Post by AimValley
5,590 followers
๐๐ฐ๐ฐ๐ฒ๐น๐ฒ๐ฟ๐ฎ๐๐ถ๐ป๐ด ๐ ๐ฅ๐-๐๐๐ถ๐ฑ๐ฒ๐ฑ ๐ฅ๐ฎ๐ฑ๐ถ๐ผ๐๐ต๐ฒ๐ฟ๐ฎ๐ฝ๐ ๐๐ถ๐๐ต ๐๐ฃ๐๐ ๐ฃ๐ผ๐๐ฒ๐ฟ ๐๐ถ๐ณ๐ฏ๐ช๐ฏ๐จ ๐ค๐ฐ๐ฎ๐ฑ๐ถ๐ต๐ข๐ต๐ช๐ฐ๐ฏ๐ข๐ญ ๐ฃ๐ฐ๐ต๐ต๐ญ๐ฆ๐ฏ๐ฆ๐ค๐ฌ๐ด ๐ช๐ฏ๐ต๐ฐ ๐ฃ๐ณ๐ฆ๐ข๐ฌ๐ต๐ฉ๐ณ๐ฐ๐ถ๐จ๐ฉ๐ด ๐ช๐ฏ ๐ต๐ณ๐ฆ๐ข๐ต๐ฎ๐ฆ๐ฏ๐ต ๐ด๐ฑ๐ฆ๐ฆ๐ฅ. Proud to share the results of our work on the ART-PIVOT project, in collaboration with UMC Utrecht and NWO (Dutch Research Council), focused on accelerating adaptive MRI-guided radiotherapy planning and reducing overall treatment planning time. ๐ง๐ต๐ฒ ๐ฐ๐ต๐ฎ๐น๐น๐ฒ๐ป๐ด๐ฒ UMC Utrechtโs MR-linac research pipeline runs entirely on CPU (MATLAB, FP64), where matrixโvector multiplication dominates compute time. With matrices reaching sizes of up to 3M ร 25K, the goal was to offload the most demanding kernels to an FPGA acceleratorโwithout compromising output quality. ๐ช๐ต๐ฎ๐ ๐๐ฒ ๐ฏ๐๐ถ๐น๐ On the AMD Versal VCK190 (VC1902, 400 AIE tiles), we developed two accelerators targeting the primary bottlenecks: โข A dense matrixโvector (GEMM) accelerator using an 8ร23 AIE tile layout, combined with a custom sort engine and add engine in the programmable logic โข A sparse matrixโvector (SpMM) accelerator featuring a novel compression scheme that eliminates zero-padding overhead and leverages the row-sparse structure of UMC Utrechtโs matrices ๐ง๐ต๐ฒ ๐ฟ๐ฒ๐๐๐น๐ Full-system VSS co-simulation shows a 4.16ร speed-up over the MATLAB baseline for GEMM on a 3M-row workload. Interestingly, the main bottleneck was not compute, but DDR read bandwidthโhighlighting that the AIE array still has headroom for further performance gains. ๐๐ฒ๐ ๐๐ฎ๐ธ๐ฒ๐ฎ๐๐ฎ๐ AMD Versal with AIE tiles proves to be a strong fit for floating-point, compute-intensive medical imaging workloads. While DDR bandwidth remains the primary architectural constraint, there is significant opportunity to exploit parallelism within those limits. Curious how advanced compute architectures like FPGA and AI Engines can accelerate your most demanding workloads? At AimValley, we turn complex algorithms into efficient, high-performance implementations. Letโs connect and explore whatโs possible. #umcu #nwo #mri #fpgaacceleration #fpgapower MathWorks #versal #AIengines AimValley is a Technolution Company!