Conference Paper

Improving Energy-Efficiency by Bypassing Trivial Computations.

DOI: 10.1109/IPDPS.2005.253 Conference: 19th International Parallel and Distributed Processing Symposium (IPDPS 2005), CD-ROM / Abstracts Proceedings, 4-8 April 2005, Denver, CO, USA
Source: DBLP

ABSTRACT We study the energy efficiency benefits of bypassing trivial computations in high-performance processors. Trivial computations are those computations whose output can be determined without performing the computation. We show that bypassing trivial instructions reduces energy consumption while improving performance. Our study shows that by bypassing trivial instructions and for the subset of SPEC'2K benchmarks studied here, on average, it is possible to improve energy and energy-delay by up to 4.5% and 11.8% over a conventional processor.

  • [Show abstract] [Hide abstract]
    ABSTRACT: In this work, we discuss several drawbacks of the conventional wide-width redundant operation table such as the waste of area cost and power consumption. We found that the waste of area cost and power consumption is caused by storing meaningless bits of the narrow-width operand values. Based on this analysis, we propose a way to avoid these storing of meaningless information of the narrow-width operands. The proposed method, partial resolution method, divides the conventional wide-width redundant operation table into two tables as the wide-width table for the half entries and the narrow-width table for the other half entries. The wide-width and the narrow-width redundant operation tables store different dynamic instructions whose operand values are wide and narrow, respectively. Since the narrow-width redundant operation table stores smaller number of bits, it requires lower area cost and also power consumption compared with the wide-width redundant operation table. The partial resolution method decreases the area cost by about 7 percent and 20 percent for the integer and the floating-point tables, respectively, and reduces the dynamic power consumption by about 34 percent and 30 percent for the integer and the floating-point tables, respectively, compared with the conventional wide-width redundant operation table with 2K entries. Meanwhile, the performance simulation with a high-end microarchitecture model and SPEC2000 benchmarks shows that the partial resolution method affects the performance very little, and even increases slightly in terms of IPC (Instruction per Cycle) value.
    Microprocessors and Microsystems - Embedded Hardware Design. 01/2008; 32:79-94.
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: Instruction reuse (IR) and trivial computation (TC) elimination are two architectural techniques that aim at eliminating redundant code to better exploit instruction-level parallelism. While they have been extensively studied in isolation, this paper is the first to compare their relative efficiency. This is done using applications from the embedded domain. This paper establishes the relationship between the two techniques by framing the arithmetic instructions detected by each of them. While TC can only eliminate instructions where one of the operands is zero or one, IR has potentially a wider scope as it can potentially eliminate any instruction given that it has been executed before with the same set of operand values. Despite the wider scope, we have found that IR and TC can eliminate about the same fraction of instructions even if an infinitely large instruction reuse buffer is assumed (IR and TC can eliminate 26% and 22% of the instructions, respectively). Another quite surprising finding is that the two techniques target quite different sets of instructions suggesting that they can provide almost additive gains if combined. In combination, they can eliminate 40% of the instructions they target. In terms of energy-efficiency, we finally find that if an instruction reuse buffer of 256 entries is used, it uses 1% more energy than a processor without IR and TC reduces the energy consumption by 5.6%.
    IEEE Second International Symposium on Industrial Embedded Systems - SIES'2007, Hotel Costa da Caparica, Lisbon, Portugal, 4-6 July 2007; 01/2007
  • Source
    [Show abstract] [Hide abstract]
    ABSTRACT: This paper addresses the issue of improving the energy efficiency of processors by eliminating trivial operations. The paper provides a new classification of trivial operations and quantifies their relative frequency in desktop and embedded applications. It then presents a hardware technique to remove trivial operations as early as at the decode stage of the pipeline to save energy. This paper shows that 13.6% and 8.6% of the instructions are identity-trivial in the selected applications in the SPEC CPU2000 and EEMBC1.1 benchmark suites, respectively. Early detection and elimination of trivial operations reduce the average energy consumption of the core pipeline by 9% and 6%, respectively.
    Microprocessors and Microsystems - Embedded Hardware Design. 01/2008; 32:183-196.