• Journal of KIISE:Computer Systems and Theory
• /
• v.36 no.6
• /
• pp.476-487
• /
• 2009

The virtual memory system in 4.4 BSD operating systems exploits a clustering scheme to reduce disk I/Os in paging out (or flushing) modified pages that are intended to be replaced in order to make free rooms in memory. Upon the page out of a victim page, the scheme stores a cluster (or group) of modified pages contiguous with the victim in the virtual address space to swap disk at a single disk write. However, it fails to find large clusters of contiguous pages if applications change pages not adjacent with each other in the virtual address space. To address the problem, we propose a new clustering scheme called Multiple-Object Clustering (MOC), which together stores multiple clusters in the virtual address space at a single disk write instead of paging out the clusters to swap space at separate disk I/Os. This multiple-cluster transfer allows the virtual memory system to significantly decrease disk writes, thus improving the page-out performance. Our experiments in the FreeBSD 6.2 show that MOC improves the execution times of realistic benchmarks such as NS2, Scimark2 SOR, and nbench LU over the traditional clustering scheme ranging from 9 to 45%.

• Journal of IKEEE
• /
• v.23 no.1
• /
• pp.80-90
• /
• 2019

Even though Non-Volatile Memory (NVM) is used for data storage, a page table should be built to access data in it. And this observation leads us to the Persistent Page Table (PPT) scheme that keeps the page table in NVM persistently. By the way, processors have different page table structures and really operational page table cannot be built without virtual and physical addresses of NVM. However, those addresses are determined dynamically when NVM storage is attached to the system. Thus, the PPT should have system-independent and also address-independent structure and really working system-dependent page table should be built from the PPT. Moreover, entries of PPT should be updated atomically and, in this paper, we describe the design of PPT that meets those requirements. And we investigate how file systems can decrease the journaling overhead with the swap operation, which is a new operation created by the PPT. We modified the Ext4 file system in Linux and experiments conducted with Filebench workloads show that the swap operation enhances file system performance up to 60%.

• Proceedings of the Korean Institute of Communication Sciences Conference
• /
• 1988.10a
• /
• pp.101-108
• /
• 1988

In this paper packet-swap Accptant Queveing system with synchronous single server and finite storage space is proposed for throughput improvement. Queueling systems are analyzed with Minisint Approximation reported by J.F CHANG and R.F Chang. Comparison between PSA. Queveing system and First-Come First Acceptant Queveing system via throughput and blocking probabilliy of test octet was performed The comparison showed that PAS Queweing system perfumes better than j.F ChANG’s Queveing system.

• Proceedings of the Korea Contents Association Conference
• /
• 2014.11a
• /
• pp.3-4
• /
• 2014

현대 운영 체제에서 가상 메모리 관리 기술은 응용 프로그램에게 가상의 큰 주소 공간을 제공하는 방법이다. 이러한 기술은 메인 메모리와 저장 장치를 이용하여 자주 접근되는 데이터는 메인 메모리에 덜 접근되는 데이터는 저장 장치에 저장한다. 본 연구는 차세대 저장 장치를 메모리 확장을 위한 장치로 가정했을 때의 성능 향상 기법을 제안한다. 본 연구진은 제안된 방법을 Linux 3.14.3을 구현하였고, 초고속 저장 장치를 이용하여 평가한 결과 기존 SWAP 시스템에 비해 20% 정도의 성능 향상 효과가 있음을 보였다.

• Proceedings of the Korea Information Processing Society Conference
• /
• 2008.11a
• /
• pp.1035-1038
• /
• 2008

가상 메모리의 스왑 저장 장치로서 플래시 메모리를 사용하는 경우, 시스템을 기동할 때 스왑 영역의 초기화를 위한 삭제 연산이 요구되어 기동 시간이 오래 걸리는 문제점이 있다. 본 논문에서는 스왑 영역의 플래시 메모리 내용을 모두 삭제하지 않고 일부만을 삭제함으로서 기동 시간을 줄일 수 있는 방법을 연구하였다.

