Digital Commons Network^™

A Deep Reinforcement Learning Approach With Prioritized Experience Replay And Importance Factor For Makespan Minimization In Manufacturing, Jose Napoleon Martinez

LSU Doctoral Dissertations

In this research, we investigated the application of deep reinforcement learning (DRL) to a common manufacturing scheduling optimization problem, max makespan minimization. In this application, tasks are scheduled to undergo processing in identical processing units (for instance, identical machines, machining centers, or cells). The optimization goal is to assign the jobs to be scheduled to units to minimize the maximum processing time (i.e., makespan) on any unit.

Machine learning methods have the potential to "learn" structures in the distribution of job times that could lead to improved optimization performance and time over traditional optimization methods, as well as to adapt …

Decentralized Scheduling For Many-Task Applications In The Hybrid Cloud, Brian Lyle Peterson

LSU Doctoral Dissertations

While Cloud Computing has transformed how we solve many computing tasks, some scientific and many-task applications are not efficiently executed on cloud resources. Decentralized scheduling, as studied in grid computing, can provide a scalable system to organize cloud resources and schedule a variety of work. By measuring simulations of two algorithms, the fully decentralized Organic Grid, and the partially decentralized Air Traffic Controller from IBM, we establish that decentralization is a workable approach, and that there are bottlenecks that can impact partially centralized algorithms. Through measurements in the cloud, we verify that our simulation approach is sound, and assess the …

Automated Generation And Visualization Of Initial Construction Schedules From Building Information Models, Yibrah Weldemihret Weldu

LSU Doctoral Dissertations

Recent advances in digital technology have had a significant influence on the quality and speed of sharing and communicating project information in the architecture, engineering, and construction (AEC) industry. The process of acquiring the design intent in order to develop and communicate project schedules, as critical components of project delivery, have similarly been benefitting from such progress. With the relatively recent techniques of Building Information Modeling (BIM) and its capability to integrate the facility design with its construction schedule, meaningul strides have been made in improving the information flow and eventually visualizing the final schedule in 4D. However, the need …

New Identification And Decoding Techniques For Low-Density Parity-Check Codes, Tian Xia

LSU Doctoral Dissertations

Error-correction coding schemes are indispensable for high-capacity high data-rate communication systems nowadays. Among various channel coding schemes, low-density parity-check (LDPC) codes introduced by pioneer Robert G. Gallager are prominent due to the capacity-approaching and superior error-correcting properties. There is no hard constraint on the code rate of LDPC codes. Consequently, it is ideal to incorporate LDPC codes with various code rate and codeword length in the adaptive modulation and coding (AMC) systems which change the encoder and the modulator adaptively to improve the system throughput. In conventional AMC systems, a dedicated control channel is assigned to coordinate the encoder/decoder changes. …

Exploiting Heterogeneity In Chip-Multiprocessor Design, Ying Zhang

LSU Doctoral Dissertations

In the past decade, semiconductor manufacturers are persistent in building faster and smaller transistors in order to boost the processor performance as projected by Moore’s Law. Recently, as we enter the deep submicron regime, continuing the same processor development pace becomes an increasingly difficult issue due to constraints on power, temperature, and the scalability of transistors. To overcome these challenges, researchers propose several innovations at both architecture and device levels that are able to partially solve the problems. These diversities in processor architecture and manufacturing materials provide solutions to continuing Moore’s Law by effectively exploiting the heterogeneity, however, they also …

Integrating Multiple Clusters For Compute-Intensive Applications, Zhifeng Yun

LSU Doctoral Dissertations

Multicluster grids provide one promising solution to satisfying the growing computational demands of compute-intensive applications. However, it is challenging to seamlessly integrate all participating clusters in different domains into a single virtual computational platform. In order to fully utilize the capabilities of multicluster grids, computer scientists need to deal with the issue of joining together participating autonomic systems practically and efficiently to execute grid-enabled applications. Driven by several compute-intensive applications, this theses develops a multicluster grid management toolkit called Pelecanus to bridge the gap between user's needs and the system's heterogeneity. Application scientists will be able to conduct very large-scale …

Quality Of Service Based Data-Aware Scheduling, Archit Kulshrestha

LSU Doctoral Dissertations

Distributed supercomputers have been widely used for solving complex computational problems and modeling complex phenomena such as black holes, the environment, supply-chain economics, etc. In this work we analyze the use of these distributed supercomputers for time sensitive data-driven applications. We present the scheduling challenges involved in running deadline sensitive applications on shared distributed supercomputers running large parallel jobs and introduce a ``data-aware'' scheduling paradigm that overcomes these challenges by making use of Quality of Service classes for running applications on shared resources. We evaluate the new data-aware scheduling paradigm using an event-driven hurricane simulation framework which attempts to run …

Power And Memory Optimization Techniques In Embedded Systems Design, Atef Khalil Allam

LSU Doctoral Dissertations

Embedded systems incur tight constraints on power consumption and memory (which impacts size) in addition to other constraints such as weight and cost. This dissertation addresses two key factors in embedded system design, namely minimization of power consumption and memory requirement. The first part of this dissertation considers the problem of optimizing power consumption (peak power as well as average power) in high-level synthesis (HLS). The second part deals with memory usage optimization mainly targeting a restricted class of computations expressed as loops accessing large data arrays that arises in scientific computing such as the coupled cluster and configuration interaction …