Heterogeneous computing denotes a scenario where different computing platforms are exploited for specific applications. While the demand for computational resources continues to grow with increasing need for querying and analyzing the volumes and rates of Big Data, energy efficiency is limiting the traditional approach to improve the compute capabilities of a data center by adding thousands of state-of-the-art x86 machines to an existing infrastructure in favor of adopting accelerators. The result is that the computing nodes in data centers have different execution models, ranging from the traditional x68 architecture to GPUs, FPGA and then even other processor types like the ARM ones or more specialized processors as TPUs. For example, GPUs are used a lot in deep learning, especially the training part. They are also used in many scientific applications based on regular domains and are delivering performance that is orders of magnitude better than traditional cores. The FPGA instead tries to close the gap between hardware and software. They are circuits that can be configured by the programmer to implement a certain function. In this Research Topic we are interested in use cases, methodological approaches, etc. discussing the pros and cons of adopting heterogeneous architectures for AI and Big Data applications in High Energy Physics.
Heterogeneous computing denotes a scenario where different computing platforms are exploited for specific applications. While the demand for computational resources continues to grow with increasing need for querying and analyzing the volumes and rates of Big Data, energy efficiency is limiting the traditional approach to improve the compute capabilities of a data center by adding thousands of state-of-the-art x86 machines to an existing infrastructure in favor of adopting accelerators. The result is that the computing nodes in data centers have different execution models, ranging from the traditional x68 architecture to GPUs, FPGA and then even other processor types like the ARM ones or more specialized processors as TPUs. For example, GPUs are used a lot in deep learning, especially the training part. They are also used in many scientific applications based on regular domains and are delivering performance that is orders of magnitude better than traditional cores. The FPGA instead tries to close the gap between hardware and software. They are circuits that can be configured by the programmer to implement a certain function. In this Research Topic we are interested in use cases, methodological approaches, etc. discussing the pros and cons of adopting heterogeneous architectures for AI and Big Data applications in High Energy Physics.