• Proceedings of the Korea Information Processing Society Conference
• /
• 2005.05a
• /
• pp.1757-1760
• /
• 2005

요구 페이징 방식의 가상메모리 시스템에서 페이지 참조 스트링에 따른 페이지 교체 알고리즘의 성능평가를 위한 시뮬레이션을 개발한다. 참조 집약성을 기반으로 워킹세트(working set) 모델을 수정한 MWS 교체 알고리즘을 제안하였고 다양한 교체 알고리즘(FIFO, SC, LFU, LRU, Rand)들과 비교 실험한 결과 MWS는 발생 페이지 부재 수 측면에서 다른 교체 정책 보다 성능이 우수하였다.

• The Korean Journal of Emergency Medical Services
• /
• v.11 no.2
• /
• pp.87-101
• /
• 2007

Health and disease related characteristics of 226 selected by systematic sampling from 452 personal emergency response system(PERS) beneficiaries and actual conditions of using PERS by them are as follows. Over 86% of PERS beneficiaries have not good health conditions and 70.7% of them have chronic diseases. On social supports family was highest as 52.2% and cases having social workers' assistance were 15.2%, but 53 as 23.7% had not visit or call from anyone and showed very low social supports. 86.2% was given PERS within 3years and 79.1% had it by the recommendation from related agencies and 4.0% was by their demand. On wearing it, 78.3% didn't bring it with them and 92% of them answered they were not sick and then it was found that they didn't use it because they had not special emergency. On satisfaction with paging system's operation, 81.3% answered they were satisfied with it, 48.8% used it for 'acute and emergency diseases' and 29.3% called ambulance for 'appointed medical treatment'. Time required for ambulance to arrive at the field was within 10 min. in 87.8% and after 10 min. in 12.2% and emergency service for beneficiaries by fire service was very good. On satisfaction with use of PERS, 85.4% were satisfied with it, 81.9% who requested repair or replacement of radio paging got it back after one or two days of their request and they answered they were satisfied with A/S. 45.5% answered they powered off it because 'they didn't use it' and 12.1% had 'economical reason of phone charge'.

• Proceedings of the Korean Information Science Society Conference
• /
• 2000.10c
• /
• pp.97-99
• /
• 2000

운영체제의 성능을 향상시키기 위해서는 효율적인 메모리 관리가 필수적이다. 시스템에서는 존재하는 크기의 메모리보다 더 큰 크기의 메모리 공간을 요구하는 경우가 빈번히 발생한다. 이 문제를 극복하기 위해서 운영체제에서는 가상 메모리(virtual memory)를 사용한다. 리눅스 가상 메모리 지원 방식에는 요구 페이징(demand paging), 스왑핑(swapping), 공유 가상 메모리(shared virtual memory)가 있다. 이런 방식을 효율적으로 지원하기 위해서 리눅스는 캐시를 사용한다. 본 논문에서는 페이지 해시 리스트에서 페이지를 찾는 알고리즘을 수정하여 새로운 페이지 캐시 관리 기법을 제안한다. 이 방법은 탐색한 페이지를 해시 리스트의 헤드(head)로 옮김으로써 다음 탐색 때 그 페이지를 찾는데 필요한 탐색 회수를 줄일 수 있다는 장점을 갖는다. 또한 다른 프로세스에 의해서 동시에 많이 참조되는 페이지들은 탐색시간이 줄어들게 된다. 시뮬레이션 프로그램을 통해 본 논문에서 제안한 수정된 페이지 캐시 관리 기법을 이용하면, 기존의 방법에 비해서 페이지를 찾는데 필요한 탐색 회수와 탐색 시간의 측면에서 성능이 향상됨을 보인다.

• The KIPS Transactions:PartA
• /
• v.8A no.4
• /
• pp.469-478
• /
• 2001

In the virtual memory system, page replacement policy exerts a great influence on the performance of demand paging. There are LRU(Least Recently Used) and LFU (Least Frequently Used) as the typical replacement policies. The LRU policy performs effectively in many cases and adapts well to the changing workloads compared to other policies. It however cannot distinguish well between frequently and infrequently referenced pages. The LFU policy requires that the page with the smallest reference count be replaced. Though it considers all the references in the past, it cannot discriminate between references that occurred far back in the past and the more recent ones. Thus, it cannot adapt well to the changing workload. In this paper, we first analyze memory reference patterns of eight applications. The patterns show that the recently referenced pages or the frequently referenced pages are accessed continuously as the case may be. So it is rather hard to optimize page replacement scheme by using just one of the LRU or LFU policy. This paper makes an attempt to combine the advantages of the two policies and proposes a new page replacement policy. In the proposed policy, paging list is divided into two lists (LRU and LFU lists). By keeping the two lists in recency and reference frequency order respectively, we try to restrain the highly referenced pages in the past from being replaced by the LRU policy. Results from trace-driven simulations show that there exists points on the spectrum at which the proposed policy performs better than the previously known policies for the workloads we considered. Especially, we can see that our policy outperforms the existing ones in such applications that have reference patterns of re-accessing the frequently referenced pages in the past after some time.