Expander development continues to advance with steady optimization efforts. Here’s this week’s update from the engineering team on FPGA acceleration for Expander.
2/
GF2-128 Hardware Performance: Completed the hardened hardware implementation of the GF2-128 arithmetic pipeline, achieving stable operation at 277 MHz on FPGA.
3/
End-to-End Proof Throughput: Achieved 75,000–77,000 proofs per second for the Keccak-based circuit in end-to-end FPGA testing, demonstrating substantial gains in overall proof generation throughput.
4/
NoC Performance Limitation: Current system performance is constrained by the on-chip Network-on-Chip (NoC) bandwidth, which is limited to around 250 MHz.
5/
Ongoing Optimization: Began exploring the use of soft NoC resources to improve data movement efficiency and relax routing congestion, with the goal of enabling higher sustained frequencies for large-scale proof workloads.
Stay tuned for more updates.
3,018
16
本页面内容由第三方提供。除非另有说明，欧易不是所引用文章的作者，也不对此类材料主张任何版权。该内容仅供参考，并不代表欧易观点，不作为任何形式的认可，也不应被视为投资建议或购买或出售数字资产的招揽。在使用生成式人工智能提供摘要或其他信息的情况下，此类人工智能生成的内容可能不准确或不一致。请阅读链接文章，了解更多详情和信息。欧易不对第三方网站上的内容负责。包含稳定币、NFTs 等在内的数字资产涉及较高程度的风险，其价值可能会产生较大波动。请根据自身财务状况，仔细考虑交易或持有数字资产是否适合您。