• 한국정보과학회논문지:시스템및이론
• /
• 제37권3호
• /
• pp.161-171
• /
• 2010

임베디드 시스템이 모놀리식(monolithic) 커널을 사용하면서, NAND 플래시 메모리는 가상 메모리 시스템의 스왑(swap) 공간을 위해 사용되고 있다. 플래시 메모리는 저전력 소비, 충격 내구성, 비 휘발성의 장점을 가지지만, '쓰기 전 삭제'의 특징 때문에 가비지 컬렉션(GC) 작업이 필요하다. GC 기법의 효율성은 플래시 메모리 성능에 큰 영향을 미친다. 본 논문에서는 플래시 메모리를 기반으로 하는 가상 메모리 시스템에서 메인 메모리와 플래시 메모리 사이에 중복된 데이터를 활용한 새로운 GC 기법을 제안한다. 제안된 기법은 GC 부하를 최소화하기 위해 데이터의 지역성을 고려한다. 실험 결과는 제안된 GC 기법이 이전의 기법과 비교하여 평균적으로 37%의 성능을 향상시킴을 보여준다.

• 대한임베디드공학회논문지
• /
• 제2권3호
• /
• pp.184-193
• /
• 2007

Flash memory adoption in the mobile devices is increasing or vanous multimedia services such as audio, videos, and games. Although the traditional research issues such as out-place update, garbage collection, and wear-leveling are important, the performance, memory usage, and fast mount issues of flash memory file system are becoming much more important than ever because flash memory capacity is rapidly increasing. In this paper, we address the problems of the existing log-based flash memory file systems analytically and propose an efficient log-based file system, which produces higher performance, less memory usage and mount time than the existing log-based file systems. Our ideas are applied to a well-known log-based flash memory file system (YAFFS2) and the performance tests are conducted by comparing our prototype with YAFFS2. The experimental results show that our prototype achieves higher performance, less system memory usage, and faster mounting than YAFFS2, which is better than JFFS2.

• ETRI Journal
• /
• 제33권2호
• /
• pp.230-239
• /
• 2011

Dynamic memory allocators for real-time embedded systems need to fulfill three fundamental requirements: bounded worst-case execution time, fast average execution time, and minimal fragmentation. Since embedded systems generally run continuously during their whole lifetime, fragmentation is one of the most important factors in designing the memory allocator. This paper focuses on minimizing fragmentation while other requirements are still satisfied. To minimize fragmentation, a part of a memory region is segregated by the proposed budgeting method that exploits the memory profile of the given application. The budgeting method can be applied for any existing memory allocators. Experimental results show that the memory efficiency of allocators can be improved by up to 18.85% by using the budgeting method. Its worst-case execution time is analyzed to be bounded.

• International Journal of Internet, Broadcasting and Communication
• /
• 제14권1호
• /
• pp.188-193
• /
• 2022

The energy efficiency of memory systems is an important task in designing future computer systems as memory capacity continues to increase to accommodate the growing big data. In this article, we present an energy-efficient last-level cache management policy for future mobile systems. The proposed policy makes use of low-power PCM (phase-change memory) as the main memory medium, and reduces the amount of data written to PCM, thereby saving memory energy consumptions. To do so, the policy keeps track of the modified cache lines within each cache block, and replaces the last-level cache block that incurs the smallest PCM writing upon cache replacement requests. Also, the policy considers the access bit of cache blocks along with the cache line modifications in order not to degrade the cache hit ratio. Simulation experiments using SPEC benchmarks show that the proposed policy reduces the power consumption of PCM memory by 22.7% on average without degrading performances.

• 대한임베디드공학회논문지
• /
• 제13권5호
• /
• pp.235-243
• /
• 2018

NAND flash memory has been widely used for embedded systems as storage device and the flash memory file systems such as JFFS2, YAFFS/YAFFS2 have been adopted by these embedded systems. The flash memory file systems provide the high performance and overcome the limitations of flash memory. However, these file systems don't solve the slow mount time problem when a sudden power failure happens. In this paper, we proposed a flash memory management method for enhancing the recovery performance. The proposed method manages the flash memory block type and stores the block type information at recovery image block. When file operations are occurred, our method stores the file information at the metadata block before and after the file operation. When mounting the flash memory, our method only scans the recovery image blocks and metadata blocks. The proposed method reduces the mount time by seeking the metadata block locations fast by using the recovery image blocks. We implemented the proposed method and evaluation results show that our method reduces the mount time 13 ~ 46 % compared with YAFFS2.

