Designers and engineers invariably want more from their simulations—more detail, more variables, greater accuracy, and faster time to results. ANSYS Fluent* 18.1 and Intel® Xeon® Scalable processors address these needs by delivering powerful performance gains for engineering simulations, including per-core gains that help to speed performance while containing software licensing costs.
A Major Leap in Simulation Performance
The Intel® Xeon® Gold 6148 processor includes more cores, higher memory bandwidth, and an enhanced cache structure compared to the previous-generation Intel® Xeon® processor E5 v4 product family. ANSYS and Intel worked together to optimize ANSYS Fluent 18.1 for these and other new hardware features, using Intel® software development products to help ensure that the additional processing power delivers meaningful performance gains for real-world simulations.
“ANSYS teamed with Intel to make sure software and hardware improvements go hand in hand. The latest combination of ANSYS Fluent* 18.1 and Intel® Xeon® Gold 6148 processor is a clearly testament of impressive overall performance gains achieved for customers who want to increase their engineering productivity.”
A key focus of the optimization effort was to improve vectorization in the solver code to take better advantage of the advanced vector processing capabilities of Intel® Xeon® processors. The improved vectorization, which is available as a runtime option in Fluent 18.1, was used in these benchmarks.
To verify performance with the new processors, benchmark tests were run using a variety of models targeting different industries and ranging in size from two million to 33 million cells. Results showed that a two-socket server based on the new Intel® Xeon® Gold 6148 processor can improve performance for ANSYS Fluent by as much as 41 percent1 versus a previous-generation server based on the Intel® Xeon® processor E5-2697 v4, and by as much as 60 percent1 versus a comparable server based on the earlier Intel® Xeon® processor E5-2698 v3.
Significant Per-Core Performance Improvement
Tests were also run to measure per-core performance across each of the three processor generations (using 32 cores in each test). The Intel® Xeon® Gold 6148 processor provided up to 34 percent1 higher performance per core than the previous generation Intel® Xeon® processor E5-2697 v4 and up to 46 percent1 higher performance per core than the Intel® Xeon® processor E5-2698 v3. Because ANSYS licenses its software based on the number of cores in use, these results indicate the types of gains customers can potentially achieve without increasing their licensing costs.
Efficient Cluster Scaling to Support the Most Demanding CFD Models
Intel® MPI is integrated into the ANSYS Fluent release. Together, ANSYS Fluent and Intel® MPI are designed to provide high performance that scales seamlessly from multi-core workstations to clusters with thousands of cores (the benchmark results demonstrate performance in single-node scenarios).
For customers moving to clustered architectures, Intel® Omni-Path Architecture (Intel® OPA) provides a high performance, low-latency fabric that helps to resolve the performance, scalability, and cost challenges of traditional InfiniBand* solutions.
Take the Next Step
With ANSYS 18.1, the Intel® Xeon® Gold 6148 processor, and Intel OPA, engineering and design teams can get higher value from their engineering simulations today, and scale their computing infrastructure as needed to maintain fast runtimes as their models grow in complexity.