Extending the Cell SPE with Energy Efficient Branch Prediction
نویسندگان
چکیده
Energy-efficient dynamic branch predictors are proposed for the Cell SPE, which normally depends on compiler-inserted hint instructions to predict branches. All designed schemes use a Branch Target Buffer (BTB) to store the branch target address and the prediction, which is computed using a bimodal counter. One prediction scheme predecodes instructions when they are fetched from the local store and accesses the BTB only for branch instructions, thereby saving power compared to conventional dynamic predictors that access the BTB for every instruction. In addition, several ways to leverage the existing hint instructions for the dynamic branch predictor are studied. We also introduce branch warning instructions which initiate branch prediction before the actual branch instruction is fetched. They allow fetching the instructions starting at the branch target and thus completely remove the branch penalty for correctly predicted branches. For a 256-entry BTB, a speedup of up to 18.8% is achieved. The power consumption of the branch prediction schemes is estimated at 1% or less of the total power dissipation of the SPE and the average energy-delay product is reduced by up to 6.2%.
منابع مشابه
Energy Efficient Branch Prediction on the Cell SPE
We propose novel power efficient branch predictors for the Cell SPU, which normally depends on compiler inserted hint instructions to predict taken branches. Several prediction schemes were designed all using a Branch Target Buffer (BTB) to store the branch target address and the prediction, which is computed using a bimodal counter. One prediction scheme pre-decodes instructions once they are ...
متن کاملExtending the Super Efficiency Method to Rank the Non-Extreme Efficient Units
This article will address the extension of super efficiency method to rank the non-extreme efficient decision making units. Many methodologies have introduced methods that can rank efficient units, amongst which, the super efficiency method due to its ability to provide meaningful geometrical as well as economic analyses has a significant place. But the common problem with all the super efficie...
متن کاملElectrical Energy Storage on the Hybrid Grid of Renewable Energy System Using Fuzzy Controller Optimization Algorithm
The main risks of arising from the using fossil fuels can be referred to environmental pollution, the effects of greenhouse gases, climate change and acid rain. For this reason, efficient use of energy in economic development has always been considered as an important goal of sustainable development. In this study, the effects of time-varying electricity prices in the energy storage components ...
متن کاملDesign and Simulation of a Highly Efficient InGaN/Si Double-Junction Solar Cell
A solar cell is an electronic device which not only harvests photovoltaic effect but also transforms light energy into electricity. In photovoltaic phenomenon, a P-N junction is created to form an empty region. The presented paper aims at proposing a new highly efficient InGaN/Si double-junction solar cell structure. This cell is designed to be used in a real environmental situation, so only s...
متن کاملDesigning Energy-Efficient Fetch Engines
This dissertation evaluates factors that affect the energy-efficiency of the fetch engine in a programmable uniprocessor. The central thesis is that branch prediction is one of the key factors affecting overall processor energy-efficiency. Cooling costs, extending battery life in mobile devices, and reducing utility costs for wall-powered systems, especially data centers are growing concerns. T...
متن کامل