize even more cost-effective GPU graph processing while still achievingperformance close to using the host DRAM.

GPU Graph Processing on CXL-Based Microsecond-Latency External Memory