https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Title-AMFE-SIMD-Ready-Branchless-Filtering-for-High-Volume-Prime/m-p/1748763#M37597 <P>Hi Intel oneMKL team,</P><P>I developed AMFE, a lightweight branchless filtering layer for primality testing pipelines. It is designed to be SIMD-Ready and highly parallel.</P><P>Benchmark Results (10M RSA-grade candidates):<BR />• ~33% rejection rate of composite candidates<BR />• 2.2x speedup (4.25ms vs 10.43ms baseline)<BR />• 1.88B candidates/sec throughput<BR />• 0 false negatives</P><P>Technical Highlights:<BR />• Branchless LUT (no runtime modulo/division)<BR />• SIMD-Ready for AVX2/AVX-512<BR />• Rust + Rayon (parallel execution)<BR />• Production-ready, Dockerized</P><P>Technical PDF attached.</P><P>I'm looking for feedback from Intel performance engineers on:<BR />1. SIMD/AVX2 optimization opportunities<BR />2. Potential integration with Intel IPP Cryptography or oneMKL<BR />3. Performance tuning for Xeon Scalable processors</P><P>Thank you for your time.</P><P>Best regards,<BR />Azeddine Ennabati</P> Fri, 22 May 2026 01:18:03 GMT aze_ennabati 2026-05-22T01:18:03Z https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Title-AMFE-SIMD-Ready-Branchless-Filtering-for-High-Volume-Prime/m-p/1748763#M37597 <P>Hi Intel oneMKL team,</P><P>I developed AMFE, a lightweight branchless filtering layer for primality testing pipelines. It is designed to be SIMD-Ready and highly parallel.</P><P>Benchmark Results (10M RSA-grade candidates):<BR />• ~33% rejection rate of composite candidates<BR />• 2.2x speedup (4.25ms vs 10.43ms baseline)<BR />• 1.88B candidates/sec throughput<BR />• 0 false negatives</P><P>Technical Highlights:<BR />• Branchless LUT (no runtime modulo/division)<BR />• SIMD-Ready for AVX2/AVX-512<BR />• Rust + Rayon (parallel execution)<BR />• Production-ready, Dockerized</P><P>Technical PDF attached.</P><P>I'm looking for feedback from Intel performance engineers on:<BR />1. SIMD/AVX2 optimization opportunities<BR />2. Potential integration with Intel IPP Cryptography or oneMKL<BR />3. Performance tuning for Xeon Scalable processors</P><P>Thank you for your time.</P><P>Best regards,<BR />Azeddine Ennabati</P> Fri, 22 May 2026 01:18:03 GMT https://community.intel.com/t5/Intel-oneAPI-Math-Kernel-Library/Title-AMFE-SIMD-Ready-Branchless-Filtering-for-High-Volume-Prime/m-p/1748763#M37597 aze_ennabati 2026-05-22T01:18:03Z