Title: Accelerated Path to Heterogenous Computing – A SYCLOMatic Way!
While hardware innovation has led to a diverse heterogeneous architectural landscape for computing, software development has become increasingly complex, making it difficult to take full advantage of CPUs and accelerators. Today’s developers and their teams are strapped for time, money and resources to accommodate the rewriting and testing of code to boost application performance for these different architectures. Developers are looking for open alternatives that improve time-to-value, and Intel is providing an easier, shorter pathway to enabling hardware choice.
To free developers from a single-vendor proprietary ecosystem and to accelerate cross-architecture programming for heterogeneous architectures so as to harness the full power of new hardware innovations, a unified programming model is required that’s both built on standards and is extensible and that enables the developers in selecting the optimal hardware for the task in hand. SYCL addresses these challenges by extending C++ capabilities to support multiarchitecture and disjoint memory configurations.
To make it easier for developers to adapt SYCL and to advance the migration capabilities for producing more SYCL-based applications by allowing the reuse of code across architectures, the SYCLomatic project assists developers in porting CUDA code to SYCL. Hosted on GitHub and being open source the SYCLomatic project enables community collaboration to advance adoption of the SYCL standard and offers a community for developers to contribute and provide feedback to advance the tool’s evolution thus further advancing heterogeneous development across CPUs, GPUs and FPGAs.
Speaker’s Bio and Pic
Jyotsna Khemka is a Software Engineering Manager at Intel. She has got more than 18 years of work experience in software application development, applied research in parallel systems, and High-Performance Computing. She has a passion for optimizing & scaling real-world applications on large-scale parallel & distributed systems. In her current role, she is responsible to enable and support strategic customers in APJ region, in optimizing their applications on heterogenous platforms using Intel Software Tools. She holds a Master’s degree in Computational Engineering from Friedrich Alexander University, Germany and specialization on Computational Fluid Dynamics.