• Proceedings of the Korean Society of Computer Information Conference
• /
• 2017.07a
• /
• pp.3-4
• /
• 2017

DRAM, HDD의 성장세가 둔화되면서 NVM이 그 대안으로 떠오르고 있다. Xpoint, Z-SSD등 여러 고성능 NVM이 발표되고 있으며 DRAM의 성능을 목표로 하고 있다. 본 논문에서는 DRAM 수준의 고성능 NVM이 상용화 될 때 어느 정도의 성능과 가격지표를 보일 것이며 이에 대한 가격과 저장공간의 트레이드오프는 어느 정도일지 알아보기 위해 실험을 진행하였다. RAM disk로 NVM환경을 시뮬레이션 하였고 SSD, HDD 간 I/O 성능을 측정하였다. NVM과 SSD는 70배에서 200배, NVM과 디스크는 98배에서 3000배 이상 차이가 있음을 확인했다. 이 결과를 바탕으로 RAM disk와 SSD 가격지표를 조사하여 IOPS/$를 도출해 보았고 NVM이 7배의 IOPS/$를 가지지만 SSD가 18배의 GB/$를 가지는 것을 확인할 수 있었다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2001.10a
• /
• pp.283-286
• /
• 2001

전통의 자기 디스크 형태 저장 장치로의 접근 지연 시간을 줄이고자, 최근에 서로 네트워크로 연결된 컴퓨터간의 메모리 자원들을 파일을 저장하는 장치로써 활용해 볼 수 있는 가상 장치(Network RamDisk)가 개발되었다. 이 시스템은 메모리 공간을 제공해 주는 서버의 노드들이 패리티 그룹(parity group)을 이루는 일종의 software RAID 시스템으로 한 서버의 노드 crash 에 대해 대응하는 것을 고려하지 않아 제 이의 서버가 노드 crash, 프로세스 crash 혹은 데이터 손실을 발생시키면 완전히 데이터를 잃어버리는, 신뢰성이 낮은 문제점이 있다. 본 논문은 클라이언트가 서버측의 노드 crash를 적절히 감지하고 새로운 서버로 fail over 할 수 있는 여러 가지 기법들을 설계하고 구현하였다. 또한 서버 풀(server pool)이라는 가상의 공간을 따로 관리하여 서버들에 대한 메모리 자원 요구 부하를 균등히 분산 시키는 효과도 얻도록 하였다.

• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.3
• /
• pp.171-180
• /
• 2009

Slowly, but surely, we are seeing the emergence of a variety of embedded systems that are employing Storage Class RAM (SCRAM) such as FeRAM, MRAM and PRAM, SCRAM not only has DRAM-characteristic, that is, random byte-unit access capability, but also Disk-characteristic, that is, non-volatility. In this paper, we propose a new software architecture that allows SCRAM to be used both for main memory and for secondary storage simultaneously- The proposed software architecture has two core modules, one is a SCRAM driver and the other is a SCRAM manager. The SCRAM driver takes care of SCRAM directly and exports low level interfaces required for upper layer software modules including traditional file systems, buddy systems and our SCRAM manager. The SCRAM manager treats file objects and memory objects as a single object and deals with them in a unified way so that they can be interchanged without copy overheads. Experiments conducted on real embedded board with FeRAM have shown that the SCRAM driver indeed supports both the traditional F AT file system and buddy system seamlessly. The results also have revealed that the SCRAM manager makes effective use of both characteristics of SCRAM and performs an order of magnitude better than the traditional file system and buddy system.

• Journal of Astronomy and Space Sciences
• /
• v.13 no.2
• /
• pp.149-162
• /
• 1996

The results of space radiation experiments carried out on board the first two Korean technology demonstration microsatellites are presented in this paper. The first satellite, KITSAT-1, launched in August 1992, carries a radiation monitoring payload called cosmic ray experiment(CRE) for characterizing the low-earth orbit(LEO) radiation environment. The CRE consists of two sub-systems: the cosmic particle experiment (CPE) and the total dose experiment(TDE). In addition, single event upset(SEU)rates of the program memory and the RAM disk are also monitored. The second satellite, KITSAT-2, launched in September 1993, carries a newly developed 32-bit on-board computer(OBC), KASCOM(KAIST satellite computer in addition to OBC186. SEUs ocurred in the KASCOM, as well as in the program memory and RAM disk memory, have been monitored since the beginning of the satellite operation. These two satellites, which are very similar in structures but different in orbits, provide a unique opportunity to study the effects of the radiation environment characterized by the orbit.

• The KIPS Transactions:PartD
• /
• v.12D no.4 s.100
• /
• pp.543-552
• /
• 2005

As random access memory chip gets cheaper, it becomes affordable to realize main memory-based database systems. The disk-based spatial indexing techniques, however, cannot direct apply to main memory databases, because the main purpose of disk-based techniques is to reduce the number of disk accesses. In main memory-based indexing techniques, the node access time is much faster than that in disk-based indexing techniques, because all index nodes reside in a main memory. Unlike disk-based index techniques, main memory-based spatial indexing techniques must reduce key comparing time as well as node access time. In this paper, we propose an efficient spatial index structure for main memory-based databases, called Tmr-tree. Tmr-tree integrates the characteristics of R-tree and T-tree. Therefore, Nodes of Tmr-tree consist of several entries for data objects, main memory pointers to left and right child, and three additional fields. First is a MBR of a self node, which tightly encloses all data MBRs (Minimum Bounding Rectangles) in a current node, and second and third are MBRs of left and right sub-tree, respectively. Because Tmr-tree needs not to visit all leaf nodes, in terms of search time, proposed Tmr-tree outperforms R-tree in our experiments. As node size is increased, search time is drastically decreased followed by a gradual increase. However, in terms of insertion time, the performance of Tmr-tree was slightly lower than R-tree.

