An approach based on algebraic treatment of programs and advanced transition system operational semantics is described for efficient synchronization and communication of parallel coarse-grained distributed/shared memory programs. The approach combines compile/run time data flow analysis and buffering tech- Niques and improves communication and synchronization features for these pro- Grams. This is accomplished in tow ways: 1) construction of a number of semantic Models of data exchanges of increasing power for asynchronous communication in Distributed environment; and 2) exploiting user-based specification of shared mem- Ory access order, known a priori, by means of synchronization facilities orthogonal To computation part of parallel programs that are more expressive and efficient than Barrier-like facilities.
展开▼
