Computer and Systems Architecture Commons^™

Blockchain Accelerators And An Application On Covid-19 Contact Tracing, Tao Lu

LSU Doctoral Dissertations

Blockchain technology has been an emerging technology in recent years. Its nature of decentralization and anonymity enables many applications to be built in a trustless environment but can still be validated and agreed upon by all the participants. Ever since the invention of blockchain technology, its concepts and forms are explored extensively, and it has shown the potential to be used in many industries. However, the low efficiency and the poor performance have prevented the technology to be adopted in practical largescale usage. In this dissertation, we present three major works around blockchain technology. The first one is BPU, an …

Compilation Optimizations To Enhance Resilience Of Big Data Programs And Quantum Processors, Travis D. Lecompte

LSU Doctoral Dissertations

Modern computers can experience a variety of transient errors due to the surrounding environment, known as soft faults. Although the frequency of these faults is low enough to not be noticeable on personal computers, they become a considerable concern during large-scale distributed computations or systems in more vulnerable environments like satellites. These faults occur as a bit flip of some value in a register, operation, or memory during execution. They surface as either program crashes, hangs, or silent data corruption (SDC), each of which can waste time, money, and resources. Hardware methods, such as shielding or error correcting memory (ECM), …

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 …

Characterizing And Optimizing Asynchronous Event-Driven Architecture For Modern Cloud Systems, Shungeng Zhang

LSU Doctoral Dissertations

Achieving good performance and high efficiency simultaneously is an essential requirement for emerging modern cloud systems such as e-commerce due to their business impact. For example, Akamai reported that every 100ms delay in website load time could lead to a 6% drop in sales. Unfortunately, achieving good performance (e.g., low latency) for modern cloud systems at high resource utilization is significantly challenging. Despite continuing efforts by cloud professionals, however, they have consistently experienced performance degradation problems (e.g., the long-tail latency problem) due to the bursty workload in the cloud. To resolve the performance degradation problems, many previous research efforts have …

Understanding And Optimizing Flash-Based Key-Value Systems In Data Centers, Yichen Jia

LSU Doctoral Dissertations

Flash-based key-value systems are widely deployed in today’s data centers for providing high-speed data processing services. These systems deploy flash-friendly data structures, such as slab and Log Structured Merge(LSM) tree, on flash-based Solid State Drives(SSDs) and provide efficient solutions in caching and storage scenarios. With the rapid evolution of data centers, there appear plenty of challenges and opportunities for future optimizations.

In this dissertation, we focus on understanding and optimizing flash-based key-value systems from the perspective of workloads, software, and hardware as data centers evolve. We first propose an on-line compression scheme, called SlimCache, considering the unique characteristics of key-value …

On I/O Performance And Cost Efficiency Of Cloud Storage: A Client's Perspective, Binbing Hou

LSU Doctoral Dissertations

Cloud storage has gained increasing popularity in the past few years. In cloud storage, data are stored in the service provider’s data centers; users access data via the network and pay the fees based on the service usage. For such a new storage model, our prior wisdom and optimization schemes on conventional storage may not remain valid nor applicable to the emerging cloud storage.

In this dissertation, we focus on understanding and optimizing the I/O performance and cost efficiency of cloud storage from a client’s perspective. We first conduct a comprehensive study to gain insight into the I/O performance behaviors …

Secure And Efficient Bft Consensus For Blockchains, Mohammad Mussadiq Jalalzai

LSU Doctoral Dissertations

Blockchains are simple data structures, containing transactions organized into blocks, in which each block points to a previous block using its hash. Thus, by following the chain, we can follow the history of transactions. Blocks are added to the chain through a consensus mechanism. Byzantine Fault Tolerant (BFT) consensus protocols that were designed before blockchains were introduced are usually considered appropriate for use in small scale networks of size 10-20 replicas. Blockchains have changed this trend as the blockchain networks usually require a larger number of replicas and classic BFT protocols cannot provide acceptable performance in large networks. One of …

A Study Of Scalability And Cost-Effectiveness Of Large-Scale Scientific Applications Over Heterogeneous Computing Environment, Arghya K. Das

LSU Doctoral Dissertations

Recent advances in large-scale experimental facilities ushered in an era of data-driven science. These large-scale data increase the opportunity to answer many fundamental questions in basic science. However, these data pose new challenges to the scientific community in terms of their optimal processing and transfer. Consequently, scientists are in dire need of robust high performance computing (HPC) solutions that can scale with terabytes of data.

In this thesis, I address the challenges in three major aspects of scientific big data processing as follows: 1) Developing scalable software and algorithms for data- and compute-intensive scientific applications. 2) Proposing new cluster architectures …

A Study Of Very Short Intermittent Ddos Attacks On The Performance Of Web Services In Clouds, Huasong Shan

LSU Doctoral Dissertations

Distributed Denial-of-Service (DDoS) attacks for web applications such as e-commerce are increasing in size, scale, and frequency. The emerging elastic cloud computing cannot defend against ever-evolving new types of DDoS attacks, since they exploit various newly discovered network or system vulnerabilities even in the cloud platform, bypassing not only the state-of-the-art defense mechanisms but also the elasticity mechanisms of cloud computing.

In this dissertation, we focus on a new type of low-volume DDoS attack, Very Short Intermittent DDoS Attacks, which can hurt the performance of web applications deployed in the cloud via transiently saturating the critical bottleneck resource of the …

Conflict Reduction Techniques For Gpu Transactional Memory Systems, Sui Chen

LSU Doctoral Dissertations

The continued evolution of GPUs have enabled the use of irregular algorithms which involve fine-grained data sharing between threads, as well as transaction processing applications such as databases. Transactional Memory (TM) is derived from databases, which by itself is a programming construct that simplifies the programming of parallel workloads and combines the advantages of traditional approaches including fine-grained and coarse-grained locking. While hardware support for TM has started to enter mainstream commodity products, it is much farther from becoming reality on the GPU and is still being researched. In this dissertation, we study the challenges for supporting transactional workloads on …

A Study Of Application-Awareness In Software-Defined Data Center Networks, Chui-Hui Chiu

LSU Doctoral Dissertations

A data center (DC) has been a fundamental infrastructure for academia and industry for many years. Applications in DC have diverse requirements on communication. There are huge demands on data center network (DCN) control frameworks (CFs) for coordinating communication traffic. Simultaneously satisfying all demands is difficult and inefficient using existing traditional network devices and protocols. Recently, the agile software-defined Networking (SDN) is introduced to DCN for speeding up the development of the DCNCF. Application-awareness preserves the application semantics including the collective goals of communications. Previous works have illustrated that application-aware DCNCFs can much more efficiently allocate network resources by explicitly …