Abstract:Architectures based on the data flow computing model provide an alternative to the conventional Von-Neumann architecture that are widely used for general purpose computing. Processors based on the data flow architecture employ fine-grain data-driven parallelism. These architectures have the potential to exploit the inherent parallelism in compute intensive applications like signal processing, image and video processing and so on and can thus achieve faster throughputs and higher power efficiency. In this paper, several data flow computing architectures are explored, and their main architectural features are studied. Furthermore, a classification of the processors is presented based on whether they employ either the data flow execution model exclusively or in combination with the control flow model and are accordingly grouped as exclusive data flow or hybrid architectures. The hybrid category is further subdivided as conjoint or accelerator-style architectures depending on how they deploy and separate the data flow and control flow execution model within their execution blocks. Lastly, a brief comparison and discussion of their advantages and drawbacks is also considered. From this study we conclude that although the data flow architectures are seen to have matured significantly, issues like data-structure handling and lack of efficient placement and scheduling algorithms have prevented these from becoming commercially viable.

Microgrid - The microthreaded many-core architecture

Design space exploration in the microthreaded many-core architecture

The Data-Driven Multithreading Programming Model Based On Ddf

The microarchitecture of a multi-threaded RISC-V compliant processing core family for IoT end-nodes

Novel many-core architecture design for real-time image processing

High level programming abstractions for leveraging hierarchical memories with micro-core architectures

The performance model of Hyper-Threading Technology in Intel Nehalem microarchitecture

Multi-core: Adding a New Dimension to Computing

Heterothread: Hybrid Thread Level Parallelism on Heterogeneous Multicore Architectures.

Multithreaded Computing Model,Architecture and Compiling Technique

A hierarchical reconfigurable micro-coded multi-core processor for IoT applications

FuMicro: A Fused Microarchitecture Design Integrating In-Order Superscalar and VLIW

Exploring the Approaches to Data Flow Computing

Programming Bare-Metal Accelerators with Heterogeneous Threading Models: A Case Study of Matrix-3000

Microarchitectural comparison and in-core modeling of state-of-the-art CPUs: Grace, Sapphire Rapids, and Genoa

Research on Formal Design of Multi-thread Mechanism Based on Microkernel Architecture

Characterizing Fine-Grain Parallelism on Modern Multicore Platform

An evaluation of a microprocessor with two independent hardware execution threads coupled through a shared cache

Characteristics of multithreading models for high-performance IO driven network applications

An 800mhz 320mw 16-Core Processor with Message-Passing and Shared-Memory Inter-Core Communication Mechanisms

Compiler-directed scratchpad memory data transfer optimization for multithreaded applications on a heterogeneous many-core architecture