Data-triggered Multithreading for Near-Data Processing
نویسندگان
چکیده
Data-centric computing becomes increasingly important because of the rapid growth of application data. In this work, we introduce the DTM (Data-Triggered Multithreading) programming model that extends the DTT (Data-Triggered Thread) model and is fully compatible with existing C/C++ programs. The DTM model naturally attaches computation to data. Therefore, the runtime system can dynamically allocate the computing resource that provides affinity and locality. We demonstrate the potential of DTM model to improve response time and improve scalability over the traditional multithreaded programming model.
منابع مشابه
A Multithreaded CGRA for Convolutional Neural Network Processing
Convolutional neural network (CNN) is an essential model to achieve high accuracy in various machine learning applications, such as image recognition and natural language processing. One of the important issues for CNN acceleration with high energy efficiency and processing performance is efficient data reuse by exploiting the inherent data locality. In this paper, we propose a novel CGRA (Coar...
متن کاملImproving Latency Tolerance of Network Processors Through Simultaneous Multithreading
Existing multithreaded network processors architecture with multiple processing engines (PEs), aims at taking advantage of blocked multithreading technique which executes instructions of different user-defined threads in the same PE pipeline, in explicit and interleave way. Multiple PEs, each of which is a multithreaded processor core, process several packets in parallel to hide long memory acc...
متن کاملQuantitative Analysis on Caching Effect of I-Structure Data in Frame-Based Multithreaded Processing
Since long latency due to remote memory access could be tolerated by rapidly switching to another thread in multithreaded processing, caching I-structure data is expected to have less benejicial effect on the pe$ormance than caching ordinary data. In this papel; we show that caching I-structure data could improve the overall performance in spite of latency tolerating property of multithreading....
متن کاملNew debugging concept for symmetric multiprocessing (SMP)
However, for the parallelization of tasks not necessarily a multi-core processor is required. Hardware multithreading, for example, is an approach that enables parallelization also for single-core processors. Here, we deal with a basic problem of cores with pipeline architecture: cache misses or data dependencies between the instructions mean that the pipelined instruction processing has to be ...
متن کاملA Study of Compiler-Directed Multithreading for Embedded Applications
Growing demand for high performance in embedded systems is creating new opportunities to leverage techniques such as pipelining and instruction-level parallel processing, which were originally developed for general-purpose processors. In this paper, we investigate the applicability of compiler-directed multithreading in speeding up embedded applications. In particular, we take programs from the...
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 2013