Compiler Controlled Speculation for Power Aware ILP Extraction in Dataflow Architectures

Abstract

Abstract. Traditional predicated execution uses two techniques: top predication -in which only the head of the dependence chain is predicated, and bottom predication -in which only the tail of the dependence chain is predicated. Top predication prevents speculative execution, thus delivering minimum performance at minimum energy cost, while bottom predication allows full speculation of the dependence chain, resulting in maximum performance at maximum energy cost. In this paper, we propose a novel power-aware ILP extraction technique, denoted the 'elasticblock', that combines these two extremes, exposing superior energy vs. performance trade-offs. Each instruction in the elastic-block is explicitly guarded by two predicates: the speculative, and the final. Instruction's final predicate is generated using traditional if-conversion technique, while the speculative predicate has its default value statically assigned by the compiler, enabling it to make power-performance trade-offs in the code. Several energy saving code optimizations are proposed for the elasticblock structure.

Islamic Economic Crisis: A Systematic Review Using Bibliometric Analysis

“…There are several causes of the financial crisis from an Islamic economic perspective, including; Social instability, namely shocks that cause fluctuation in social aspects, such as rampant crime, corruption, and poverty, which hinder or stop economic activity (Ghoul, 2011;Hassan & Kayed, 2009), Speculation is the practice of financial transactions to benefit from short-term fluctuations. Or medium in the market value of tradable goods, such as financial instruments (Ahmed, Rosser Jr, & Uppal, 2010;Almoharby, 2011;Chapra, 2008;Farooq, John, & Jacome, 2009;Seidu, 2009;Siddiqi, 2008;Trabelsi, 2011),…”

Autocorrelation analysis: A new and improved method for branch predictability characterization

“…Its benefit largely depends on the accuracy in identifying those branches. Compilers typically identify the candidates for predication by profiling the program and collecting the branches with taken rate near 50% [8] or with low prediction accuracy for a certain branch predictor [13] [19]. Since the proposed DPI metric has higher accuracy in identifying hard-to-predict branches than taken/transition rate, it can replace taken rate or transition rate in identifying the hard-topredict branches and improve the effectiveness of predication.…”

A Method to Reduce Energy Consumption of Conditional Operations with Execution Probabilities

“…Though these tools perform many kinds of optimization, minimizing energy consumption with respect to the speculative execution is not mentioned. There also have been much research on power-or energyaware speculation in processors [15], [16], [17], [18], [19]. In Ref.…”

“…In Ref. [19] a special data driven architecture is targeted. These techniques are difficult to apply to minimizing the energy consumption of operations with the constraints of time and resources.…”