2015, Vol.10, No.2

Journal of Knowledge Information Technology and Systems 2023 KCI Impact Factor : 0.89

pISSN : 1975-7700 / eISSN : 2734-0570
https://journal.kci.go.kr/kkits

Reduction in cache miss rates continues to be an important issue in processor design, especially at the last level cache(LLC). The Least Recently Used(LRU) replacement policy has been widely adapted in processors for the past decades. The LRU replacement policy represents the cache blocks in a set as LRU stack. So, it only use the recency information. And it is expensive to implement in hardware. Also in some workload, it raises the thrashing, i.e., blocks with high reuse evicting each other from the cache. We propose an new replacement policy to solve problems of LRU replacement policy. In the proposed policy, a incoming block is inserted at the bottom of the set. To exploit the frequency information, a hit block exchanges its position with its adjacent block above in the set. A victim block can be chosen from bottom. Selecting the bottom block as the victim makes the proposed policy to protect the cache from polluting by less frequently used blocks. It also proposes to divide the blocks in a cache set into groups to resolve the thrashing. The victim block is selected alternately from each group. The proposed policy reduces the average MPKI(Miss Per Kilo Instructions) of the baseline 1MB 16-way LLC cache using LRU replacement policy by 12.9%. It reduces the storage requirement by 47% compared to LRU replacement policy.

There are mass productions and collections of information in different fields with real time in our present society. Big data is not limited as a specific single system in its information collection process and produces useful information through collecting different data from multiple systems. Thus, there are rapid increases in users who abuse such information through accessing specific distributed systems by collecting and combining the information. Therefore, it is considered that damages caused by the information abuse are very serious. In this study, trace back information has primarily used in a process for collecting data in the different information environments that manage the data as a distributed storage manner in order to react currently increased illegal data collections. In addition, the data is to be classified according to values of the data and the values are also evaluated and managed through applying proper security classes. Then, the data that causes significant damages from combinations of it due to leaks of the data is to be managed. Based on the management, a way of performing stable information provides and collections under the current situation that represents rapidly and differently changed environments for collecting information is proposed.

It is very important for a book reader to choose an appropriate book which is suitable for the reader's reading-level for a successful reading. Especially, when a reader choose a foreign book, more attention needs to be paid because the book's readability-level and the reader's reading-level have to be considered simultaneously. In case of English books that are sold or published in Korea, somewhat objective readability levels of them can be estimated to some degree by referring to the levels of difficulty which are developed by institutes in the U.S.A. or Korea. However, because reader's reading-levels can be estimated through a long English test, many readers avoid taking such a test and subjectively estimate their reading-level. They are therefore likely to choose an English book which are not suitable for their reading-level. This paper, for Korean readers whose mother tongue is not English, proposes an algorithm which can help a reader find out his/her reading-level suitability for an English book by a short time English vocabulary test. We present our systematic process for extracting test words from a target English book and for estimating the level suitability between the book and the reader. In order to evaluate our algorithm, we implement it in a Web application and show survey results.