Multi-thread Acceleration Research Articles

SPARTA: High-Level Synthesis of Parallel Multi-Threaded Accelerators

SPARTA: High-Level Synthesis of Parallel Multi-Threaded Accelerators

Svelto: High-Level Synthesis of Multi-Threaded Accelerators for Graph Analytics

Graph analytics are an emerging class of irregular applications. Operating on very large datasets, they present unique behaviors, such as fine-grained, unpredictable memory accesses, and highly unbalanced task level parallelism, that make existing high-performance general-purpose processors or accelerators (e.g., GPUs) suboptimal. To address these issues, research and industry are developing a variety of custom accelerator designs for this application area, including solutions based on reconfigurable devices (Field Programmable Gate Arrays). These new approaches often employ High-Level Synthesis (HLS) to Speed up the development of the accelerators. In this paper, we propose a novel architecture template for the automatic generation of accelerators for graph analytics and irregular applications. The architecture template includes a dynamic task scheduling mechanism, a parallel array of accelerators that enables supporting task-level parallelism with context switching, and a related multi-channel memory interface that decouples communication from computation and provides support for fine-grained atomic memory operations. We discuss the integration of the architectural template in an HLS flow, presenting the necessary modifications to enable automatic generation of the custom architectures starting from OpenMP annotated code. We evaluate our approach first by synthesizing and exploring triangle counting, a common graph algorithm, and then by synthesizing custom designs for a set of graph database benchmark queries, representing series of graph pattern matching routines. We compare the synthesized accelerators with previous state-of-the-art methodologies for the synthesis of parallel architectures, showing that the proposed approach allows reducing resource usage by optimizing the number of accelerators replicas without any performance penalty.

DESIGN AND APPLICATION OF BOUNDING VOLUME HIERARCHY COLLISION DETECTION ALGORITHM BASED ON VIRTUAL SPHERE

The result of collision detection is closely related to the further deformation or cutting action of soft tissue. In order to further improve the efficiency and stability of collision detection, in this paper, a collision detection algorithm of bounding volume hierarchy based on virtual sphere was proposed. The proposed algorithm was validated and the results show that the detection efficiency of the bounding volume hierarchy algorithm based on virtual sphere is higher than that of the serial hybrid bounding volume hierarchy algorithm and the parallel hybrid bounding volume hierarchy algorithm. Different collision detection algorithms were tested and the results show that the collision detection algorithm based on virtual sphere has high detection efficiency and good stability. As the number of triangular patches increased, the advantage was more and more obvious. Finally, the proposed algorithm was applied to two large and medium-sized virtual scenes to implement the collision detection between the vastus lateralis muscle, thigh and surgical instrument. Based on the virtual sphere, the collision detection algorithm of bounding volume hierarchy can implement efficient and stable collision detection in a virtual surgery system. Meanwhile, the algorithm can be combined with other acceleration algorithms (such as the multithread acceleration algorithm) to further improve detection efficiency.

A Pipelined-Based Multi-Thread Acceleration Method for Remote Sensing Image Progressive Transmission

Haicheng Qu, Junping Zhang and Yu Meng Dept. of Information Engineering,Harbin Institute of Technology, Harbin, China Dept. of Software Engineering,Software College of Liaoning Technical University, Huludao,China Dept.of Computer Application Technology,College of Information Science and Engineering of Northeastern University,Shenyang, China haichengqu@hit.edu.cn, zhangjp@hit.edu.cn, perfectmeng@gmail.com