2015, Vol.20, No.9

Journal of The Korea Society of Computer and Information 2023 KCI Impact Factor : 0.65

pISSN : 1598-849X / eISSN : 2383-9945
https://journal.kci.go.kr/jksci

In this paper, we propose a new variable latency L1 data cache architecture for multi-core processors. Our proposed architecture extends the traditional variable latency cache to be geared toward the multi-core processors. We added a specialized data structure for recording the latency of the L1 data cache. Depending on the added latency to the L1 data cache, the value stored to the data structure is determined. It also tracks the remaining cycles of the L1 data cache which notifies data arrival to the reservation station in the core. As in the variable latency cache of the single-core architecture, our proposed architecture flexibly extends the cache access cycles considering process variation. The proposed cache architecture can reduce yield losses incurred by L1 cache access time failures to nearly 0%. Moreover, we quantitatively evaluate performance, power, energy consumption, power-delay product, and energy-delay product when increasing the number of cache access cycles.

In order to prevent the elderly people from committing suicide attempts, it is necessary to verify attributes that affect the suicide attempts. It is noted that previous studies have focused on qualitative approaches, and simple correlation analyses to determine the attributes related to the suicide attempts in the elderly people. However, such previous approaches had led to insufficient performance when facing with complicated data sets. In this sense, this study suggests an alternative method in which attribute selection techniques are adopted to determine more relevant attributes of the suicide attempts occurring in Korean elderly people. To verify empirical validity of our proposed method, we used Korea National Health and Nutrition Examination Survey (KNHANES) from January 2007 to December 2012. Empirical results proved that the proposed attribute selection techniques showed better predictive effectiveness; 94.4% compared to the simple statistical methods. This study proposes a way to determining the elderly suicide and preventing it to happen.

The set partitioning problem is a well-known NP-hard combinatorial optimization problem, and it is formulated as an integer programming model. This paper proposes an Integer Programming-based Local Search for solving the set partitioning problem. The key point is to solve the set partitioning problem as the set covering problem. First, an initial solution is generated by a simple heuristic for the set covering problem, and then the solution is set as the current solution. Next, the following process is repeated. The original set covering problem is reduced based on the current solution, and the reduced problem is solved by Integer Programming which includes a specific element in the objective function to derive the solution for the set partitioning problem. Experimental results on a set of OR-Library instances show that the proposed algorithm outperforms pure integer programming as well as the existing heuristic algorithms both in solution quality and time.