Mao3x4-225x300

Dinh-Mao Bui


Department: Computer Science
Position: Assistant Professor
Degree: PhD in Computer Science and Engineering, Kyung Hee University, Republic of Korea
Office: 7e.442
Email: mao.bui@nu.edu.kz
Phone: + 7 (7172) 694664
Website: https://cs-sst.github.io/faculty/bui

CURRENT ACTIVITY

Currently, I have finished the Ph.D. in Computer Engineering from the Ubiquitous Computing Lab, Department of Computer Science and Engineering, Kyung Hee University, Korea. My current research interests are machine learning (supervised learning, especially the prediction technique) and optimization for the distributed system, big data, and networking. Some of my works, which include the adaptive replication management in HDFS, the multiuser detection in DS-CDMA, and the energy efficiency in cloud computing, are based on my improved Gaussian process prediction method. In the past, my previous research works are related to modeling and optimizing the operation of Infrastructure-as-a-Services (IaaS) cloud computing in the datacenter. For the working experience, I have three years full-time working as the system developer and four years working as the Ph.D. researcher on many projects listed as follows


EMPLOYMENT HISTORY

Jan. 2014 – Aug. 2018

ICNS Lab, Kyung Hee University

Korea

   Postdoctoral

  • Advising PhD and Master students.
  • Teaching assistance for professor.
  • Continue enhancing the energy efficiency approach for cloud computing.
  • Preparing government project proposal on the area of distributed system, 5G communication system.

 

Mar. 2014 – Jan. 2018

Ubiquitous Computing Lab, Kyung Hee University

Korea

   Ph.D. Researcher

  • Develop performance oriented prediction technique based on improved Gaussian process regression.
  • Enhance the performance of Hadoop system based on adaptive replication management.
  • Improve energy efficiency in both low-level (CPU multicores) and high-level (cloud computing).
  • Propose new multiuser detection in DS-CDMA system based on regression technique.
  • Enhance fault tolerance in cloud computing based on proactive approach.
  • Develop scalable data acquisition and synchronization for Mining Mind project.
  • Writing project proposals.

 

Nov. 2010 – 2013

Vietnam Data Communication Company (VDC)

Vietnam

   System Engineer in R&D Section of Online Application Center

  • Develop and propose project plans related to IaaS Cloud Computing.
  • Establish the cloud infrastructure for testing and developing based on Open Nebula.
  • Implement the component for computational resource and IP allocation.
  • Propose service leasing based on virtual machine provision.
  • Optimize resource utilization with regard to the Service Level Agreement.
  • Implement proactive fault detection and disaster recovery for infrastructure.
  • Develop Ruby-based Restful API services for resource management.

 


EDUCATION

Jan. 2018

Kyung Hee University, Korea

Advisor: Professor Sungyoung Lee, Kyung Hee University, Korea.

Thesis: Performance-oriented prediction based on improved Gaussian process regression.

In this thesis, I thoroughly analyze the nature of Bayesian learning and Gaussian process regression (GPR). The analysis is not only in theory but also broadened to some potential applications in real world. Subsequently, the proposal of enhancement is introduced to each phase of GPR, which are hyper-parameters learning phase and training phase. In detail, my contribution and uniqueness focus on:-      Propose a complexity reduction to hyper-parameters learning phase of GPR. This method is a cooperation of fast Fourier transform, convergence law of log determinant and stochastic gradient descent. The target of this cooperation is the possibility of indirectly optimizing and approximating the hyper-parameters. By applying this method, I can significantly improve the speed of finding the hyper-parameters with a slight degradation in accuracy.-      Introduce the ‘divide and conquer’ coupled with parallel processing to the training phase to improve the processing speed, rather than relying only on iterative gradient methods like other related research.

 

Ph.D. in CS


PROJECTS

2010 – 2013

_IaaS cloud computing resource management system solution, supported by the VDC2, VNPT.

2014 – 2016

_EAOS – Development of general-purpose OS and virtualization technology to reduce 30 % of energy for high-density servers based on low-power processors, funded by the IT R&D program of Korean Ministry of Science.

_PESS – Profiling-based energy saving system for energy-efficient cloud platform”, funded by the National Research Foundation of Korea (NRF), Korean Ministry of Science.

2016-2018

_Mining Mind – Development of mining minds core technology exploiting personal big data for health and wellness support, funded by the Korean Ministry of Trade, Industry and Energy.

 


SELECTED PUBLICATIONS

–      Dinh-Mao Bui, Shujaat Hussain, Eui-Nam Huh, and Sungyoung Lee. “Adaptive replication management in HDFS based on supervised learning.” IEEE Transactions on Knowledge and Data Engineering (SCI, IF: 3.438), 2016.

https://doi.org/10.1109/TKDE.2016.2523510

 

–      Dinh-Mao Bui and Sungyoung Lee. “Fast Gaussian Process Regression for Multiuser Detection in DS-CDMA.” IEEE Communications Letters (SCI, IF: 1.988), 2017.

https://doi.org/10.1109/LCOMM.2016.2620430

 

–      Dinh-Mao Bui, YongIk Yoon, Eui-Nam Huh, SungIk Jun, and Sungyoung Lee. “Energy efficiency for cloud computing system based on predictive optimization.” Journal of Parallel and Distributed Computing (SCI, IF:1.93), 2017.

https://doi.org/10.1016/j.jpdc.2016.11.011

 

–      Dinh-Mao Bui, Thien Huynh-The and Sungyoung Lee. “Early fault detection in IaaS cloud computing based on fuzzy logic and prediction technique.” The Journal of Supercomputing (SCI, IF: 1.326), 2017.

https://doi.org/10.1007/s11227-017-2053-3

 

–      Dinh-Mao Bui, Huu-Quoc Nguyen, YongIk Yoon, SungIk Jun, Muhammad Bilal Amin, and Sungyoung Lee. “Gaussian process for predicting CPU utilization and its application to energy efficiency.” Applied Intelligence (SCI, IF: 1.215), 2015.

https://doi.org/10.1007/s10489-015-0688-4