New functionality in Intel MKL 11.0 now allows you to balance performance with reproducible results by allowing greater flexibility in code path choice and by ensuring that algorithms are deterministic. To learn more about Conditional Numerical Reproducibility (CNR) see the following resources:
- Webinar archive: Getting Reproducible Results with Intel(R) MKL 11.0
- Webinar archive: Run-to-Run Reproducibility with Intel® MKL and the Intel® Compilers
- Intel® MKL User's Guide
- Read the Parallel Universe article: Using the Intel® Math Kernel Library (Intel® MKL) and Intel® Compilers to Obtain Run-to-Run Numerical Reproducible Results
- Visit this CNR article for an introduction to the feature
- Answer this two question survey to influence how this feature will be extended in the future.
- Submit questions and feedback on the Intel MKL User Forum or Intel® Premier Support
All of Intel MKL supports this new capability except the following: the vsRngBeta random number generator on 32-bit operating systems.