• Proceedings of the Korean Information Science Society Conference
• /
• 2001.04a
• /
• pp.22-24
• /
• 2001

슈퍼스칼라 프로세서에서는 분기 명령의 결과 지연으로 명령의 공급이 중단되는 것을 방지하고 지속적인 파이프라인 처리를 위해서 분기의 결과를 미리 예측하여 명령을 폐치하고 있다. 본 논문에서는 심플스칼라 툴 셋을 사용하여 슈퍼스칼라 프로세서에서 사용되는 대표적인 동적 분기예측 방법 시뮬레이션 환경을 구축한다. 동적 분기예측 방법으로 분기 타겟버퍼(Branch Target Buffer, BTB) 상에서 분기명령의 자기 히스토리에 근거한 BTB 방식과 이전 분기명령의 히스토리와의 상관관계를 고려한 Gshare 분기예측기를 적용 구현한다. 심플스칼라 시뮬레이터에 SPEC95 벤치마크 프로그램을 실행시켜 디자인 파라미터 변화에 따른 분기 예측기의 예측정확도를 실험한다. 또한 BTB와 Gshare 분기예측기를 VHDL로 구현하고 Synopsys 툴을 이용하여 시뮬레이션 및 합성 과정을 거쳐 게이트 크기와 파워 소모량을 측정한다.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.44 no.1
• /
• pp.33-40
• /
• 2007

Processor pipelines have been growing deeper and issue widths wider over the years. If this trend continues, the branch misprediction penalty will become very high. Branch misprediction is the single most significant performance limiter for improving processor performance using deeper pipelining. Therefore, more accurate branch predictor becomes an essential part of modern processors. Several branch predictors combine a part of the branch address with a fixed amount of global branch history to make a prediction. These predictors cannot perform uniformly well across all programs because the best amount of branch history to be used depends on the program and branches in the program. Therefore, predictors that use a fixed history length are unable to perform up to their potential performance. In this paper, we propose a branch prediction mechanism, using variable length history, which predicts using a bank having higher prediction accuracy among predictions from five banks. Bank 0 is a bimodal predictor which is indexed with the 12 least significant bits of the branch address. Banks 1, 2, 3 and 4 are predictors which are indexed with different global history bits and the branch PC. In simulation results, the proposed mechanism outperforms gshare predictors using fixed history length of 12 and 13 , up to 6.34% in prediction accuracy. Furthermore, the proposed mechanism outperforms gshare predictors using best history lengths for benchmarks, up to 2.3% in prediction accuracy.

• Journal of Korean Society of Forest Science
• /
• v.84 no.2
• /
• pp.121-130
• /
• 1995

Forecasting accuracy is examined in the context of Michigan travel demand. Eight different annual models are used to forecast up to two years ahead, and nine different quarterly models up to four quarters. In the evaluation of annual models' performance, multiple regression performed better than the other methods in both the one year and two year forecasts. For quarterly models, Winters exponential smoothing and the Box-Jenkins method performed better than naive 1 s in the first quarter ahead, but these methods in the second, third, and fourth quarters ahead performed worse than naive 1 s. The sophisticated models did not outperform simpler models in producing quarterly forecasts. The best model, multiple regression, performed slightly better when fitted to quarterly rather than annual data: however, it is not possible to strongly recommend quarterly over annual models since the improvement in performance was slight in the case of multiple regression and inconsistent across the other models. As one would expect, accuracy declines as the forecasting time horizon is lengthened in the case of annual models, but the accuracy of quarterly models did not confirm this result.

• The KIPS Transactions:PartA
• /
• v.16A no.5
• /
• pp.347-356
• /
• 2009

Precise branch prediction is a critical factor in the IPC Improvement of modern microprocessor architectures. In addition to the branch prediction accuracy, branch prediction delay have a profound impact on overall system performance as well. However, it tends to be overlooked when the architects design the branch predictor. To tolerate branch prediction delay, this paper proposes Early Start Prediction (ESP) technique. The proposed solution dynamically identifies the start instruction of basic block, called as Basic Block Start Address (BB_SA), and the solution uses BB_SA when predicting the branch direction, instead of branch instruction address itself. The performance of the proposed scheme can be further improved by combining short interval hiding technique between BB_SA and branch instruction. The simulation result shows that the proposed solution hides prediction latency, with providing same level of prediction accuracy compared to the conventional predictors. Furthermore, the combination with short interval hiding technique provides a substantial IPC improvement of up to 10.1%, and the IPC is actually same with ideal branch predictor, regardless of branch predictor configurations, such as clock frequency, delay model, and PHT size.

