http://chineseinput.net/에서 pinyin(병음)방식으로 중국어를 변환할 수 있습니다.
변환된 중국어를 복사하여 사용하시면 됩니다.
GPU기반의 보안 로그 이벤트 고속필터링기법에 대한 실증적 연구
안혜선(Hyeseon An),박제원(Jeawon Park),최재현(Jaehyun Choi),이남용(Namyong Lee) 한국정보기술학회 2013 한국정보기술학회논문지 Vol.11 No.9
Big data problem are expanded to a general issue not limited to some special areas. Security Management Systems put their data on relational DB. But if the number or size of data becomes bigger, its processing time is dramatically increased as their size, it isnt suitable structure for collecting big log data and may violates security policy. In this paper, in order to analyze big security logs and events in real time without high-cost of updating the system, we try a high speed filtering methods based on GPU(Graphic Processing Unit). After experimentation, it showed about 3,000 times more faster filtering execution time and there are more improvements possible if we can process maximum execution lines more than 30 lines by solving the memory assignment problem of GPU shared memory.