Topic
Degree of parallelism
About: Degree of parallelism is a research topic. Over the lifetime, 1515 publications have been published within this topic receiving 25546 citations.
Papers published on a yearly basis
Papers
More filters
TL;DR: A new SISO-decoder architecture is proposed that leads to significant throughput gains and better hardware efficiency compared to existing architectures for the full range of code rates.
Abstract: Turbo decoders for modern wireless communication systems have to support high throughput over a wide range of code rates. In order to support the peak throughputs specified by modern standards, parallel turbo-decoding has become a necessity, rendering the corresponding VLSI implementation a highly challenging task. In this paper, we explore the implementation trade-offs of parallel turbo decoders based on sliding-window soft-input soft-output (SISO) maximum a-posteriori (MAP) component decoders. We first introduce a new approach that allows for a systematic throughput comparison between different SISO-decoder architectures, taking their individual trade-offs in terms of window length, error-rate performance and throughput into account. A corresponding analysis of existing architectures clearly shows that the latency of the sliding-window SISO decoders causes diminishing throughput gains with increasing degree of parallelism. In order to alleviate this parallel turbo-decoder predicament, we propose a new SISO-decoder architecture that leads to significant throughput gains and better hardware efficiency compared to existing architectures for the full range of code rates.
39 citations
TL;DR: This work proposes a light-weight asynchronous processing framework called Frog with a preprocessing/hybrid coloring model based on the Pareto principle about coloring algorithms, and finds that a majority of vertices are colored with only a few colors, such that they can be read and updated in a very high degree of parallelism without violating the sequential consistency.
Abstract: GPUs have been increasingly used to accelerate graph processing for complicated computational problems regarding graph theory. Many parallel graph algorithms adopt the asynchronous computing model to accelerate the iterative convergence. Unfortunately, the consistent asynchronous computing requires locking or atomic operations, leading to significant penalties/overheads when implemented on GPUs. As such, the coloring algorithm is adopted to separate the vertices with potential updating conflicts, guaranteeing the consistency/correctness of the parallel processing. Common coloring algorithms, however, may suffer from low parallelism because of a large number of colors generally required for processing a large-scale graph with billions of vertices. We propose a light-weight asynchronous processing framework called Frog with a preprocessing/hybrid coloring model. The fundamental idea is based on the Pareto principle (or 80-20 rule) about coloring algorithms as we observed through masses of real-world graph coloring cases. We find that a majority of vertices (about 80 percent) are colored with only a few colors, such that they can be read and updated in a very high degree of parallelism without violating the sequential consistency. Accordingly, our solution separates the processing of the vertices based on the distribution of colors. In this work, we mainly answer three questions: (1) how to partition the vertices in a sparse graph with maximized parallelism, (2) how to process large-scale graphs that cannot fit into GPU memory, and (3) how to reduce the overhead of data transfers on PCIe while processing each partition. We conduct experiments on real-world data (Amazon, DBLP, YouTube, RoadNet-CA, WikiTalk, and Twitter) to evaluate our approach and make comparisons with well-known non-preprocessed (such as Totem, Medusa, MapGraph, and Gunrock) and preprocessed (Cusha) approaches, by testing four classical algorithms (BFS, PageRank, SSSP, and CC). On all the tested applications and datasets, Frog is able to significantly outperform existing GPU-based graph processing systems except Gunrock and MapGraph. MapGraph gets better performance than Frog when running BFS on RoadNet-CA. The comparison between Gunrock and Frog is inconclusive. Frog can outperform Gunrock more than 1.04X when running PageRank and SSSP, while the advantage of Frog is not obvious when running BFS and CC on some datasets especially for RoadNet-CA.
38 citations
23 Sep 1994
TL;DR: Two methods for synthesis of VHDL specifications containing concurrent processes are presented to preserve simulation/synthesis correspondence during high-level synthesis and to produce hardware that operates with a high degree of parallelism.
Abstract: This paper presents two methods for synthesis of VHDL specifications containing concurrent processes. Our main objective is to preserve simulation/synthesis correspondence during high-level synthesis and to produce hardware that operates with a high degree of parallelism. The first method supports an unrestricted use of signals and wait statements and synthesizes synchronous hardware with global control of process synchronization for signal update. The second method allows hardware synthesis without the strict synchronization imposed by the VHDL simulation cycle. Experimental results have shown that the proposed methods are efficient for a wide spectrum of digital systems.
38 citations
TL;DR: An analytic modeling technique based on the use of Markovian Agents and Mean Field Analysis is proposed that allows the effective description of different concurrent Big Data applications on a same, multi‐site cloud infrastructure, accounting for mutual interactions, in order to support the careful evaluation of several elements in terms of real costs/risks/benefits.
Abstract: Big Data applications are characterized by a non-negligible number of complex parallel transactions on a huge amount of data that continuously varies, generally increasing over time. Because of the amount of needed resources, the ideal runtime scenario for these applications is based on complex cloud computing and storage infrastructures, providing a scalable degree of parallelism together with isolation between different applications and resource abstraction. However, such additional abstraction degree also introduces significant complexity in performance modeling and decision making. Potential concurrency of many applications on the same cloud infrastructure has to be evaluated, and, simultaneously, scalability of applications over time has to be studied through proper modeling practices, in order to predict the system behavior as the usage patterns evolve and the load increases. For this purpose, in this paper, we propose an analytic modeling technique based on the use of Markovian Agents and Mean Field Analysis that allows the effective description of different concurrent Big Data applications on a same, multi-site cloud infrastructure, accounting for mutual interactions, in order to support the careful evaluation of several elements in terms of real costs/risks/benefits for correctly dimensioning and allocating the resources and verifying the existing service level agreements. Copyright © 2014 John Wiley & Sons, Ltd.
38 citations
TL;DR: The goal for optical information processing is to use the unique characteristics of light which are not readily achieved with electronic devices, namely, ultrahigh speed (picoseconds), a high degree of parallelism (image processing), and conductor-free interconnection.
Abstract: The goal for optical information processing is to use the unique characteristics of light which are not readily achieved with electronic devices, namely, ultrahigh speed (picoseconds), a high degree of parallelism (image processing), and conductor-free interconnection. The requirements of the nonlinear materials to perform such functions, using all-optical interactions, are discussed and the limitations of the nonlinear mechanisms are outlined.
38 citations