Intel® Integrated Performance Primitives Cryptography Multi-buffer

 Some of the cryptography algorithms cannot be parallelized due to
 their specificity (eg the RSA algorithm).
 The way to get high-performance implementations of such cryptographic
 algorithms is a parallel processing using *single instruction,
 multiple data* (SIMD) architecture. The usage model implies
 presence of several independent and homogeneous (encryption or decryption)
 requests for RSA operations, using of SIMD leads to performance
 improvement.
 .
 Multi-buffer processing collects up to eight RSA operations.
 Each request is computed independently from the others, so it
 is possible to process several packets at the same time.
 .
 This library consists of highly-optimized kernels taking advantage
 of Intel’s multi-buffer processing and Intel® AVX-512 instruction set.