• Journal of the Korea Institute of Information Security & Cryptology
• /
• v.11 no.2
• /
• pp.3-12
• /
• 2001

Currently, the superscalar architecture dominates todays microprocessor marketplase. As, more transistors are integrated onto larger die, however, an on-chip multiprocessor is regarded as a promising alternative to the superscalar microprocessor. This paper examines the behavior of the next generation block encryption algorithms RC6 and Rijndael on the on-chip multiprocessing microprocessor. Based on the simulation results by using a program-driven simulator, the on-chip multiprocessor can exploit thread level parallelism effectively and overcome the limitation of instruction level parallelism in the next generation block encryption algorithms.

• International Journal of Control, Automation, and Systems
• /
• v.1 no.3
• /
• pp.339-350
• /
• 2003

In this paper, we propose a simultaneous multithreading (SMT) architecture that improves instruction throughput by exploiting instruction level parallelism (ILP) and thread level parallelism (TLP). The proposed architecture issues and completes instructions belonging to the same thread in exact program order. The issue and completion policy greatly reduces the design complexity and hardware cost of our architecture, compared with others that employ out-of-order issue and completion. On the other hand, when the instructions belong to different threads, the issue and completion orders for those instructions may not necessarily be identical to the fetch order. The processor issues instructions simultaneously from multiple threads to functional units by exploiting ILP and TLP, and by dynamic resource sharing. That parallel execution notably improves performance and resource utilization with minimal additional hardware cost over the conventional superscalar processors. This paper proposes an SMT architecture with grouping as well as one without grouping. Without grouping, all threads dynamically and flexibly share most resources. On the other hand, in the SMT architecture with grouping, in which resources and threads are divided into several groups for design simplification, resources are shared only among threads belonging to the same group as those resources. Simulation results show that our processors with four and eight threads improve performance by three or more times over the conventional superscalar processor with comparable execution resources and policies, and that reasonable grouping reduces the design complexity of SMT processors with little negative effect on performance.

• The KIPS Transactions:PartA
• /
• v.15A no.5
• /
• pp.259-268
• /
• 2008

Multi-Core processors have become main-stream microprocessors in recent years. Servers based on these multi-core processors are widely adopted in High Performance Computing (HPC) and commercial business applications as well. These servers provide increased level of parallelism, thus can potentially boost the performance for applications. However, the shared resources among multiple cores on the same chip can become hot spots and act as performance bottlenecks. Therefore it is essential to optimize the use of shared resources for high performance and scalability for the multi-core servers. In this paper, we conduct experimental studies to analyze the positive and negative effects of the resource sharing on the performance of HPC applications. Through the analyses we also characterize the performance of multi-core servers.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.49 no.8
• /
• pp.1-9
• /
• 2012

Quantum-Inspired Evolutionary Algorithm(QEA) contains sufficient data-level parallelism to be naturally accelerated on GPUs. For an efficient reduction of execution time, however, careful task-mapping should be done to properly reflect the characteristics of CPU and GPU. Furthermore, when deciding which part of the application should run on GPU, we need to consider the data transfer between CPU and GPU memory spaces as well as the data-level parallelism. In addition, the usage of zero-copy host memory, proper choice of the execution configuration, and thread organization considering memory coalescing is important to further reduce the execution time. With all these techniques, we could run QEA 3.69 times faster on average in comparison with the multi-threading CPU for the case of 0-1 knapsack problem with 30,000 items.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.44 no.2
• /
• pp.61-71
• /
• 2007

The instruction level parallelism, which has been used to improve the performance of processors, expose its limit. The change of a control flow by a branch miss prediction is one of the obstacles that restrict the instruction level parallelism. The single chip multiprocessors have been developed to utilize the thread level parallelism. However, we could not use the maximum performance of the single chip multiprocessor in case of executing the coded programs without considering the multi-thread. In order to overcome the two performance degradation factors, in this paper, we suggest the concurrent branch execution method that applies to the multi-path execution method at a single chip multiprocessor. We executes all two flows of the conditional branch using the idle core processor. Through this, we can improve the processor's efficiency with blocking the control flow termination by the branch instruction and reducing the idle time. We analyze the effects of concurrent branch execution proposed in this paper through the simulation. As a result of that, concurrent branch execution reduces about 20% of idle time and improves the maximum 10% of the branch prediction accuracy. We show that our scheme improves the overall performance of maximum 39% compared to the normal single chip multiprocessor and maximum 27% compared to the superscalar processor.

