Since the beginning of 2016, Solvency II is in force in the European Union – harmonising the regulation for insurers with an economic and risk-based approach. Now, life insurers and pension funds need to calculate the economic value of liabilities for long-term contracts regularly, considering scenarios such as movements of interest rates or changes in mortality. This is computationally demanding and insurers are looking for ways to accelerate their implementations for Solvency II.
This white paper analyses typical implementations and highlights opportunities for optimisation – including parallelism, vectorisation, and data access efficiency. It finds that typical portfolios are suitable for execution in parallel on CPU as well as on accelerator processors such as GPUs or Intel’s Xeon Phi. The paper gives an example implementation and shows that significant performance gains are achievable.