T
Tara M. Madhyastha
Researcher at University of California, Santa Cruz
Publications - 18
Citations - 751
Tara M. Madhyastha is an academic researcher from University of California, Santa Cruz. The author has contributed to research in topics: File system & Input/output. The author has an hindex of 14, co-authored 18 publications receiving 734 citations. Previous affiliations of Tara M. Madhyastha include University of Illinois at Urbana–Champaign.
Papers
More filters
Proceedings ArticleDOI
Data mining meets performance evaluation: fast algorithms for modeling bursty traffic
TL;DR: A simple, parsimonious method, the b-model, which solves both problems: it requires just one parameter, and can easily generate large traces, and has many more attractive properties.
Proceedings ArticleDOI
Scalable Performance Environments for Parallel Systems
Daniel A. Reed,R.D. Olson,Ruth A. Aydt,Tara M. Madhyastha,T. Birkett,David W. Jensen,Bobby A.A. Nazief,B. K. Totty +7 more
TL;DR: The environment prototype contains a set of performance data transformation modules that can be interconnected in user-specified ways and allows users to interconnect and configure modules graphically to form an acyclic, directed data analysis graph.
Proceedings ArticleDOI
Input/output access pattern classification using hidden Markov models
TL;DR: A new method for access pattern classification is examined that uses hidden Markov models, trained on access patterns from previous executions, to create a probabilistic model of input/output accesses, and is compared to a neural network classification &n-rework.
Journal ArticleDOI
Data sonification: do you see what I hear?
TL;DR: The authors assert that, despite great strides in developing the graphical dimension of user interfaces, the auditory dimension has been neglected, and offer a tool for using sound to complement visual cues when working with complex data.
Journal ArticleDOI
Learning to classify parallel input/output access patterns
TL;DR: This paper examines and compares two novel input/output access pattern classification methods based on learning algorithms, and proposes a method for forming global classifications from local classifications for parallel file system performance.