Communication-Computation overlap in massively parallel System on Chip
نویسندگان
چکیده
This paper presents the concept and the implementation of a communication-computation overlap in massively parallel System on Chip (mpSoC). This paradigm allows to decrease the execution time of parallel programs using specific strategies in the programming level and partially decoupled system control in the hardware level. The related experiment in VHDL language for system design and in assembly language for case study implementation are described.
منابع مشابه
Massively parallel neural computation
Reverse-engineering the brain is one of the US National Academy of Engineering’s “Grand Challenges.” The structure of the brain can be examined at many different levels, spanning many disciplines from low-level biology through psychology and computer science. This thesis focusses on real-time computation of large neural networks using the Izhikevich spiking neuron model. Neural computation has ...
متن کاملScalable communication architectures for massively parallel hardware multi-processors
Modern complex embedded applications inmultiple application fields impose stringent and continuously increasing functional and parametric demands. To adequately serve these applications, massively parallel multi-processor systems on a single chip (MPSoCs) are required. This paper is devoted to the design of scalable communication architectures of massively parallel hardware multi-processors for...
متن کاملHardware Support for Explicit Communication in Scalable CMP’s
Programming models with explicit communication between parallel tasks allow the runtime system to schedule task execution and data transfers ahead of time. Explicit communication is not limited to message passing and streaming applications: recent proposals in parallel programming allow such explicit communication in other task-based scenarios too. Scheduling of data transfers allows the overla...
متن کاملMassively Parallel Inner-Product Array Processor
We present a hardware architecture for parallel innerproduct array computation in very high dimensional feature spaces, towards a general-purpose kernel-based classiJer and function approximator: The architecture is internally analog with fully digital interface. On-chip analog jinegrain parallel processing yields real-time throughput levels for high-dimensional (over 1,000per chip) classificat...
متن کامل12 A Critical Analysis of MultigridMethods on Massively
The hierarchical nature of multigrid algorithms leaves domain parallel strategies with a deeciency of parallelism as the computation moves to coarser and coarser grids. To introduce more parallelism several strategies have been designed to project the original problem space into non-interfering subspaces, allowing all grids to relax concurrently. Our objective is to understand the potential eec...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2017