Adaptive eager-lazy hybrid evaluation of event patterns for low latency빠른 지연 시간을 위한 이벤트 패턴의 적응형 Eager-Lazy 하이브리드 평가

Cited 0 time in webofscience Cited 0 time in scopus
  • Hit : 614
  • Download : 0
Event pattern detection refers to identifying combinations of events matched to a user-specified query event pattern from a real-time event stream. Major applications of event pattern detection include fraud detection, anomaly detection, outlier detection, and intrusion detection. Latency, which is defined as the time elapsed between the arrival of the last event of an event combination matched to a query event pattern and the point at which the system identifies the event combination matched, is an important measure of the performance of an event pattern detection system. Existing methods can be classified into the eager evaluation method and the lazy evaluation method depending on when each event arrival is evaluated. These methods have advantages and disadvantages in terms of latency depending on the event arrival rate. The eager evaluation method evaluates all events immediately when they arrive while the lazy evaluation method defers evaluations of arriving events and later evaluates them in batch. The latency of the eager evaluation method is lower when the event arrival rate is slow while that of the lazy evaluation method, where optimization techniques can be used for batch processing, is lower when the event arrival rate is fast. In this paper, we propose a hybrid eager-lazy evaluation method that combines the advantages of both methods. For each event type, the hybrid method, which we call APAM (Adaptive Partitioning-And-Merging), determines which method to use: eager or lazy. It improves latency by using the eager evaluation method for more event types when the event arrival rate is slower and by using the lazy evaluation method for more event types when the event arrival rate is faster. That is, event types are partitioned into an eager evaluation group and a lazy evaluation group. We also propose a formal cost model to estimate the latency and propose a method of finding the optimal partition based on the cost model. Finally, we show through experiments that our method can improve the latency by up to 361.48 times over the eager evaluation method and 27.94 times over the lazy evaluation method using a synthetic and real-world data set.
Advisors
황규영researcherLee, Jae-Gilresearcher이재길researcherWhang, Kyu-Youngresearcher
Description
한국과학기술원 :전산학부,
Publisher
한국과학기술원
Issue Date
2016
Identifier
325007
Language
eng
Description

학위논문(박사) - 한국과학기술원 : 전산학부, 2016.8 ,[iv, 43 p. :]

Keywords

Complex event processing; Event pattern detection; Latency; Optimization; Hybrid; 컴플렉스 이벤트 처리; 이벤트 패턴 탐지; 지연 시간; 최적화; 하이브리드

URI
http://hdl.handle.net/10203/222429
Link
http://library.kaist.ac.kr/search/detail/view.do?bibCtrlNo=663215&flag=dissertation
Appears in Collection
CS-Theses_Ph.D.(박사논문)
Files in This Item
There are no files associated with this item.

qr_code

  • mendeley

    citeulike


rss_1.0 rss_2.0 atom_1.0