Error Tolerant Multimedia Stream Processing: There's Plenty of Room at the Top (of the System Stack)

作者: Yiannis Andreopoulos

DOI: 10.1109/TMM.2012.2232649

关键词: Computer scienceVideo processingSystems designStream processingRobustness (computer science)Logic synthesisMultimediaSoftware

摘要: There is a growing realization that the expected fault rates and energy dissipation stemming from increases in CMOS integration will lead to abandonment of traditional system reliability favor approaches offer hardware-induced errors across application, runtime support, architecture, device integrated-circuit (IC) layers. Commercial stakeholders multimedia stream processing (MSP) applications, such as information retrieval, mining systems, high-throughput image video systems already feel strain inadequate system-level scaling robustness under always-increasing user demand. While applications can tolerate certain imprecision their results, today's MSP do not support systematic way exploit this aspect for cross-layer resilience. However, research currently emerging attempts utilize error-tolerant nature purpose. This achieved by modifications all layers stack, algorithms software architecture layer, even IC digital logic synthesis itself. Unlike conventional aims worst-case performance accuracy guarantees, provide guarantees accuracy. In paper we review recent advances field an (layer-by-layer) perspective, attempt foresee some components future design may influence general computing landscape within next ten years.

参考文章(101)
Steven Hand, Derek G. Murray, Spread-spectrum computation hot topics in system dependability. pp. 5- 5 ,(2008)
Cordelia Schmid, Roger Mohr, Christian Bauckhage, Evaluation of Interest Point Detectors International Journal of Computer Vision. ,vol. 37, pp. 151- 172 ,(2000) , 10.1023/A:1008199403446
Zheng Zhu, Ingemar J. Cox, Mark Levene, Ranked-Listed or Categorized Results in IR: 2 Is Better Than 1 applications of natural language to data bases. pp. 111- 123 ,(2008) , 10.1007/978-3-540-69858-6_12
H. Asthana, Ruoxun Fu, Ingemar J. Cox, On the feasibility of unstructured peer-to-peer information retrieval international conference on the theory of information retrieval. pp. 125- 138 ,(2011) , 10.1007/978-3-642-23318-0_13
Brian Foo, Yiannis Andreopoulos, Mihaela van der Schaar, Analytical Rate-Distortion-Complexity Modeling of Wavelet-Based Video Coders IEEE Transactions on Signal Processing. ,vol. 56, pp. 797- 815 ,(2008) , 10.1109/TSP.2007.906685
S.L. Regunathan, P.A. Chou, J. Ribas-Corbera, A generalized video complexity verifier for flexible decoding international conference on image processing. ,vol. 3, pp. 289- 292 ,(2003) , 10.1109/ICIP.2003.1247238
Kosmas Petridis, Dionysios Anastasopoulos, Carsten Saathoff, Norman Timmermann, Yiannis Kompatsiaris, Steffen Staab, M-OntoMat-Annotizer: Image Annotation Linking Ontologies and Multimedia Low-Level Features Lecture Notes in Computer Science. pp. 633- 640 ,(2006) , 10.1007/11893011_80
Thomas Y. Yeh, Glenn Reinman, Sanjay J. Patel, Petros Faloutsos, Fool me twice ACM Transactions on Graphics. ,vol. 29, pp. 1- 11 ,(2009) , 10.1145/1640443.1640448
A. Chandrakasan, T. Xanthopoulos, V. Gutnik, Data driven signal processing: an approach for energy efficient computing international symposium on low power electronics and design. pp. 347- 352 ,(1996) , 10.5555/252493.252633