STAC Webinar: Four steps to accelerate Monte Carlo calculations
In this 30-minute webinar, Intel engineers describe four steps you can take to improve your Monte Carlo performance today. This includes specific code optimizations from Intel's experience with the STAC-A2 Benchmark suite (Heston-based Greeks for a path-dependent, multi-asset option with early exercise) and performance comparisons of Sandy Bridge, Ivy Bridge, and Xeon Phi processors.
Below the webinar recording are a link to the slides, along with other links to more information.
• Learn more about STAC-A2
• Review Intel's STAC-A2 Reports
• Request the STAC-A2 Pack for Intel Composer XE (benchmark implementation code)*
• Request the inital (2012) version of the Intel STAC-A2 implementation.* (This version is superseded by the STAC Pack above. It's available for developers who wish to study the optimizations in the latter code relative to the former.
• Get information about Intel Server Products
• Get information on Intel Composer XE
* For firms with a premium subscription. Contact STAC for information on your firm's membership status.
NOTE: The slides below have one update since the webinar. The results in slide 11 for 25,000 paths have been updated to correspond to the baseline GREEKS.TIME benchmarks for the respective platforms. GREEKS.TIME reflects the mean of 5 test runs, whereas the results for the other problem sizes are from just one test run. We've updated these results to avoid confusion when comparing this table to the STAC Report Card in the STAC Report for each platform.
If you do not see an attachment below, please log in. If you are not registered, you may register for no charge by clicking here.