The Community for Technology Leaders
2012 IEEE Fifth International Conference on Cloud Computing (2012)
Honolulu, HI, USA USA
June 24, 2012 to June 29, 2012
ISSN: 2159-6182
ISBN: 978-1-4673-2892-0
pp: 147-154
As cloud diversifies into different application fields, understanding and characterizing the specific workloadsand application requirements play important roles in thedesign of efficient cloud infrastructure and system softwaresupport. Video analytic is a rapidly advancing field and it iswidely used in many application domains (i.e., health, medicalcare, surveillance, and defense). To support video analyticapplications efficiently in cloud, one has to overcome manychallenges such as lack of understanding of the relationship andtradeoff between analytic performance metrics and resourcerequirements. Furthermore, cloud computing has grown fromthe early model of resource sharing to data sharing andworkflow sharing. To address the challenges and to leverageemerging trends, we propose and experiment with a domainspecific cloud environment for video analytic applications. Wedesign a cloud infrastructure framework for sharing videodata, analytic software, and workflow. In addition, we create avideo analytic quality aware resource plan model to guaranteeusers QoS and optimize usage of resources based on predictiveknowledge of video analytic softwares performance metrics anda resource planning model that optimizes the overall analyticservice quality under users constraints (i.e., time and cost).The predictive knowledge is represented as input and analyticsoftware specific predictors. The experimental results show thatthe video analytic quality aware resource planning model canbalance the tradeoff between analytic quality and resourcerequirements, and achieve optimal or near-optimal planning forvideo analytic workloads with constraints in a resource sharedenvironment. Simulation studies show that resource planningresults using ground truth and video analytic performancepredictions are very similar, which indicates that our analytic quality/resource predictors are very accurate.
Software, Streaming media, Prediction algorithms, Planning, Analytical models, Algorithm design and analysis, Measurement, Planning, Cloud Computing, Video Analytic, Quality Prediction

J. Lee, T. Feng, W. Shi, A. Bedagkar-Gala, S. K. Shah and H. Yoshida, "Towards Quality Aware Collaborative Video Analytic Cloud," 2012 IEEE Fifth International Conference on Cloud Computing(CLOUD), Honolulu, HI, USA USA, 2012, pp. 147-154.
185 ms
(Ver 3.3 (11022016))