Efficient Run-Time System Support for High Performance Reliable Reconfigurable Systems

2012 ◽  
Vol 1 (2) ◽  
pp. 213-219
Author(s):  
Chuan Hong ◽  
Khaled Benkrid ◽  
Xabier Iturbe ◽  
Hanaa Hussain
Keyword(s):  
High Performance ◽  
Reconfigurable Systems ◽  
System Support ◽  
Time System ◽  
Run Time

scholarly journals An Efficient Framework for Floor-plan Prediction of Dynamic Runtime Reconfigurable Systems

2015 ◽  
Vol 4 (2) ◽  
pp. 99
Author(s):  
Ahmed Al-Wattar ◽  
Shawki Areibi ◽  
Gary Grewal
Keyword(s):  
Operating System ◽  
High Performance ◽  
Reconfigurable Systems ◽  
Floor Plan ◽  
Reconfigurable Logic ◽  
Machine Learning Approach ◽  
Hidden Knowledge ◽  
Run Time ◽  
Time Systems ◽  
Run Time Reconfiguration

<p>Several embedded application domains for reconfigurable systems tend to combine <br />frequent changes with high performance demands of their workloads such as image processing, wearable computing and<br />network processors.  Time multiplexing of reconfigurable hardware resources raises a number of new issues, ranging <br />from run-time systems to complex programming models that usually form a Reconfigurable<br />hardware Operating System (ROS).  The Operating System performs online task scheduling and handles resource management.<br />There are many challenges in adaptive computing and dynamic reconfigurable systems. One of the major understudied challenges<br />is estimating the required resources in terms of soft cores, Programmable Reconfigurable Regions (PRRs), <br />the appropriate communication infrastructure, and to predict a near optimal layout and floor-plan of the reconfigurable logic fabric. <br />Some of these issues are specific to the application being designed, while others are more general and relate to the underlying run-time environment.<br />Static resource allocation for Run-Time Reconfiguration (RTR) often leads to inferior and unacceptable results. <br />In this paper, we present a novel adaptive and dynamic methodology, based on a Machine Learning approach, for predicting and<br />estimating the necessary resources for an application based on past historical information.<br />An important feature of the proposed methodology is that the system is able to learn and generalize and, therefore, is expected to improve <br />its accuracy over time.  The goal of the entire process is to extract useful hidden knowledge from the data. This knowledge is the prediction <br />and estimation of the necessary resources for an unknown or not previously seen application.<br /><br /></p>

scholarly journals An Efficient Evolutionary Task Scheduling/Binding Framework for Reconfigurable Systems

2016 ◽  
Vol 2016 ◽  
pp. 1-24
Author(s):  
A. Al-Wattar ◽  
S. Areibi ◽  
G. Grewal
Keyword(s):  
High Performance ◽  
Computation Time ◽  
Reconfigurable Systems ◽  
Random Allocation ◽  
Early Design Stages ◽  
Run Time ◽  
Reconfiguration Overhead ◽  
Reconfigurable Platform ◽  
Performance Area ◽  
Time Systems

Several embedded application domains for reconfigurable systems tend to combine frequent changes with high performance demands of their workloads such as image processing, wearable computing, and network processors. Time multiplexing of reconfigurable hardware resources raises a number of new issues, ranging from run-time systems to complex programming models that usually form a reconfigurable operating system (ROS). In this paper, an efficient ROS framework that aids the designer from the early design stages all the way to the actual hardware implementation is proposed and implemented. An efficient reconfigurable platform is implemented along with novel placement/scheduling algorithms. The proposed algorithms tend to reuse hardware tasks to reduce reconfiguration overhead, migrate tasks between software and hardware to efficiently utilize resources, and reduce computation time. A supporting framework for efficient mapping of execution units to task graphs in a run-time reconfigurable system is also designed. The framework utilizes an Island Based Genetic Algorithm flow that optimizes several objectives including performance, area, and power consumption. The proposed Island Based GA framework achieves on average 55.2% improvement over a single-GA implementation and an 80.7% improvement over a baseline random allocation and binding approach.

SARANA: language, compiler and run-time system support for spatially aware and resource-aware mobile computing

2008 ◽  
Vol 366 (1881) ◽  
pp. 3699-3708 ◽  
Author(s):  
Pradip Hari ◽  
Kevin Ko ◽  
Emmanouil Koukoumidis ◽  
Ulrich Kremer ◽  
Margaret Martonosi ◽  
...  
Keyword(s):  
Mobile Computing ◽  
System Development ◽  
Cost Model ◽  
Personal Digital Assistants ◽  
System Support ◽  
Time System ◽  
Systems Research ◽  
Trade Offs ◽  
Run Time ◽  
Resource Aware

Increasingly, spatial awareness plays a central role in many distributed and mobile computing applications. Spatially aware applications rely on information about the geographical position of compute devices and their supported services in order to support novel functionality. While many spatial application drivers already exist in mobile and distributed computing, very little systems research has explored how best to program these applications, to express their spatial and temporal constraints, and to allow efficient implementations on highly dynamic real-world platforms. This paper proposes the SARANA system architecture, which includes language and run-time system support for spatially aware and resource-aware applications. SARANA allows users to express spatial regions of interest, as well as trade-offs between quality of result (QoR), latency and cost. The goal is to produce applications that use resources efficiently and that can be run on diverse resource-constrained platforms ranging from laptops to personal digital assistants and to smart phones. SARANA's run-time system manages QoR and cost trade-offs dynamically by tracking resource availability and locations, brokering usage/pricing agreements and migrating programs to nodes accordingly. A resource cost model permeates the SARANA system layers, permitting users to express their resource needs and QoR expectations in units that make sense to them. Although we are still early in the system development, initial versions have been demonstrated on a nine-node system prototype.

Sign in / Sign up

Export Citation Format

Share Document