Conference material: "Scientific service & Internet: proceedings of the 25th All-Russian Scientific Conference (September 18-21, 2023, online)"
On the creation of parallelizing compilers for distributed memory computing systems
This work describes the conditions for creating optimizing parallelizing compilers for computing systems with distributed memory. Optimizing program transformations as specific to distributed systems memory, and converting programs that are needed for distributed memory computing systems and can improve compilers for computer systems and shared memory. Are given arguments in favor of creating parallelizing compilers for distributed memory computing systems follows based on a high-level internal representation and with high-level output language.
parallelization automation, distributed memory, program conversions, data placement, transfers data