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

Xilinx's AI Engine is a recent industry example of energy-efficient vectorprocessing that includes novel support for 2D SIMD datapaths and shuffleinterconnection network. The current approach to programming the AI Enginerelies on a C/C++ API for vector intrinsics. While an advance overassembly-level programming, it requires the programmer to specify a number oflow-level operations based on detailed knowledge of the hardware. To addressthese challenges, we introduce Vyasa, a new programming system that extends theHalide DSL compiler to automatically generate code for the AI Engine. Weevaluated Vyasa on 36 CONV2D and 6 CONV3D workloads, and achieved geometricmeans of 7.6 and 23.3 MACs/cycle for 32-bit and 16-bit operands (whichrepresent 95.9% and 72.8% of the peak performance respectively). For 4 of theseworkloads for which expert-written codes were available to us, Vyasademonstrated a geometric mean performance improvement of 1.10x with 50x smallercode relative to the expert-written codes.

Vyasa: A High-Performance Vectorizing Compiler for Tensor Convolutions on the Xilinx AI Engine