Enabling Distributed Applications Optimization in Cloud Environment

Authors

Pinchao LiuFollow

Document Type

Dissertation

Degree

Doctor of Philosophy (PhD)

Major/Program

<--Please Select Department-->

First Advisor's Name

Liting Hu

First Advisor's Committee Title

committee chair

Second Advisor's Name

S. S. Iyengar

Second Advisor's Committee Title

committee member

Third Advisor's Name

Jason Liu

Third Advisor's Committee Title

committee member

Fourth Advisor's Name

Deng Pan

Fourth Advisor's Committee Title

committee member

Fifth Advisor's Name

Gang Quan

Fifth Advisor's Committee Title

committee member

Keywords

Cloud computing, Stream processing, Distributed systems

Date of Defense

3-25-2021

Abstract

The past few years have seen dramatic growth in the popularity of public clouds, such as Infrastructure-as-a-Service (IaaS), Platform-as-a-Service (PaaS), and Container-as-a-Service (CaaS). In both commercial and scientific fields, quick environment setup and application deployment become a mandatory requirement. As a result, more and more organizations choose cloud environments instead of setting up the environment by themselves from scratch. The cloud computing resources such as server engines, orchestration, and the underlying server resources are served to the users as a service from a cloud provider. Most of the applications that run in public clouds are the distributed applications, also called multi-tier applications, which require a set of servers, a service ensemble, that cooperate and communicate to jointly provide a certain service or accomplish a task. Moreover, a few research efforts are conducting in providing an overall solution for distributed applications optimization in the public cloud.

In this dissertation, we present three systems that enable distributed applications optimization: (1) the first part introduces DocMan, a toolset for detecting containerized application’s dependencies in CaaS clouds, (2) the second part introduces a system to deal with hot/cold blocks in distributed applications, (3) the third part introduces a system named FP4S, a novel fragment-based parallel state recovery mechanism that can handle many simultaneous failures for a large number of concurrently running stream applications.

Identifier

FIDC009678

ORCID

0000-0001-7506-3506

Previously Published In

• Pinchao Liu, Liting Hu, Hailu Xu, Zhiyuan Shi, Jason Liu, Qingyang Wang, JaiDayal, and Yuzhe Tang, ”A Toolset for Detecting Containerized Application’s De-pendencies in CaaS Clouds”,2018 IEEE International Conference on Cloud Com-puting (IEEE CLOUD), June 2018.
• Pinchao Liu, Adnan Maruf, Farzana Beente Yusuf, Labiba Jahan, Hailu Xu, BoyuanGuan, Liting Hu, and Sitharama S. Iyengar, Towards Adaptive Replication forHot/Cold Blocks in HDFS using MemCached”,In Proceedings of 2019 Interna-tional Conference on Data Intelligence and Security (ICDIS 2019), June 2019.
• Pinchao Liu, Hailu Xu, Dilma Da Silva, QingyangWang, Sarker Tanzir Ahmed, andLiting Hu. ”FP4S: Fragment-based Parallel State Recovery for Stateful Stream Ap-plications”,34th IEEE International Parallel & Distributed Processing Symposium(IPDPS 2020).

Recommended Citation

Liu, Pinchao, "Enabling Distributed Applications Optimization in Cloud Environment" (2021). FIU Electronic Theses and Dissertations. 4653.
https://digitalcommons.fiu.edu/etd/4653