• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.11 no.2
• /
• pp.670-686
• /
• 2017

MapReduce (MRV1), a popular programming model, proposed by Google, has been well used to process large datasets in Hadoop, an open source cloud platform. Its new version MapReduce 2.0 (MRV2) developed along with the emerging of Yarn has achieved obvious improvement over MRV1. However, MRV2 suffers from long finishing time on certain types of jobs. Speculative Execution (SE) has been presented as an approach to the problem above by backing up those delayed jobs from low-performance machines to higher ones. In this paper, an adaptive SE strategy (ASE) is presented in Hadoop-2.6.0. Experiment results have depicted that the ASE duplicates tasks according to real-time resources usage among work nodes in a cloud. In addition, the performance of MRV2 is largely improved using the ASE strategy on job execution time and resource consumption, whether in a multi-job environment.

• The Transactions of the Korea Information Processing Society
• /
• v.6 no.2
• /
• pp.497-511
• /
• 1999

Recently, many parallel processing technologies were developed, ILP(Instruction level Parallelism) processor's performance have been growed very rapidly. especially, EPIC(Explicitly Parallel Instruction computing) architectures attempt to enhance the performance in the predicated execution and speculative execution with the hardware. In this paper to improve the code scheduling possibility by applying to the characteristics of EPIC architectures, a new register allocation algorithm is proposed. And we proves that proposed register allocation algorithm is more efficient scheme than the conventional scheme when predicated execution is applied to our scheme by experiments. In experimental results, it shows much more performance enhancement, about 19% in proposed scheme than the conventional scheme. So, our scheme is verified that it is an effective register allocation method.

• The KIPS Transactions:PartA
• /
• v.13A no.6 s.103
• /
• pp.489-494
• /
• 2006

This paper improves the algorithms for Speculative Partial Redundancy Elimination(SPRE) proposed by Knoop et al. This research brings up an issue concerning a field to which SPRE cannot be applied, and suggests a solution to the problem. The Improved SPRE algorithm performs the execution speed optimization based on the information on the execution frequency from profiling and the memory space optimization.

• The KIPS Transactions:PartA
• /
• v.15A no.4
• /
• pp.217-226
• /
• 2008

Branch prediction accuracy is critical for system performance in modern microprocessor architectures. The use of speculative update branch history provides substantial accuracy improvement in branch prediction. However, speculative update branch history is the information about uncommitted branch instruction and thus it may hurts program correctness, in case of miss-speculative execution. Therefore, speculative update branch history requires suitable recovery mechanisms to provide program correctness as well as performance improvement. In this paper, we propose recovery logics for speculative update branch history. The proposed solutions are recovery logics for both global history and local history. In simulation results, our solution provides performance improvement up to 5.64%. In addition, it guarantees the program correctness and almost 90% of additional hardware overhead is reduced, compared to previous works.

• The KIPS Transactions:PartA
• /
• v.19A no.2
• /
• pp.77-84
• /
• 2012

Speculative execution for improving instruction-level parallelism is widely used in high-performance processors. In the speculative execution technique, the most important factor is the accuracy of branch predictor. Unfortunately, complex branch predictors for improving the accuracy can cause serious thermal problems in 3D multicore processors. Thermal problems have negative impact on the processor performance. This paper analyzes two methods to solve the thermal problems in the branch predictor of 3D multi-core processors. First method is dynamic thermal management which turns off the execution of the branch predictor when the temperature of the branch predictor exceeds the threshold. Second method is thermal-aware branch predictor placement policy by considering each layer's temperature in 3D multi-core processors. According to our evaluation, the branch predictor placement policy shows that average temperature is $87.69^{\circ}C$, and average maximum temperature gradient is $11.17^{\circ}C$. And, dynamic thermal management shows that average temperature is $89.64^{\circ}C$ and average maximum temperature gradient is $17.62^{\circ}C$. Proposed branch predictor placement policy has superior thermal efficiency than the dynamic thermal management. In the perspective of performance, the proposed branch predictor placement policy degrades the performance by 3.61%, while the dynamic thermal management degrades the performance by 27.66%.