• Journal of KIISE:Computing Practices and Letters
• /
• v.14 no.5
• /
• pp.537-541
• /
• 2008

Log-Structured File System(LFS) is a disk based file system that is optimized for improving the write performance. LFS gathers dirty data in memory as long as possible, and flushes all dirty data sequentially at once. In a real system, however, maintaining dirty data in memory should be flushed into a disk to meet file system consistency issues even if more memory is still available. This synchronizations increase the cleaner overhead of LFS and make LFS to write down more metadata into a disk. In this paper, by adapting Non-volatile RAM(NV-RAM) we modifies LFS and virtual memory subsystem to guarantee that LFS could gather enough dirty data in the memory and reduce small disk writes. By doing so, we improves the performance of LFS by around 2.5 times than the original LFS.

• Proceedings of the Safety Management and Science Conference
• /
• 2002.05a
• /
• pp.151-159
• /
• 2002

Real-Time Linux를 이용하여 실시간 운영체제가 요구하는 특성과 요구조건을 분석하고 이러한 요구조건에 부합하도록 리눅스를 하드웨어에 이식하고 하드웨어에 이식하기 위한 방법을 제시하였다. 다른 상용 실시간 운영체제 (RTOS)와는 달리 리눅스는 특정 하드웨어를 지원하기 위한 별도의 개발환경을 제공하지 않는다. 이에 개발환경을 구축하고 부트로더를 개발하기 위해 목표 시스템에 부합하도록 리눅스 커널을 이식하였다. 또한 응용 개발의 유연성을 제공하기 위하여 램디스크를 이용한 파일 시스템을 지원하도록 하였으며, GPIO(general purpose I/O)를 통한 디바이스 드라아버를 제작하는 등의 실험을 통해 시스템의 안정성을 검증하였다 실험에서는 StrongArm SAl110 마이크로프로세서를 이용하였으며 이 실험을 통해 실시간 운영체제로서의 리눅스의 활용 가능성을 확인하였다.

• Proceedings of the Korean Society of Computer Information Conference
• /
• 2014.07a
• /
• pp.333-335
• /
• 2014

페이스북, 트위터와 같은 클라우드 및 웹 서비스 제공회사와 클라우드 및 웹 서비스 제공자는 수많은 사용자들에 의해 발생되는 무수한 데이터를 빠르게 처리하기 위해 하드 디스크보다는 램에 저장 할 필요가 있다. 그러한 좋은 도구로서 분산메모리 객체 캐싱 소프트웨어인 멤캐시드가 있다. 멤캐시드의 성능은 저장공간의 크기에 따라 많은 차이를 보이는데, 하드웨어의 비용, 전력소비와 온도조절 등 공간의 제약을 감안했을 때, 무작정 개별 서버에 많은 RAM을 장착하거나, 서버 배열을 확장하는 것은 효율적인 방법이 아니다. 따라서 많은 양의 데이터가 메모리에 저장이 가능하도록 RAM과 SSD를 같이 확장한 SSD 기반 하이브리드 메모리를 제안한다. 하이브리드 메모리는 객체 캐시로 동작하고 페이지 단위로 할당하는 것보다 객체 단위로 자원할당을 함으로서 SSD에서 빠른 무작위 읽기를 할 수 있게 해 객체의 접근속도를 향상시켰다.

• The KIPS Transactions:PartD
• /
• v.11D no.1
• /
• pp.11-22
• /
• 2004

As random access memory chip gets cheaper, it becomes affordable to realize main memory-based database systems. Consequently, reducing cache misses emerges as the most important issue in current main memory databases, in which CPU speeds have been increasing at 60% per year, compared to the memory speeds at 10% per you. In this paper, we design and implement a main-memory database system for real-time mobile GIS. Our system is composed of 5 modules： the interface manager provides the interface for PDA users； the memory data manager controls spatial and non-spatial data in main-memory using virtual memory techniques； the query manager processes spatial and non-spatial query : the index manager manages the MR-tree index for spatial data and the T-tree index for non-spatial index : the GIS server interface provides the interface with disk-based GIS. The MR-tree proposed propagates node splits upward only if one of the internal nodes on the insertion path has empty space. Thus, the internal nodes of the MR-tree are almost 100% full. Our experimental study shows that the two-dimensional MR-tree performs search up to 2.4 times faster than the ordinary R-tree. To use virtual memory techniques, the memory data manager uses page tables for spatial data, non- spatial data, T-tree and MR-tree. And, it uses indirect addressing techniques for fast reloading from disk.

• Journal of the Korean Institute of Intelligent Systems
• /
• v.14 no.4
• /
• pp.424-430
• /
• 2004

This paper explains the construction of the filesystems which could be utilized in embedded systems as an implementation of ubiquitous computing. It includes the formal architecture of filesystem hierarchy for the DOC (Disk-On-Chip) filesystem and the flash filesystem based on the MTD (Memory Technology Devices). For DOC, the root filesystem and the user filesystem are constructed by the TrueFFS supported by the M-Systems. For MTD filesystem, the root filesystem is implemented in the fast RAM disk, and the user filesystem is implemented in the JFFS2 that supports large capacity. In order to support the GUI filesystem, the porting process of Qt/E is also included in this paper.