Background traffic modeling for large-scale network simulation

Network simulation is an indispensable tool for studying Internet-scale networks due to the heterogeneous structure, immense size and changing properties. It is crucial for network simulators to generate representative traffic, which is necessary for effectively evaluating next-generation network protocols and applications. With network simulation, we can make a distinction between foreground traffic, which is generated by the target applications the researchers intend to study and therefore must be simulated with high fidelity, and background traffic, which represents the network traffic that is generated by other applications and does not require significant accuracy. The background traffic has a significant impact on the foreground traffic, since it competes with the foreground traffic for network resources and therefore can drastically affect the behavior of the applications that produce the foreground traffic. This dissertation aims to provide a solution to meaningfully generate background traffic in three aspects. First is realism. Realistic traffic characterization plays an important role in determining the correct outcome of the simulation studies. This work starts from enhancing an existing fluid background traffic model by removing its two unrealistic assumptions. The improved model can correctly reflect the network conditions in the reverse direction of the data traffic and can reproduce the traffic burstiness observed from measurements. Second is scalability. The trade-off between accuracy and scalability is a constant theme in background traffic modeling. This work presents a fast rate-based TCP (RTCP) traffic model, which originally used analytical models to represent TCP congestion control behavior. This model outperforms other existing traffic models in that it can correctly capture the overall TCP behavior and achieve a speedup of more than two orders of magnitude over the corresponding packet-oriented simulation. Third is network-wide traffic generation. Regardless of how detailed or scalable the models are, they mainly focus on how to generate traffic on one single link, which cannot be extended easily to studies of more complicated network scenarios. This work presents a cluster-based spatio-temporal background traffic generation model that considers spatial and temporal traffic characteristics as well as their correlations. The resulting model can be used effectively for the evaluation work in network studies.

Optimization of three-dimensional branching networks of microchannels for thermal management of microelectronics

The aim of this work is to present a methodology to develop cost-effective thermal management solutions for microelectronic devices, capable of removing maximum amount of heat and delivering maximally uniform temperature distributions. The topological and geometrical characteristics of multiple-story three-dimensional branching networks of microchannels were developed using multi-objective optimization. A conjugate heat transfer analysis software package and an automatic 3D microchannel network generator were developed and coupled with a modified version of a particle-swarm optimization algorithm with a goal of creating a design tool for 3D networks of optimized coolant flow passages. Numerical algorithms in the conjugate heat transfer solution package include a quasi-ID thermo-fluid solver and a steady heat diffusion solver, which were validated against results from high-fidelity Navier-Stokes equations solver and analytical solutions for basic fluid dynamics test cases. Pareto-optimal solutions demonstrate that thermal loads of up to 500 W/cm2 can be managed with 3D microchannel networks, with pumping power requirements up to 50% lower with respect to currently used high-performance cooling technologies.

Background Traffic Modeling for Large-Scale Network Simulation

Network simulation is an indispensable tool for studying Internet-scale networks due to the heterogeneous structure, immense size and changing properties. It is crucial for network simulators to generate representative traffic, which is necessary for effectively evaluating next-generation network protocols and applications. With network simulation, we can make a distinction between foreground traffic, which is generated by the target applications the researchers intend to study and therefore must be simulated with high fidelity, and background traffic, which represents the network traffic that is generated by other applications and does not require significant accuracy. The background traffic has a significant impact on the foreground traffic, since it competes with the foreground traffic for network resources and therefore can drastically affect the behavior of the applications that produce the foreground traffic. This dissertation aims to provide a solution to meaningfully generate background traffic in three aspects. First is realism. Realistic traffic characterization plays an important role in determining the correct outcome of the simulation studies. This work starts from enhancing an existing fluid background traffic model by removing its two unrealistic assumptions. The improved model can correctly reflect the network conditions in the reverse direction of the data traffic and can reproduce the traffic burstiness observed from measurements. Second is scalability. The trade-off between accuracy and scalability is a constant theme in background traffic modeling. This work presents a fast rate-based TCP (RTCP) traffic model, which originally used analytical models to represent TCP congestion control behavior. This model outperforms other existing traffic models in that it can correctly capture the overall TCP behavior and achieve a speedup of more than two orders of magnitude over the corresponding packet-oriented simulation. Third is network-wide traffic generation. Regardless of how detailed or scalable the models are, they mainly focus on how to generate traffic on one single link, which cannot be extended easily to studies of more complicated network scenarios. This work presents a cluster-based spatio-temporal background traffic generation model that considers spatial and temporal traffic characteristics as well as their correlations. The resulting model can be used effectively for the evaluation work in network studies.

The effect of reciprocal mapping on high-risk sixth-grade students' social studies achievement

Reading deficits in students in Grades 4 to 12 are evident in American schools. Informational text is particularly difficult for students. This quasi-experimental study (N=138) investigated sixth-grade students' achievement in social studies using the Reciprocal Mapping instructional routine, compared to sixth-grade students' achievement taught with a traditional approach. The Reciprocal Mapping instructional routine incorporated explicit instruction in text structure using graphic organizers. Students created their own graphic organizers and used them to write about social studies content. The comparison group used a traditional approach, students' reading the textbook and answering questions. Students for this study included sixth-graders in the seven sixth-grade classrooms in two public schools in a small, rural south Florida school district. A focus of this study was to determine the helpfulness of the intervention for at-risk readers. To determine students considered to be at-risk, the researcher used data from the reading portion of the Florida Comprehensive Assessment Test (FCAT), 2011-2012, that considers Level 1 and 2 as at-risk readers. The quasi-experimental study used a pretest-posttest control group design, with students assigned to treatment groups by class. Two teachers at the two rural sites were trained on the Reciprocal Mapping instructional routine and taught students in both the experimental and control groups for an equivalent amount of time over a 5-week period. Results of the 3 x 2 factorial ANCOVA found a significant positive difference favoring the experimental group's social studies achievement as compared to that of the comparison group as measured by the pre/post unit test from the social studies series (McGraw-Hill, 2013), when controlling for initial differences in students' reading FCAT scores. Interactions for high-risk struggling readers were investigated using the significance level p < .05. Due to no significant interaction the main effects of treatment were interpreted. The pretest was used as a covariate and the multivariate analysis was found to be significant. Therefore, analysis of covariance was run on each of the dependent variable as a follow-up. Reciprocal Mapping was found to be significant in posttest scores, independent of gender and level of risk, and while holding the pretest scores constant. Findings showed there was a significant difference in the performance of the high-risk reading students taught with the Reciprocal Mapping intervention who scored statistically better than students in the control group. Further study findings showed that teacher fidelity of implementation of the treatment had a statistically significant relationship in predicting posttest scores when controlling for pretest scores. Study results indicated that improving students' use of text structure through the Reciprocal Mapping instructional routine positively supported sixth-grade students' social studies achievement.