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 Tools annual

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 monthly

Global: Zendy Plus monthly

Presents the access of premium content as premium feature

Premium Content

Global: Zendy Plus annual

Global: Zendy Free Plan

Molecular dynamics (MD) simulation is one of the past decade's most importanttools for enabling biology scientists and researchers to explore human healthand diseases. However, due to the computation complexity of the MD algorithm,it takes weeks or even months to simulate a comparatively simple biology entityon conventional multicore processors. The critical path in molecular dynamicssimulations is the force calculation between particles inside the simulatedenvironment, which has abundant parallelism. Among various accelerationplatforms, FPGA is an attractive alternative because of its low power and highenergy efficiency. However, due to its high programming cost using RTL, none ofthe mainstream MD software packages has yet adopted FPGA for acceleration.  In this paper we revisit the FPGA acceleration of MD in high-level synthesis(HLS) so as to provide affordable programming cost. Our experience with the MDacceleration demonstrates that HLS optimizations such as loop pipelining,module duplication and memory partitioning are essential to improve theperformance, achieving a speedup of 9.5X compared to a 12-core CPU. Moreimportantly, we observe that even the fully optimized HLS design can still be2X slower than the reference RTL architecture due to the common dynamic(conditional) data flow behavior that is not yet supported by current HLStools. To support such behavior, we further customize an array of processingelements together with a data-driven streaming network through a common RTLtemplate, and fully automate the design flow. Our final experimental resultsdemonstrate a 19.4X performance speedup and 39X energy efficiency for thewidely used ApoA1 MD benchmark on the Convey HC1ex FPGA compared to a 12-coreIntel Xeon server.

Revisiting FPGA Acceleration of Molecular Dynamics Simulation with Dynamic Data Flow Behavior in High-Level Synthesis