Interleaved generic-point parallel scalar multiplication

This paper presents an efficient scheme for generic-point postcomputation-based parallel scalar multiplication. The proposed scheme processes more than one request simultaneously by interleaving parallel scalar multiplications. The results show that the proposed scheme outperforms the previous postcomputation-based methods when the consecutive requests number is eight or more. Furthermore, the proposed scheme is scalable for any number of parallel processors and performs better as the number of consecutive requests increases. Accordingly, the proposed scheme is very attractive for use in high-performance end servers that use parallel elliptic curve processors.