NECSST(Next-generation Embedded/Computer System Software Technology) Lab

차세대 임베디드 컴퓨터 시스템 소프트웨어 연구

관련기사 바로가기

The NECSST(Next-generation Embedded/Computer System Software Technology) Lab has been conducting research on system software issues with a focus on flash memory and non-volatile memory (aka NVRAM, NVM, Storage Class Memory (SCM), New Memory, etc.). We are currently being supported by various government ministries and agencies as well as by industry such as Samsung Electronics, SKhynix, and Intel. Our current focus of research is as follows:
•Systems Support for big data and deep learning
•Efficient file system design and implementation for NVM
•Memory management issues in tiered memory and disaggregated environments
•QoS issues for flash memory and SSDs
•Operating system redesign for non-volatile memory deployed systems

초대량의 데이터를 다루는 빅데이터와 딥러닝 응용에서 컴퓨팅 시스템의 효율성은 더욱 중요해지고 있다.
본 연구실은 플래시 메모리와 비휘발성메모리와 같은 차세대메모리를 통해 시스템의 효율을 극대화하는 연구를 수행하고 있다.

Major research field

Operating system, flash memory, SSD, persistent memory, Systems for big data and deep learning

Desired field of research

Operating system, flash memory, SSD, persistent memory, Systems for big data and deep learning

Research Keywords and Topics

system software, operating system, file system, storage

Research Publications
MORE

• USENIX Annual Technical Conference (ATC ’21) / First Responder: Persistent Memory Simultaneously as High Performance Buffer Cache and Storage
• USENIX Annual Technical Conference (ATC ’20) / HetPipe: Enabling Large DNN Training on (Whimpy) Heterogeneous GPU Clusters through Integration of Pipelined Model Parallelism and Data Parallelism
• IEEE International Conference on Data Engineering (ICDE 2020) / Doubleheader Logging: Eliminating Journal Write Overhead for Mobile DBMS
• USENIX Conference on File and Storage Technologies (FAST 2020) / Write-Optimized Dynamic Hashing for Persistent Memory
• USENIX Conference on File and Storage Technologies (FAST 2020) / SLM-DB : Single - Level Key-Value Store with Persistent Memory
• USENIX Annual Technical Conference (ATC ’19) / Alleviating Garbage Collection Interference Through Spatial Separation in All Flash Arrays

Patents

• Apparatus and method of managing multi solid state disk system / 노삼혁, 김병석, 김재호, 임광현 / 2019-12
• 비휘발성 메모리를 위한 단일 레벨 키-벨류 스토어 / 노삼혁, 최영리, Olzhas Kaiyrakhmet, 이송이 / 2019-11
• 뉴메모리 기반 메모리 관리 방법 및 이를 수행하는 장치들 (METHOD OF MANAGING MEMORY BASED ON NEW MEMORY, AND APPARATUSES PERFORMING THE SAME) / 노삼혁, 송현호 / 2019-06
• 전자 장치 및 제어 방법 (Electronic device and controlling method thereof) / 노삼혁, 박재형, 김정현, 문영제, 송현섭 / 2019-04

국가과학기술표준분류

  • EE. 정보/통신
  • EE02. 소프트웨어
  • EE0202. S/W 솔루션

국가기술지도분류

  • 정보-지식-지능화 사회 구현
  • 010700. 고성능정보처리/저장 장치 기술

녹색기술분류

  • 녹색기술관련 과제 아님
  • 녹색기술관련 과제 아님
  • 999. 녹색기술 관련과제 아님

6T분류

  • IT 분야
  • 정보처리 시스템 및 S/W
  • 010311. 멀티미디어 단말기 및 운영체계기술