• ETRI Journal
• /
• v.30 no.4
• /
• pp.576-586
• /
• 2008

In most parallel loops of embedded applications, every iteration executes the exact same sequence of instructions while manipulating different data. This fact motivates a new compiler-hardware orchestrated execution framework in which all parallel threads share one fetch unit and one decode unit but have their own execution, memory, and write-back units. This resource sharing enables parallel threads to execute in lockstep with minimal hardware extension and compiler support. Our proposed architecture, called multithreaded lockstep execution processor (MLEP), is a compromise between the single-instruction multiple-data (SIMD) and symmetric multithreading/chip multiprocessor (SMT/CMP) solutions. The proposed approach is more favorable than a typical SIMD execution in terms of degree of parallelism, range of applicability, and code generation, and can save more power and chip area than the SMT/CMP approach without significant performance degradation. For the architecture verification, we extend a commercial 32-bit embedded core AE32000C and synthesize it on Xilinx FPGA. Compared to the original architecture, our approach is 13.5% faster with a 2-way MLEP and 33.7% faster with a 4-way MLEP in EEMBC benchmarks which are automatically parallelized by the Intel compiler.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.27 no.5C
• /
• pp.512-521
• /
• 2002

Recently, conventional superscalar RISC processors arrive their performance limit, and many researches on the next-generation architecture are concentrated on SMT(Simultaneous Multi-Threading). In SMT processors, multiple threads are executed simultaneously and share hardware resources dynamically. In this case, it is more important to supply instructions from multiple threads to processor core efficiently than ever. Because SMT architecture shows higher IPC(Instructions per cycle) than superscalar architecture, performance is influenced by fetch bandwidth and the size of fetch queue. Moreover, to use TLP(Thread Level Parallelism) efficiently, fetch thread selection algorithm and fetch bandwidth for each selected threads must be carefully designed. Thus, in this paper, the performance values influenced by these factors are analyzed. Based on the results, an optimal instruction fetch strategy for SMT processors is proposed.

• Journal of the Institute of Electronics Engineers of Korea SD
• /
• v.46 no.8
• /
• pp.102-116
• /
• 2009

There have been lots of researches for H.264/AVC performance enhancement on a multi-core processor. The enhancement has been performed through parallelization methods. Parallelization methods can be classified into a task-level parallelization method and a data level parallelization method. A task-level parallelization method for H.264/AVC decoder is implemented by dividing H.264/AVC decoder algorithms into pipeline stages. However, it is not suitable for complex and large bitstreams due to poor load-balancing. Considering load-balancing and performance scalability, we propose a horizontal data level parallelization method for H.264/AVC decoder in such a way that threads are assigned to macroblock lines. We develop a mathematical performance expectation model for the proposed parallelization methods. For evaluation of the mathematical performance expectation, we measured the performance with JM 13.2 reference software on ARM11 MPCore Evaluation Board. The cycle-accurate measurement with SoCDesigner Co-verification Environment showed that expected performance and performance scalability of the proposed parallelization method was accurate in relatively high level

• ETRI Journal
• /
• v.22 no.4
• /
• pp.13-24
• /
• 2000

As more transistors are integrated onto bigger die, an on-chip multiprocessor will become a promising alternative to the superscalar microprocessor that dominates today's microprocessor marketplace. This paper describes key parts of a new on-chip multiprocessor, called Raptor, which is composed of four 2-way superscalar processor cores and one graphic co-processor. To obtain performance characteristics of Raptor, a program-driven simulator and its programming environment were developed. The simulation results showed that Raptor can exploit thread level parallelism effectively and offer a promising architecture for future on-chip multi-processor designs.

• Journal of the Semiconductor & Display Technology
• /
• v.12 no.3
• /
• pp.13-18
• /
• 2013

Motion estimation is a key technique of modern video processing that significantly improves the coding efficiency significantly by exploiting the temporal redundancy between successive frames. Thread-level parallelism is a promising method to accelerate the motion estimation process for multithreading general-purpose processors. In this paper, we propose a parallel motion estimation algorithm which parallelizes the motion search process of the current H.264/AVC encoder. The proposed algorithm is implemented using the OpenMP application programming interface (API) and can be easily integrated into the current encoder. The experimental results show that the proposed parallel algorithm can reduce the processing time of the motion estimation up to 65.08% without any penalty in the rate-distortion (RD) performance.