• 센서학회지
• /
• 제28권4호
• /
• pp.216-220
• /
• 2019

In this paper, we propose a finite memory filter (estimator) for discrete-time stochastic linear systems with delays in state and measurement. A novel filtering algorithm is designed based on finite memory strategies, to achieve high estimation accuracy and stability under parametric uncertainties. The new finite memory filter uses a set of recent observations with appropriately chosen initial horizon conditions. The key contribution is the derivation of Lyapunov-like equations for finite memory mean and covariance of system state with an arbitrary number of time delays. A numerical example demonstrates that the proposed algorithm is more robust and accurate than the Kalman filter against dynamic model uncertainties.

• 제어로봇시스템학회논문지
• /
• 제10권2호
• /
• pp.178-184
• /
• 2004

In this paper, we designed an image processing system for the high speed line-scan camera which adopts the new memory model we proposed. As a resolution and a data rate of the line-scan camera are becoming higher, the faster image processing systems are needed. But many conventional systems are not sufficient to process the image data from the line-scan camera during a very short time. We designed the memory controller which eliminates the time for transferring image data from the line-scan camera to the main memory with high-speed SRAM and has a dual-port configuration therefore the DSP can access the main memory even though the memory controller are writing the image data. The memory controller is implemented by VHDL and Xilinx SPARTAN-IIE FPGA.

• 산업기술연구
• /
• 제15권
• /
• pp.51-62
• /
• 1995

Databases can crash due to various failures in computer systems. Recovery is a mechanism for restoring consistent data from damages caused by the by the failures and is an essential feature in database systems. This paper surveys recovery techniques for memory resident database systems. We first describe the basic architecture for memory resident database systems, and point out the main factors affecting their performance enhancement. Next, we explain the write-ahead logging(WAL), a recovery technique widely-used in most disk resident database systems, for easy understanding of basic recovery mechanisms. And then, we discuss some new concepts employed in memory resident database systems recovery. Finally, we present a representative memory resident database recovery technique, which is based on a special purpose hardware called HALO, as a case study.

• 대한임베디드공학회논문지
• /
• 제3권3호
• /
• pp.158-166
• /
• 2008

These days, mobile embedded systems adopt flash memory capable of XIP feature since they can reduce memory usage, power consumption, and software load time. XIP provides direct access to ROM and flash memory for processors. However, using XIP incurs unnecessary degradation of applications' performance because direct access to ROM and flash memory shows more delay than that to main memory. In this paper, we propose a memory management framework, dynamic XIP, which can resolve the performance degradation of using XIP. Using a constrained RAM cache, dynamic XIP can dynamically change XIP region according to page access pattern to reduce performance degradation in execution time or energy consumption resulting from native XIP problem. The proposed framework consists of a page profiler gathering applications' memory access pattern using PMU and an XIP manager deciding that a page is accessed whether in main memory or in flash memory. The proposed framework is implemented and evaluated in Linux kernel. Our evaluation shows that our framework can reduce execution time at most 25% and energy consumption at most 22% compared with using XIP-only case adopted in general mobile embedded systems. Moreover, the evaluation shows that in execution time and energy consumption, our modified LRU algorithm with code page filters can reduce more than at most 90% and 80% respectively compared with applying just existing LRU algorithm to dynamic XIP.

• International journal of advanced smart convergence
• /
• 제12권1호
• /
• pp.76-81
• /
• 2023

Journaling file systems are widely used to keep file systems in a consistent state against crash situations. As traditional journaling file systems are designed for block I/O devices like hard disks, they are not efficient for emerging byte-addressable NVM (non-volatile memory) media. In this article, we present a new in-memory journaling file system for NVM that is different from traditional journaling file systems in two respects. First, our file system journals only modified portions of metadata instead of whole blocks based on the byte-addressable I/O feature of NVM. Second, our file system bypasses the heavy software I/O stack while journaling by making use of an in-memory file system interface. Measurement studies using the IOzone benchmark show that the proposed file system performs 64.7% better than Ext4 on average.