Exploiting Partial Runtime Reconfiguration for High-Performance Reconfigurable Computing

@article{ElAraby2009ExploitingPR,
  title={Exploiting Partial Runtime Reconfiguration for High-Performance Reconfigurable Computing},
  author={Esam El-Araby and Iv{\'a}n Gonz{\'a}lez and Tarek A. El-Ghazawi},
  journal={ACM Trans. Reconfigurable Technol. Syst.},
  year={2009},
  volume={1},
  pages={21:1-21:23}
}
Runtime Reconfiguration (RTR) has been traditionally utilized as a means for exploiting the flexibility of High-Performance Reconfigurable Computers (HPRCs). However, the RTR feature comes with the cost of high configuration overhead which might negatively impact the overall performance. Currently, modern FPGAs have more advanced mechanisms for reducing the configuration overheads, particularly Partial Runtime Reconfiguration (PRTR). It has been perceived that PRTR on HPRC systems can be the… 
Space and Time Sharing of Reconfigurable Hardware for Accelerated Parallel Processing
TLDR
A resource virtualization solution based on Partial Run-Time Reconfiguration (PRTR) that allows space, time, and/or space-time sharing of the reconfigurable (co-)processors among the (main) processors and thus increasing the overall system utilization.
AMRC: an algebraic model for reconfiguration of high performance cluster computing systems at runtime
TLDR
A vector algebra-based model for HPCS reconfiguration at runtime is presented and named AMRC, which determines the element causing the dynamic behavior and analyzes the reason regarding both software and hardware at runtime.
Configuration Mapping Algorithms to Reduce Energy and Time Reconfiguration Overheads in Reconfigurable Systems
TLDR
A configuration memory hierarchy including two on-chip memory modules with different access time and energy consumption features is presented and two configuration mapping algorithms are developed that take advantage of these memories to reduce the system energy consumption, while increasing its performance.
Optimising runtime reconfigurable designs for high performance applications
TLDR
This thesis proposes novel optimisations for high performance runtime reconfigurable designs based on Reconfiguration Data Flow Graph, a new hierarchical graph structure enabling runtime reconfigured designs to be synthesised in three steps: function analysis, configuration organisation, and runtime solution generation.
Hardware Task Scheduling for Partially Reconfigurable FPGAs
TLDR
This work proposes and implements a run time system manager for scheduling software and hardware tasks on available processor(s) and hardware (HW) tasks on any number of reconfigurable regions of a partially reconfigured FPGA, and validate its correctness using its RTSM to execute an image processing application on a ZedBoard platform.
Probabilistic Performance Modelling when Using Partial Reconfiguration to Accelerate Streaming Applications with Non-deterministic Task Scheduling
TLDR
A probabilistic approach is used to estimate the acceleration of streaming applications with unknown task schedule thanks to the application of partial reconfiguration, and the impact of how different strategies or heuristics affect to the final performance is included in this analysis.
Hardware Architectural Support for Caching Partitioned Reconfigurations in Reconfigurable Systems
TLDR
This paper presents a hardware implementation of an on-chip configuration memory controller that efficiently manages run-time reconfigurations and includes support to deal with configurations that have been divided into blocks of customizable size.
VAPRES: A Customizable and Flexible Base Architecture for Partially Reconfigurable Systems
TLDR
VAPRES is presented, a PR base architecture that provides a customizable and flexible platform for PR system and application design that offers reconfigurable and on-demand communication channel establishment, and seamless hardware module replacement.
Partial reconfiguration on FPGAs in practice — Tools and applications
TLDR
A survey on state-of-the-art trends on reconfigurable architectures and devices, application specific requirements, and design techniques and tools that are essential for implementing partial run-time reconfiguration on FPGAs, and a demonstration of the floorplanning and constraint generation tool GoAhead.
...
1
2
3
4
5
...

References

SHOWING 1-10 OF 46 REFERENCES
Dynamic precision management for loop computations on reconfigurable architectures
  • K. Bondalapati, V. Prasanna
  • Computer Science
    Seventh Annual IEEE Symposium on Field-Programmable Custom Computing Machines (Cat. No.PR00375)
  • 1999
TLDR
It is shown how the precision requirements can be analyzed for typical computations in loops by computing the precision variation curve, and algorithms to generate optimal schedules of configurations using the Precision variation curves are developed.
Exploiting processing locality for adaptive computing systems
TLDR
A virtual-memory-like technique, called Virtual Configuration Management (VCM), has been proposed, which extends cache and memory management techniques to reconfigurable platforms and augmented them with data mining concepts using association rule mining (ARM).
Exploiting dynamic and partial reconfiguration for FPGAs: toolflow, architecture and system integration
TLDR
This tutorial should help designers as well as researchers in developing dynamic and partial reconfigurable systems and increase the number of area of applications exploiting this very promising methodology.
Performance and overhead in a hybrid reconfigurable computer
TLDR
The results show that the SRC-6E can outperform a general-purpose microprocessor for computationally intensive algorithms by a factor of over 1500, however, overhead due to configuration and data transfer must be properly dealt with by the application or the system's run-time environment to achieve the full throughput potential.
Configuration caching management techniques for reconfigurable computing
TLDR
This work presents techniques to carefully manage the configurations present on the reconfigurable hardware throughout program execution, and shows that the number of required reconfigurations is reduced, lowering the configuration overhead.
Evaluating Partial Reconfiguration for Embedded FPGA Applications
TLDR
This paper analyzes the performance impact of using PR to perform remote updating, an important capability often used in embedded applications.
A system-level design methodology for reconfigurable computing applications
TLDR
A structured design life-cycle, which can be applied at the reconfigurable computing systems rather than at the chip-level as it has been traditionally the case, is proposed, formally modeled and experimentally verified.
Analytical Modeling for High Performance Reconfigurable Computers
TLDR
The initial results for an analytical modeling framework for High Performance Reconfigurable Computing systems are presented and it is evident that a mathematical modeling framework is a necessary tool for analyzing the complex interaction of design issues and performance metrics for HPRC.
Analytical modeling of high performance reconfigurable computers: prediction and analysis of system performance
TLDR
This dissertation develops and validates an analytic performance modeling methodology for a class of fork-join algorithms executing on a High Performance Reconfigurable Computing (HPRC) platform and proves to be accurate in characterizing the effects of reconfigurable devices, application load imbalance, background user load and heterogeneity for applications running on shared, homogeneous and heterogeneous HPRC resources.
Dynamic reconfiguration for management of radiation-induced faults in FPGAs
TLDR
To study single event upset (SEU) impact on signal processing applications, a novel fault injection technique to corrupt configuration bits is used, thereby simulating SEU faults and highlighting the benefits of dynamic reconfiguration for space-based reconfigurable computing.
...
1
2
3
4
5
...