-
Myungjun Son
He is a Research Assistant in the Computer Science and Engineering Department at the Pennsylvania State University. His research interests involve cloud computing, serverless computing, bigdata platforms and large-scale distributed computing resource manangement. CV dblp
Recent News
- 10/2019 - MPEC: Distributed Matrix Multiplication Performance Modeling on a Scale-out Cloud Environment for Data Mining Jobs has been accepted on Transactions on Cloud Computing!
- 09/2019 - Started RA in The Penn State!
Education
The Pennsylvania State University, University Park- Ph.D.'s in Computer Science : Sep 2019 - Present
Kookmin University, Seoul- Master's in Computer Science : March 2017 - August 2019
- Bachelor's in Computer Science : March 2010 - February 2017
Research
Currently, He is a member of Microsystems Design Lab group woring under the direction of Dr. Mahmut T. KandemirIn the past, He worked under the direction of Dr. Kyungyong Lee in the BigData lab.
- ABC2: Autonomic BigData Cloud Computing research project, 2017-2019
- Build autonomic cloud computing service that handles complex infrastructure configurations
- Provide optimal kernel layer to utilize in a wide range of big-data processing applications
- Implemented system that predicts latency of matrix multiplication of arbitrary shapes and sizes using various cloud computing resource configurations
- DeepSpotCloud , 2017-2018
- Use cloud computing service to build cost-efficient and scalable big-data analytics platform
- Implemented system of DeepSpotCloud utilizing EC2 GPU spot instances to build cost-efficient, deep learning framework
Publications
- IEEE ToCC 2019 - Jeongchul Kim, Myungjun Son, Kyungyong Lee, “Modeling Distributed Matrix Multiplication Performance on a Scale-out Cloud Environment for Data Mining Jobs,” IEEE Transactions on Cloud Computing [PDF]
- IEEE CLOUD 2018 - Myungjun Son and Kyungyong Lee, “Distributed Matrix Multiplication Performance Estimator for Machine Learning Jobs in Cloud Computing,” 2018 IEEE 11th International Conference on Cloud Computing (CLOUD) [PDF]
- IEEE CLOUD 2017 - Kyungyong Lee and Myungjun Son, “DeepSpotCloud: Leveraging Cross-Region GPU Spot Instances for Deep Learning,” 2017 IEEE 10th International Conference on Cloud Computing (CLOUD) [PDF]
Talks
- “Distributed Matrix Multiplication Performance Estimator for Machine Learning Jobs in Cloud Computing,” 2018 IEEE 11th International Conference on Cloud Computing (CLOUD)
Technical skills
- Programming Languages: Python, Scala
- Platforms: Linux, AWS, Docker, Spark, Hadoop, Kubernetes
Myungjun Son
Research Assistant.
call +1 814-826-5993 email mjson dot psu.edu© 2019