livre blanc

FPGAs in the Storm

FPGA prototype platforms make extensive use of re-programmable logic provided by embedded LUTs to virtually map any design, regardless of size, architecture and target application.
Deep learning acceleration could serve machine learning training in the data center, inference in the data center or inference at the edge. (Source: Barclays Research)

For FPGA providers to effectively manage their business they wrestle with two unique and non-overlapping use cases: support for FPGA prototype platforms (verification of SoCs) and support for the deep learning acceleration market (proliferation of AI application). Unfortunately, an FPGA roadmap to satisfy both use cases is irreconcilable, leaving the FPGA vendor in a quandary.

Partager