Fast SHA-256 Implementations on Intel® Architecture Processors
The paper describes a family of highly-optimized implementations of the Fast SHA-256 cryptographic hash algorithm, which provide industry leading performance on a range of Intel® processors for a single data buffer consisting of an arbitrary number of data blocks.
The paper describes the overall design of the Fast SHA-256 software, delves into some of the detailed optimizations, and presents a summary of the performance of some versions of the code. With our implementation a single core of an Intel® Core™ i7 processor 2600 with Intel® Hyper-Threading Technology can compute Fast SHA-256 of a large data buffer at the rate of ~11.5 cycles/byte.
Read the full Fast SHA-256 Implementations on Intel® Architecture Processors.