Performance of Iterative Computation in Self-Timed Rings

Williams, Ted E.

doi:10.1007/978-1-4615-2794-7_3

Cited by 5 publications

(6 citation statements)

References 8 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The concept of inserting empty buffers for optimizing system performance was long known in asynchronous design [43], [31]. In [2], [35], exact algorithms for slack matching on choice-free asynchronous systems were presented.…”

Section: B Recyclingmentioning

confidence: 99%

Elastic systems

Cortadella

Galceran-Oms

Kishinevsky

2010

Eighth ACM/IEEE International Conference on Formal Methods and Models for Codesign (MEMOCODE 2010)

View full text Add to dashboard Cite

Elastic systems provide tolerance to the variations in computation and communication delays. The incorporation of elasticity opens new opportunities for optimization using new correct-by-construction transformations that cannot be applied to rigid non-elastic systems.The basics of synchronous and asynchronous elastic systems will be reviewed. A set of behavior-preserving transformations will be presented: retiming, recycling, early evaluation, variablelatency units and speculative execution. The application of these transformations for performance and power optimization will be discussed. Finally, a novel framework for microarchitectural exploration will be introduced, showing that the optimal pipelining of a circuit can be automatically obtained by using the previous transformations.

show abstract

Section: B Recyclingmentioning

confidence: 99%

Elastic systems

Cortadella

Galceran-Oms

Kishinevsky

2010

Eighth ACM/IEEE International Conference on Formal Methods and Models for Codesign (MEMOCODE 2010)

View full text Add to dashboard Cite

show abstract

“…2(b) shows a recursive self-timed pipeline, in which parallel threads of computation are synchronized by joins and forks. We note that the throughput of a self-timed recursive loop does not always increase with the increase of the pipeline depth [13]. The throughput depends on the ratio of the number of tokens (or pipeline depth) to the number of bubbles.…”

Section: Introductionmentioning

confidence: 98%

“…Otherwise, the datapath is in bubble limited mode and increasing pipeline depth will decrease the throughput. Readers are referred to [5], [13] for a detailed discussion.…”

Section: Introductionmentioning

confidence: 99%

Self-timed dynamically pipelined adaptive signal processing system: a case study of DLMS equalizer for read channel

Chen

Zhang

2005

IEEE Trans. Circuits Syst. I

View full text Add to dashboard Cite

Many pipelined adaptive signal processing systems are subject to a tradeoff between throughput and signal processing performance incurred by the pipelined adaptation feedback loops. In the conventional synchronous design regime, such throughput/performance tradeoff is typically fixed since the pipeline depth is usually determined in the design phase and remains unchanged in the run time. Nevertheless, in many real-life scenarios, the overall system performance can be potentially improved if we can run-time dynamically configure this tradeoff. With this motivation, we propose to apply self-timed pipeline, an alternative to synchronous pipeline, to implement the pipelined adaptive signal processing systems, in which the pipeline depth can be dynamically changed to realize run-time configurable throughput/performance tradeoffs. Based on a well-known high speed self-timed pipeline style, we developed architecture and circuit level design techniques to implement the self-timed pipelined adaptation feedback loop with configurable pipeline depth. We demonstrate the proposed design approach using a delayed least mean square (DLMS) adaptive equalizer for magnetic recording read channel. The data transfer rate in hard disk varies as the read head moves among tracks with different distance from the center of the disk platter. By adjusting the pipeline depth on-the-fly, the DLMS equalizer can dynamically track the best equalization performance allowed by the varying data transfer rates. Simulation result shows a significant performance improvement compared with its synchronous counterpart.Index Terms-Adaptive signal processing, delayed least mean square (DLMS), self-timed, pipelining.

show abstract

“…However, methods based on theoretical analysis [5][6], graph unfolding [7] and Markov analysis [8] handle only choice-free or data-independent asynchronous pipelines. A canopy graph based approach was proposed that achieves both accurate in throughput estimation and fast in runtimes [9].…”

Section: Introductionmentioning

confidence: 99%