We research technologies for managing and analyzing truly large scale datasets. When processing large volumes of data, parallel and distributed computations, efficient storage systems, and database technologies are essential. We push high performance data processing technologies to the limit by providing innovative and practical research in the intersection of three major computer science areas - distributed and parallel computing, database systems, and storage systems.

Big Data Processing Framework

We develop Velox , a novel high-performance decentralized big data processing framework. More details about this project can be found at VeloxMR and VeloxDFS .

DBMS for NVRAM

We develop Eternity , a novel database management system for byte-addressable non-volatile memory.

Android I/O Stack

SQLite is a server-less database engine embedded in Android, which is often blamed for its poor interaction with EXT4 file system. In this project, we improve the I/O performance of SQLite.

High Performance Machine Learning

We explore the opportunity to leverage high performance computing technologies to accelerate distributed machine learning.

Scientific Dataset Indexing on GPU

In this project, we develop massively parallel tree traversal algorithms for GPUs. Project page

Recent Publications

IEEE Cloud 2017, "Coalescing HDFS Blocks to Avoid Recurring YARN Container Overhead"

ASPLOS 2017, "Failure-Atomic Slotted Paging"

USNIEX FAST 2017, "WORT: Write-Optimal Radix Tree for Persistent Storage Systems"

Full List of Publications
Upcoming Conferences

People

in UNIST DICL

Beomseok Nam

Associate Professor
WWW

Jinwoong Kim

MS/Ph.D program, 2011 ~ Present
WWW

Moohyeon Nam

MS/Ph.D. program, 2013 ~ Present
WWW

Wook-Hee Kim

MS/Ph.D. program, 2013 ~ Present
WWW

Wonbae Kim

Ph.D. program, 2017 ~ Present
WWW

Ki-Beom Jin

M.S. program, 2015 ~ Present
WWW

Jihye Seo

M.S. program, 2016 ~ Present
WWW

Deukyeon Hwang

M.S. program, 2017 ~ Present
WWW