• Proceedings of the Korean Information Science Society Conference
• /
• 2005.07a
• /
• pp.808-810
• /
• 2005

조건 분기예측실패는 많은 사이클을 낭비시키며, 비순서적 실행을 방해하고, 잘못 예측된 명령어들을 수행하게 되므로 전력을 낭비한다. gshare와 GAg같은 전역 히스토리를 기반으로 하는 예측기에서는 히스토리의 명령어 완료시간 갱신(commit update)에 의해 많은 분기예측실패가 발생한다. 이를 위해 히스토리를 모험적으로 갱신하고, 분기예측실패 시 히스토리를 복구시키는 메커니즘에 관한 연구들이 제시되었다. 본 논문에서는 기존 분기예측기에 age_Counter를 추가하여 미해결 분기명령어 수를 저장하며, 이를 분기예측실패 후 분기 히스토리 레지스터를 복구하는데 사용하는 간단한 복구 메커니즘을 제안한다. SimpleScalar 3.0/PISA 툴셋과 SPECINT95 벤치마크 프로그램에서 시뮬레이션 한 결과, 제안된 복구 메커니즘은 GAg와 gshare 예측기에서 예측정확도는 각각 $9.21\%$와 $2.14\%$가 개선되었고, IPC는 $18.08\%$와 $8.75\%$ 개선되었다.

• The KIPS Transactions:PartA
• /
• v.14A no.6
• /
• pp.347-356
• /
• 2007

In designing a branch predictor, in addition to accuracy, microarchitects should consider power consumption, especially for embedded processors. This paper proposes a power-aware branch predictor, which is based on the gshare predictor, by accessing the BTB (Branch Target Buffer) only when the prediction from the PHT (Pattern History Table) is taken. To enable the selective access to the BTB, the PHT in the proposed branch predictor is accessed one cycle earlier than the traditional PHT to prevent the additional delay. As a side effect, two predictions from the PHT are obtained through one access to the PHT, which leads to more power savings. The proposed branch predictor reduces the power consumption, not requiring any additional storage arrays, not incurring additional delay (except just one MUX delay) and never harming accuracy. Simulation results show that the proposed predictor reduces the power consumption by $35{\sim}48%$ compared to the traditional predictor.

• The Transactions of the Korea Information Processing Society
• /
• v.5 no.10
• /
• pp.2676-2685
• /
• 1998

In order to improve the branch prediction accuracy and to reduce the BTB miss rate, this paper proposes a two-level BTB structure that adds small-sized victim BTB to the convetional BTB. With small cost, two-level BTB can reduce the BTB miss rate as well as improve the prediction accuracy of the hybrid branch prediction strategy which combines dynamic prediction and static prediction. Through the trace-driven simulation of four bechmark programs, the performance improvement by the proposed two-level BTB structure is analysed and validated. Our proposed BTB structure can improve the BTB miss rate by 26.5% and the misprediction rate by 26.75%

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.43 no.4 s.310
• /
• pp.1-14
• /
• 2006

The indirect branch instruction is a most substantial obstacle in utilizing ILP of modem high performance processors. The target address of an indirect branch has the polymorphic characteristic varied dynamically, so it is very difficult to predict the accurate target address. Therefore the performance of a processor with speculative methodology is reduced significantly due to the many execution cycle delays in occurring the misprediction. We proposed the very accurate and novel indirect branch prediction scheme so called data-dependence based prediction. The predictor results in the prediction accuracy of 98.92% using 1K entries, and. 99.95% using 8K But, all of the proposed indirect predictor including our predictor has a large hardware overhead for restoring expected target addresses as well as tags for alleviating an aliasing. Hence, we propose the scheme minimizing the hardware overhead without sacrificing the prediction accuracy. Our experiment results show that the hardware is reduced about 60% without the performance loss, and about 80% sacrificing only the performance loss of 0.1% in aspect of the tag overhead. Also, in aspect of the overhead of storing target addresses, it can save the hardware about 35% without the performance loss, and about 45% sacrificing only the performance loss of 1.11%.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.13 no.3
• /
• pp.360-365
• /
• 2003

In this paper, GMDH(Croup Method of Data Handling) algorithm which is proved to be more excellent in efficiency and accuracy of practical use of data is applied to electric power demand forecasting. As a result, it became much easier to make a choice of input data and make an exact prediction based on a lot of data. Also, we considered both economy factors(GDP, export, import, number of employee, number of economically active population and consumption of oil) and climate factors(average temperature) when forecasting. We assumed target forecast period from first quarter 1999 to first quarter 2001, and suggested more accurate forecasting method of electric power demand by using 3-step computer simulation processes(first process for selecting optimum input period, second for analyzing time relation of input data and forecast value, and third for optimizing input data) for improvement of forecast precision. The proposed method can get 0.96 percent of mean error rate at target forecast period.

• Journal of KIISE:Computer Systems and Theory
• /
• v.33 no.5
• /
• pp.297-305
• /
• 2006

Trace-driven simulation is widely used for measuring the performance of a microprocessor in its initial design phase. However, since it requires much time and disk space, the statistical simulation has been studied as an alternative method. In this paper, statistical simulations are performed for a high performance superscalar microprocessor with a perceptron-based multiple branch predictor. For the verification, various hardware configurations are simulated using SPEC2000 benchmarks programs as input. As a result, we show that the statistical simulation is quite accurate and time saving for the evaluation of microprocessor architectures with multiple branch prediction.