Abstract:CPU-FPGA heterogeneous computing platforms have shown great potential for high performance, energy efficiency, and reconfigurability, and they are being built more and more for supporting various compute-intensive applications in both academia and industry. The task mapping on the heterogeneous processors is a key factor affecting computing performance in heterogeneous platforms. However, the previous methods do not incorporate the unique features for the workload executed on the CPU-FPGA heterogeneous computing platforms, and this induces the underutilized computing capability and load imbalance. In this paper, we propose a novel feature-aware task scheduling method for heterogeneous computing platforms with multicore CPUs and multiple FPGAs. First, we conduct the in-depth analysis of workload features in the CPU-FPGA platforms, and present a task speedup learning method based on the static and dynamic features extracted from the resource capabilities, code structure and running task description. Second, the tasks are assigned to the heterogeneous processors according to the speedup estimation and runtime load balancing strategy. Finally, we implement the typical heterogeneous workloads to evaluate our method, and the experimental results show the significant performance gain compared to the popular methods.

Detecting Data Hazards in Multi-Processor System-on-Chips on FPGA

Hardware Implementation on FPGA for Task-Level Parallel Dataflow Execution Engine.

A Method to Detect Hazards in Pipeline Processor

Architecture Support for Task Out-of-Order Execution in MPSoCs

Static or Dynamic: Trade-Offs for Task Dependency Analysis for Heterogeneous MPSoC

ASIP Performance Enhancement by Hazard Control through Scoreboard

Evaluation and Tradeoffs for Out-of-Order Execution on Reconfigurable Heterogeneous MPSoC

Minimizing Bank Conflict Delay for Real-Time Embedded Multicore Systems via Bank Mapping.

A novel dynamic scheduling algorithm of data hazard for embedded processor

Automatic Cache Partitioning and Time-Triggered Scheduling for Real-Time MPSoCs

System Structure for FPGA-based SOPC Design Using Hard Tasks

Data Fusion with Genetic Algorithm Based Lifetime Prediction for Dependable Multi-Processor System-on-Chips

FPGA Implementation of a Scheduler Supporting Parallel Dataflow Execution

Parallel dataflow execution for sequential programs on reconfigurable hybrid MPSoCs

Feature-Aware Task Scheduling on CPU-FPGA Heterogeneous Platforms

Multiprocessor System-on-Chip Design with Fuzzy Dynamic Scheduling for Parallel Video Processing

An FPGA-GPU Heterogeneous System and Implementation for On-Board Remote Sensing Data Processing

Automatic Data Placement For Cpu-Fpga Heterogeneous Multiprocessor System-On-Chips

A Hybrid Approach to Cache Management in Heterogeneous CPU-FPGA Platforms.

Resource-Aware Task Scheduling and Placement in Multi-FPGA System.

Parallelizing Workload Execution in Embedded and High-Performance Heterogeneous Systems