Home Page Icon
Home Page
Table of Contents for
Part 4: Power and Reliability
Close
Part 4: Power and Reliability
by Hamid Sarbazi-Azad
Advances in GPU Research and Practice
Cover image
Title page
Table of Contents
Copyright
Dedication
List of Contributors
Preface
Acknowledgments
Part 1: Programming and Tools
Chapter 1: Formal analysis techniques for reliable GPU programming: current solutions and call to action
Abstract
Acknowledgments
1 GPUs in Support of Parallel Computing
2 A quick introduction to GPUs
3 Correctness issues in GPU programming
4 The need for effective tools
5 Call to Action
Chapter 2: SnuCL: A unified OpenCL framework for heterogeneous clusters
Abstract
Acknowledgments
1 Introduction
2 OpenCL
3 Overview of SnuCL framework
4 Memory management in SnuCL Cluster
5 SnuCL extensions to OpenCL
6 Performance evaluation
7 Conclusions
Chapter 3: Thread communication and synchronization on massively parallel GPUs
Abstract
1 Introduction
2 Coarse-Grained Communication and Synchronization
3 Built-In Atomic Functions on Regular Variables
4 Fine-Grained Communication and Synchronization
5 Conclusion and Future Research Direction
Chapter 4: Software-level task scheduling on GPUs
Abstract
Acknowledgments
1 Introduction, Problem Statement, and Context
2 Nondeterministic behaviors caused by the hardware
3 SM-centric transformation
4 Scheduling-enabled optimizations
5 Other scheduling work on GPUs
6 Conclusion and future work
Chapter 5: Data placement on GPUs
Abstract
1 Introduction
2 Overview
3 Memory specification through MSL
4 Compiler support
5 Runtime support
6 Results
7 Related work
8 Summary
Part 2: Algorithms and Applications
Chapter 6: Biological sequence analysis on GPU
Abstract
1 Introduction
2 Pairwise Sequence Comparison and Sequence-Profile Comparison
3 Design aspects of GPU solutions for biological sequence analysis
4 GPU Solutions for Pairwise Sequence Comparison
5 GPU Solutions for Sequence-Profile Comparison
6 Conclusion and perspectives
Chapter 7: Graph algorithms on GPUs
Abstract
1 Graph representation for GPUs
2 Graph traversal algorithms: the breadth first search (BFS)
3 The single-source shortest path (SSSP) problem
4 The APSP problem
5 Load Balancing and Memory Accesses: Issues and Management Techniques
Chapter 8: GPU alignment of two and three sequences
Abstract
1 Introduction
2 GPU architecture
3 Pairwise alignment
4 Alignment of three sequences
5 Conclusion
Chapter 9: Augmented Block Cimmino Distributed Algorithm for solving tridiagonal systems on GPU
Abstract
1 Introduction
2 ABCD Solver for tridiagonal systems
3 GPU implementation and optimization
4 Performance evaluation
5 Conclusion and future work
Chapter 10: GPU computing applied to linear and mixed-integer programming
Abstract
Acknowledgments
1 Introduction
2 Operations Research in Practice
3 Exact Optimization Algorithms
4 Metaheuristics
5 Conclusions
Conflicts of Interest
Chapter 11: GPU-accelerated shortest paths computations for planar graphs
Abstract
1 Introduction
2 Related work
3 Partitioned Approaches
4 Computational Complexity Analysis
5 Experiments and results
About the Authors
Chapter 12: GPU sorting algorithms
Abstract
1 Introduction
2 Generic Programming Strategies for GPU
3 Sorting algorithms
Chapter 13: MPC: An effective floating-point compression algorithm for GPUs
Abstract
Acknowledgments
1 Introduction
2 Methodology
3 Experimental results
4 Summary and Conclusions
Chapter 14: Adaptive sparse matrix representation for efficient matrix-vector multiplication
Abstract
1 Introduction
2 Sparse matrix-vector multiplication
3 GPU architecture and programming model
4 Optimization principles for SpMV
5 Platform (Adaptive Runtime System)
6 Results and analysis
7 Summary
Part 3: Architecture and Performance
Chapter 15: A framework for accelerating bottlenecks in GPU execution with assist warps
Abstract
Acknowledgments
1 Introduction
2 Background
3 Motivation
4 The CABA Framework
5 A Case for CABA: Data Compression
6 Methodology
7 Results
8 Other Uses of the CABA Framework
9 Related Work
10 Conclusion
Chapter 16: Accelerating GPU accelerators through neural algorithmic transformation
Abstract
1 Introduction
2 Neural transformation for GPUs
3 Instruction-set-architecture design
4 Neural accelerator: design and integration
5 Controlling quality trade-offs
6 Evaluation
7 Related work
8 Conclusion
Chapter 17: The need for heterogeneous network-on-chip architectures with GPGPUs: A case study with photonic interconnects
Abstract
1 Introduction
2 Background
3 The Need for Heterogeneous Interconnections
4 Characterization of GPGPU Performance
5 Conclusion
Chapter 18: Accurately modeling GPGPU frequency scaling with the CRISP performance model
Abstract
Acknowledgments
1 Introduction
2 Motivation and related work
3 GPGPU DVFS performance model
4 Methodology
5 Results
6 Conclusion
Part 4: Power and Reliability
Chapter 19: Energy and power considerations of GPUs
Abstract
1 Introduction
2 Evaluation methodology
3 Power profiling of regular and irregular programs
4 Affecting power and energy on GPUs
5 Summary
Appendix
About the authors
Chapter 20: Architecting the last-level cache for GPUs using STT-MRAM nonvolatile memory
Abstract
1 Introduction
2 Background
3 Related Work
4 Two-Part L2 Cache Architecture
5 Dynamic Write Threshold Detection Mechanism
6 Implementation
7 Evaluation Result
8 Conclusion
Chapter 21: Power management of mobile GPUs
Abstract
Acknowledgments
1 Introduction
2 GPU Power Management for Mobile Games
3 GPU Power Management for GPGPU Applications
4 Future Outlook
5 Conclusions
Chapter 22: Advances in GPU reliability research
Abstract
1 Introduction
2 Evaluating GPU Reliability
3 Hardware Reliability Enhancements
4 Software Reliability Enhancements
5 Summary
Chapter 23: Addressing hardware reliability challenges in general-purpose GPUs
Abstract
1 Introduction
2 GPGPUs Architecture
3 Modeling and Characterizing GPGPUs Reliability in the Presence of Soft Errors [25]
4 RISE: Improving the Streaming Processors’ Reliability Against Soft Errors in GPGPUs [36]
5 Mitigating the Susceptibility of GPGPUs to PVs [43]
Author Index
Subject Index
Search in book...
Toggle Font Controls
Playlists
Add To
Create new playlist
Name your new playlist
Playlist description (optional)
Cancel
Create playlist
Sign In
Email address
Password
Forgot Password?
Create account
Login
or
Continue with Facebook
Continue with Google
Sign Up
Full Name
Email address
Confirm Email Address
Password
Login
Create account
or
Continue with Facebook
Continue with Google
Prev
Previous Chapter
Chapter 18: Accurately modeling GPGPU frequency scaling with the CRISP performance model
Next
Next Chapter
Chapter 19: Energy and power considerations of GPUs
Part 4
Power and Reliability
Add Highlight
No Comment
..................Content has been hidden....................
You can't read the all page of ebook, please click
here
login for view all page.
Day Mode
Cloud Mode
Night Mode
Reset