• The Journal of the Institute of Internet, Broadcasting and Communication
• /
• v.15 no.5
• /
• pp.155-162
• /
• 2015

This paper proposes a deterministic optimization algorithm to solve economic load dispatch problem with quadratic convex fuel cost function. The proposed algorithm primarily partitions a generator with prohibited zones into multiple generators so as to place them afield the prohibited zone. It then sets initial values to $P_i{\leftarrow}P_i^{max}$ and reduces power generation costs of those incurring the maximum unit power cost. It finally employs a swap optimization process of $P_i{\leftarrow}P_i-{\beta}$, $P_j{\leftarrow}P_j+{\beta}$ where $_{max}\{F(P_i)-F(P_i-{\beta})\}$ > $_{min}\{F(P_j+{\beta})-F(P_j)\}$, $i{\neq}j$, ${\beta}=1.0,0.1,0.01,0.001$. When applied to 3 different 15-generator cases, the proposed algorithm has consistently yielded optimized results compared to those of heuristic algorithms.

• Journal of KIISE
• /
• v.43 no.8
• /
• pp.841-847
• /
• 2016

As network technology advances, a larger number of devices are connected through the Internet. Recently, cloud storage services are gaining popularity, as they are convenient to access anytime and anywhere. Among cloud storage services, object storage is the representative one due to their characteristics of low cost, high availability, and high durability. One limitation of object storage services is that they can access data on the cloud only through the HTTP-based RESTful APIs. In our work, we resolve this limitation with the in-memory file system which provides a POSIX interface to the file system users and communicates with cloud object storages with RESTful APIs. In particular, our flush mechanism is compatible with existing file systems, as it is based on the swap mechanism of the Linux kernel. Our in-memory file system backed by cloud storage reduces the performance overheads and shows a better performance than S3QL by 57% in write operations. It also shows a comparable performance to tmpfs in read operations.

• The Journal of Korean Institute of Communications and Information Sciences
• /
• v.28 no.4B
• /
• pp.351-360
• /
• 2003

All-optical network provede unlimited for bandwidth, the very low bit error rate, and the transparency to IP. Optical networks promise to be the next generation networks that can meet the higher bandwidth demands. However, the number of wavelengths is often not large enough to help a large amount of nodes. The blocking by wavelength contention can be reduced by wavelength conversion, which can't perfectly resolve all situations. Because of that, a lot of groups have proposed unique wavelength assingment algorithms and protocols. Most schemes don't support the priority concepts. This paper describes the unique priority scheme based on the throughput. In this paper, we apply our priority scheme to SWAP(Simple Wavelength Assignment Protocol) and show the performance of the proposed priority scheme. Our proposed priority scheme can be a better solution to solve an important problem about the blocking by wavelength contention into WDM optical networks.

• Journal of the Korea Society of Computer and Information
• /
• v.26 no.10
• /
• pp.27-35
• /
• 2021

Local search has been used to solve various combinatorial optimization problems. One of the most important factors in local search is the method of generating a neighbor solution. In this paper, we propose neighbor generation strategies of local search for permutation-based combinatorial optimization, and compare the performance of each strategies targeting the traveling salesman problem. In this paper, we propose a total of 10 neighbor generation strategies. Basically, we propose 4 new strategies such as Rotation in addition to the 4 strategies such as Swap which have been widely used in the past. In addition, there are Combined1 and Combined2, which are made by combining basic neighbor generation strategies. The experiment was performed by applying the basic local search, but changing only the neighbor generation strategy. As a result of the experiment, it was confirmed that the performance difference is large according to the neighbor generation strategy, and also confirmed that the performance of Combined2 is the best. In addition, it was confirmed that Combined2 shows better performance than the existing local search methods.

• Journal of the Korea Institute of Information and Communication Engineering
• /
• v.23 no.7
• /
• pp.875-880
• /
• 2019

A RC4 stream cipher is widely used for security applications such as IEEE 802.11 WEP, IEEE 802.11i TKIP and so on, because it can be simply implemented to dedicated circuits and achieve a high-speed encryption. RC4 is also used for systems with limited resources like IoT, but there are performance limitations. RC4 consists of two stages, KSA and PRGA. KSA performs initialization and randomization of S-box and K-box and PRGA produces cipher texts using the randomized S-box. In this paper, we initialize the S-box and K-box in the randomization of the KSA stage to reduce the initialization delay. In the randomization, we use clusters to process swap operation between elements of S-box in parallel and can generate two cipher texts per clock. The proposed RC4 cipher hardware can initialize S-box and K-box without any delay and achieves about 2 times to 6 times improvement in KSA randomization and key stream generation.