Dynamic Reconfigurable Architectures and Transparent Optimization Techniques: Automatic Acceleration of Software Execution

ISBN-13: 9789048139125 / Angielski / Twarda / 2010 / 177 str.

Antonio Carlos Schneider Beck Fl.;Luigi Carro

Opis

Dynamic Reconfigurable Architectures and Transparent Optimization Techniques presents a detailed study on new techniques to cope with the aforementioned limitations. First, characteristics of reconfigurable systems are discussed in details, and a large number of case studies is shown. Then, a detailed analysis of several benchmarks demonstrates that such architectures need to attack a diverse range of applications with very different behaviours, besides supporting code compatibility. This requires the use of dynamic optimization techniques, such as Binary Translation and Trace reuse. Finally, works that combine both reconfigurable systems and dynamic techniques are discussed and a quantitative analysis of one them, the DIM architecture, is presented.

Spis treści

1. Introduction. 1.1 Challenges. 1.2 Main Motivations. 1.3 This Book. 2. Reconfigurable Systems. 2.1 Introduction. 2.2 Basic Principles. 2.3 How It Works. 2.4 Advantages of using Reconfigurable Logic. 2.5 Classification. 2.6 Directions. 3. Deployment of Reconfigurable Systems. 3.1 Introduction. 3.2 Examples.3.3 Recent Dataflow Architectures. 3.4 Summary and Comparative Tables. 4. Dynamic Optimization Techniques. 4.1 Introduction. 4.2 Binary Translation. 4.3 Reuse. 5. Dynamic Detection and Reconfiguration. 5.1 Warp Processing. 5.2 Configurable Compute Array. 5.3 Drawbacks. 6. The DIM Reconfigurable System. 6.1 Introduction. 6.2 The Reconfigurable Array in Details. 6.3 Translation, Reconfiguration and Execution. 6.4 The BT Algorithm in Details. 6.5 Case Studies. 6.6 DIM in Stack Machines. 6.7 On-Going and Future works. 7. Conclusions and Future Trends. 7.1 Introduction. 7.2 Decreasing the routing area of Reconfigurable Systems. 7.3 Measuring the impact of the OS in reconfigurable systems. 7.4 Reconfigurable systems to increase the Yield. 7.5 Study of the area overhead with technology scaling and future technologies. 7.6 Scheduling targeting to Low-Power. 7.7 Comparison against a Fine-Grain reconfigurable system. 7.8 Reconfigurable Systems Attacking Different levels of Instruction Granularity.

Blurb

As Moore’s law is losing steam, one already sees the phenomenon of clock frequency reduction caused by the excessive power dissipation in general purpose processors. At the same time, embedded systems are concentrating several heterogeneous applications in a single device, and hence new architectural alternatives are necessary. Reconfigurable computing has already shown to be a potential solution when it comes to accelerate specific code with a small power budget, but significant speedups are achieved only in very dedicated dataflow oriented software, without capturing the reality of nowadays complex heterogeneous systems. Moreover, any architecture solution should be able to execute legacy code, since there is already a large base of applications and standards.

Dynamic Reconfigurable Architectures and Transparent Optimization Techniques presents a detailed study on new techniques to cope with the aforementioned limitations. First, characteristics of reconfigurable systems are discussed in details, and a large number of case studies is shown. Then, a detailed analysis of several benchmarks demonstrates that such architectures need to attack a diverse range of applications with very different behaviours, besides supporting code compatibility. This requires the use of dynamic optimization techniques, such as Binary Translation and Trace reuse. Finally, works that combine both reconfigurable systems and dynamic techniques are discussed and a quantitative analysis of one them, the DIM architecture, is presented.