• Journal of the Korea Computer Industry Society
• /
• v.7 no.3
• /
• pp.145-154
• /
• 2006

This paper improves algorithms of Speculative Partial Redundancy Elimination(SPRE) proposed by Knoop et al. Improving SPRE algorithm performs the execution speed optimization based on the information of the execution frequency from profiling and the memory space optimization. The first purpose of presented algorithm is to reduce in space requirements and the second purpose is to de crease the execution time. Since too much weight on execution speed optimization may cause the explosion of the memory space, it is important to consider the size of memory. This fact can be a big advantage in the embedded system which concerns the required memory size more than the execution speed In this paper we implemented the min-cut algorithm, and this algorithm used the control flow graph is constructed with network and partitioned.

• Journal of KIISE:Computer Systems and Theory
• /
• v.30 no.12
• /
• pp.724-735
• /
• 2003

In recent high-performance superscalar processors, the result value of an instruction is predicted to improve instruction-level parallelism by breaking data dependencies. Using those predicted values, instructions are speculatively executed and substantial performance can be gained. It, however, requires additional power consumption due to the frequent access and update of the value prediction table. In this paper, first, the trade-off between the performance improvement and the increased power consumption for value prediction is measured and analyzed. And, in order to reduce additional power consumption without performance loss, the technique of controlling speculative execution with confidence counter and predicting useful instructions is developed. Also, in order to prove the validity, a tool is developed that can simulate processor behavior at cycle-level and measure total energy consumption and power consumption per cycle.

• Agribusiness and Information Management
• /
• v.1 no.2
• /
• pp.43-60
• /
• 2009

To improve the performance of wide-issue superscalar processors, it is essential to increase the width of instruction fetch and the issue rate. Removal of control hazard has been put forward as a significant new source of instruction-level parallelism for superscalar processors and the conditional branch prediction is an important technique for improving processor performance. Branch mispredictions, however, waste a large number of cycles, inhibit out-of-order execution, and waste electric power on mis-speculated instructions. Hence, the branch predictor with higher accuracy is necessary for good processor performance. In global-history-based predictors like gshare and GAg, many mispredictions come from commit update of the branch history. Some works on this subject have discussed the need for speculative update of the history and recovery mechanisms for branch mispredictions. In this paper, we present a new mechanism for recovering the branch history after a misprediction. The proposed mechanism adds an age_counter to the original predictor and doubles the size of the branch history register. The age_counter counts the number of outstanding branches and uses it to recover the branch history register. Simulation results on the SimpleScalar 3.0/PISA tool set and the SPECINT95 benchmarks show that gshare and GAg with the proposed recovery mechanism improved the average prediction accuracy by 2.14% and 9.21%, respectively and the average IPC by 8.75% and 18.08%, respectively over the original predictor.

• Journal of the Institute of Electronics Engineers of Korea CI
• /
• v.48 no.3
• /
• pp.54-68
• /
• 2011

As the prediction accuracy of conditional branch instruction is increased highly, the importance of prediction accuracy for unconditional branch instruction is also increased accordingly. Except the case of RAS(Return Address Stack) overflow, the prediction accuracy of function return address should be 100% theoretically. However, there exist some possibilities of miss-predictions for RAS return addresses, when miss-speculative execution paths are invalidated, in case of modern speculative microprocessor environments. In this paper, we propose the RAS rename technique to prevent RAS pollution, results in the reduction of RAS miss-prediction. We divide a RAS stack into a soft-stack and a hard-stack and we handle the instructions for speculative execution in the soft-stack. When some overwrites happen in the soft-stack, we move the soft-stack data into the hard-stack. In addition, we propose an enhanced version of RAS rename scheme. In simulation results, our solution provide 1/90 reduction of miss-prediction of function return address, results in up to 6.85% IPC improvement, compared to normal RAS method. Furthermore, it reduce miss-prediction ratio as 1/9, compared to previous technique.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.20 no.5
• /
• pp.143-149
• /
• 2020

Currently, a multi-core processor is mainly used as a central processing unit of a computer system, and a high-performance out-of-order processor is adopted as each core to maximize system performance. The early out-of-order execution processor with Tomasulo algorithm aimed at floating-point instructions, and it took several cycles to execute by the use of complex structures such as reorder buffer and reservation station. However, in order for the processor to properly utilize out-of-order execution and increase the throughput of instructions, it must operate in a fully pipelined manner. In this paper, a fully pipelined out-of-order processor with speculative execution is designed with VHDL and verified with GHDL. As a result of the simulation, a program composed of ARM instructions is successfully performed.