Performance Evaluation and Modeling of Cryptographic Libraries for MPI Communications
This work addresses data security for HPC applications in cloud environments, but it is incremental as it focuses on empirical evaluation and modeling of existing cryptographic libraries.
The paper tackled the problem of securing HPC applications in public clouds by evaluating the performance of four cryptographic libraries (OpenSSL, BoringSSL, Libsodium, CryptoPP) integrated into MPI communications, finding that encryption incurs overheads and providing models to guide performance improvements.
In order for High-Performance Computing (HPC) applications with data security requirements to execute in the public cloud, the cloud infrastructure must ensure the privacy and integrity of data. To meet this goal, we consider incorporating encryption in the Message Passing Interface (MPI) library. We empirically evaluate four contemporary cryptographic libraries, OpenSSL, BoringSSL, Libsodium, and CryptoPP using micro-benchmarks and NAS parallel benchmarks on two different networking technologies, 10Gbps Ethernet and 40Gbps InfiniBand. We also develop accurate models that allow us to reason about the performance of encrypted MPI communication in different situations and give guidance on how to improve encrypted MPI performance.