DHOOT, ANUBHAV V. Hybrid online/offline optimization of application binaries (Under the direction of Assistant Professor Dr. Frank Mueller). Long-running parallel applications suffer from performance limitations particu-larly due to inefficiencies in accessing memory. Dynamic optimizations, i

instrumental tool of management of new forms of life, including the building of on-line communities of support and collective learning. I see, however, a much stronger connection between networks and relationships of experience through the cultural transformations induced by communication networks

(SDSC) and Hydra at Texas A&M Supercomputing Facilities, and quantify the performance gap resulting from using different number of processors per node. We use PAIDE to instrument the source code of MrBayes to collect the performance, upload the performance data to Prophesy database, and then use

Abstract—Automating the process of parallel performance experimentation, analysis, and problem diagnosis can enhance environments for performance-directed application development, compilation, and execution. This is especially true when paramet-ric studies, modeling, and optimization strategies

Public reporting burden for the collection of information is estimated to average 1 hour per response, including the time for reviewing instructions, searching existing data sources, gathering and maintaining the data needed, and completing and reviewing the collection of information. Send comments

considered in performance predictions. Most existing approaches use established prediction models [3, 18] to estimate the performance of already existing complex software systems. Their main focus lies on the questions: i) ”How can we automatically derive or extract the models we need? ” and ii

Despite being a core networking primitive, collection protocols today often suffer from poor reliability (e.g., 70%) in practice, and heavily used protocols have never been evaluated in terms of communication efficiency. Using detailed experimental studies, we describe three challenges that cause

Vast majority of wireless appliances used in household, industry and medical field share the ISM frequency band. These devices need to coexist and thus are challenged to tolerate their mutual interference. One way of dealing with this is by using frequency hopping; where the device changes its radio channel periodically. Consequently, communications will not suffer from the same interference each time; instead, it should be fairer and more stable. This thesis investigates the aforementioned problem in the field of low power wireless sensor networks and Internet of Things where Contiki OS is used. We introduce a low-power pseudo-random frequency-hopping MAC protocol which is specifically characterized as a duty cycled asynchronous sender-initiated LPL style protocol. We illustrate two flavors of the protocol; one that does not use any dedicated channel and another which allows dedicated broadcast channels that can implement frequency-hopping as well. We implement

Abstract not found

Abstract. We review theoretical models of individual motility as well as collective dynamics and pattern formation of active particles. We focus on simple models of active dynamics with a particular emphasis on nonlinear and stochastic dynamics of such self-propelled entities in the framework