Experiments with the fresh breeze tree-based memory model

Abstract The Fresh Breeze memory model and system ar-chitecture is proposed as an approach to achieving signif-icant improvements in massively parallel computation by supporting fine-grain management of memory and process-ing resources and utilizing a global shared name space for all processors and computation tasks. Memory management and the scheduling of tasks are done by hardware realiza-tions, eliminating nearly all operating system execution cy-cles for data access, task scheduling and security. In partic-ular, the Fresh Breeze memory model uses trees of fixed-size chunks of memory to represent all data objects, which eliminates data consistency issues and simplifies memory management. Low-cost reference-count garbage collection is used to support modular programming in type-safe pro-gramming languages. The main contributions of this paper are: (1) a pro-gram exection model for massively parallel computing as the Fresh Breeze application programming interface (API) comprising a radical memory model and a scheme for ex-pressing concurrency; (2) an experimental implementation of the API through simulation using the FAST simulator of the IBM Cyclops 64 many-core chip; (3) simulation result