(Photo Taken in November 2014)



Youngjae Kim (PhD)

Assistant Professor
Department of Software
College of Information Technology
Ajou University

Email: youkim a.t. ajou.ac.kr
Office: No. 704 Paldal Hall
Phone: +82-31-219-3811
Address: 206 Worldcup-Ro, Yeongtong-Gu, Suwon-Si, Gyeonggi-Do, 443-749, Korea 
NEWS!
  • [Publication] A paper titled "Optimizing End-to-End Big Data Transfers over Terabits Network Infrastructure" is accepted to appear in IEEE Transactions on Parallel and Distributed Systems (TPDS) (Feb. 9, 2016)

  • [Funding] Dr. Kim is awarded a research grant for Young Researchers by NRF about developing a multi-tiered big data storage system (2015년도 신진연구지원사업) (Dec. 3, 2015) 

Research Interest

    System Software Design and Development, Distributed Data Storage, Data Analytics System, Non-volatile Memory 
  • I'm broadly interested in the intersection of cutting-edge technologies in hardware, system software, and application spanning a diverse spectrum of environments from cloud, enterprise computing to embedded domain. 
    Recent Research Topics: 
  • Integrating search and discovery services into file systems  
  • Image storage system for scalable training of deep neural networks  
  • UniStore: A unified big data store aggregating geo-dispersed big data storage systems 
  • Fault tolerant data transfers over terabits network  
  • End-to-end analysis-aware data placement in virtual data facility 
 
 
<ImageNet Data Example> Building an image storage system for scalable training of deep neural networks  


<Big Data+Storage+Discovery>

Integrating scientific search and discovery services in large-scale distributed file and storage systems  

 


<Big Data Coupling in HPC and Cloud>

Coupling geo-dispersed 

datasets and big data transfer over terabits network using CCI 


Employment 
  • 2015-Present: Assistant Professor, Department of Software, Ajou University, Korea 

  • 2009-2015: Research Staff Member, Oak Ridge National Laboratory for US Department of Energy, Oak Ridge, TN, USA
    • 2014-2015: Team leader (NVM File Systems), National Center for Computational Sciences
      • Explored non-volatile memory devices from several perspectives including memory extension, burst buffers, out-of-core data analytics, and fault tolerance for extreme compute and data systems
    • 2012-2015: Level III Research Staff Member
    • 2009-2011: Level II Research Staff Member

  • 2011-2014: Adjunct Professor, School of Electrical and Computer Engineering, Georgia Institute of Technology

  • 2003-3004: Researcher, Embedded OS Team, ETRI, Daejeon, Korea

Education 

Selected Papers authored or coauthored (Full List of Publications)
  • Optimizing End-to-End Big Data Transfers over Terabits Network Infrastructure, (to appear) IEEE TPDS'16 
  • AnalyzeThis: An Analysis Workflow-Aware Storage System, SC’15 
  • LADS: Optimizing Data Transfers using Layout-Aware Data SchedulingUSENIX FAST’15 (pdf) 
  • Synchronous I/O Scheduling of Independent Write Caches for an Array of SSDsIEEE CAL’15 (pdf) 
  • Best Practices and Lessons Learned from Deploying and Operating Large-Scale Data-Centric Parallel File systems, SC’14 (Best paper finalist) (pdf) 
  • Harmonia: Coordinating Garbage Collection for Arrays of Solid-state DrivesIEEE TC’14 (pdf) 
  • Active Flash: Towards Energy-Efficient, In-Situ Data Analytics on Extreme-Scale MachinesUSENIX FAST’13 (pdf) 
  • Preemptible I/O Scheduling of Garbage Collection for Solid-state Drives, IEEE TCAD13 (pdf)
  • D-Factor: A Quantitative Model of Application Slow-Down in Shared Service Systems with Multiple ResourcesSIGMETRICS’12 (pdf) (slides)
  • NVMalloc: Exposing an Aggregate SSD Store as a Memory Partition in Extreme-Scale MachinesIPDPS’12 (pdf) 
  • Workload Characterization and Performance Implications of Large-Scale Blog Servers, ACM WEB12 (pdf) 
  • Migration, Assignment, and Scheduling of Jobs in Virtualized Environment, USENIX HotCloud11 (pdf) 
  • HybridStore: A Cost-Efficient, High-Performance Storage System Combining SSDs and HDDsMASCOTS’11 (pdf) (slides)
  • Provisioning a Multi-Tiered Data Staging Area for Extreme-Scale MachinesICDCS’11 (pdf) 
  • A Semi-Preemptive Garbage Collector for Solid State Drives, ISPASS’11 (Best paper finalist) (pdf) (slides)
  • Functional Partitioning to Optimize End-to-End Performance on Many-Core ArchitecturesSC’10 (pdf) 
  • DFTL: A Flash Translation Layer Employing Demand-based Selective Caching of Page-level Address Mappings, ASPLOS’09 (Google scholar citations: > 500 in Jan. 31, 2016) (pdf) 
  • FlashSim: A Simulator for NAND Flash-based Solid-State Drives, SIMUL’09 (Google scholar citations: > 150 in Jan. 31 2016) (pdf) 
  • Managing Thermal Emergencies in Disk-Based Storage Systems, ASME JEP08 (pdf) 
  • Modeling and Managing Thermal Profiles of Rack-mounted Servers with ThermoStat, HPCA’07 (Best paper finalist) (pdf) 
  • Using STEAM for Thermal Simulation of Storage Systems, IEEE MICRO06 (pdf) 
  • Understanding the Performance-Temperature Interactions in Disk I/O of Server WorkloadsHPCA’06 (pdf)