• Annual Conference of KIPS
• /
• 2003.05b
• /
• pp.1209-1212
• /
• 2003

오늘날 인터넷 환경에서의 망은 유무선의 환경이 통합된 하나의 망으로 달수 있다. 일반적인 TCP에서는 무선망에서의 핸드오프나 비트오류 등으로 인한 패킷 손실이 발생하는 경우에도 흔잡제어 알고리즘으로 손실된 패킷을 복구하게 되며 이러한 복구는 혼잡윈도우를 줄이게 됨으로 인해 현저히 TCP의 처리량을 감소시키게 된다. 본 논문에서는 유무선이 통합된 망에서 데이터 전송 효율을 높일 수 있는 알고리즘을 제시하고자 한다. 이 알고리즘에서는 홈서버를 사용하여 무선망에서 발생한 패킷 손실이 종단간의 재전송이 아닌 홈서버에서 지역 재전송을 함으로써 유무선망의 부하를 줄이고 흔잡제어 알고리즘이 실행되는 것을 방지하여 TCP의 성능향상을 가져올 수 있으며 캐쉬메모리에 재전송 패킷을 보관하여 재전송함으로써 보다 빠른 재전송효과를 얻을 수 있다.

• The Transactions of the Korean Institute of Electrical Engineers D
• /
• v.54 no.10
• /
• pp.615-619
• /
• 2005

Token coherence protocol has many good reasons against snooping/directory-based protocol in terms of latency, bandwidth, and complexity. Token counting easily maintains correctness of the protocol without global ordering of request which is basis of other dominant cache coherence protocols. But this lack of global ordering causes starvation which is not happening in snooping/directory-based protocols. Token coherence protocol solves this problem by providing an emergency mechanism called persistent request. It enforces other processors in the competition (or accessing same shared memory block, to give up their tokens to feed a starving processor. However, as the number of processors grows in a system, the frequency of starvation occurrence increases. In other words, the situation where persistent request occurs becomes too frequent to be emergent. As the frequency of persistent requests increases, not only the cost of each persistent matters since it is based on broadcasting to all processors, but also the increased traffic of persistent requests will saturate the bandwidth of multiprocessor interconnection network. This paper proposes a new request mechanism that defines order of requests to reduce occurrence of persistent requests. This ordering mechanism has been designed to be decentralized since centralized mechanism in both snooping-based protocol and directory-based protocol is one of primary reasons why token coherence protocol has advantage in terms of latency and bandwidth against these two dominant Protocols.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.9 no.12
• /
• pp.4856-4871
• /
• 2015

Backpressure based scheduling has been considered as a promising technique for improving the throughput of a wide range of communication networks. However, this scheduling technique has not been well studied for heterogeneous wireless networks. In this paper, we propose a virtual-queue based backpressure scheduling (VQB) algorithm for heterogeneous multi-hop wireless networks. The VQB algorithm introduces a simple virtual queue for each flow at a node for backpressure scheduling, whose length depends on the cache size of the node. When calculating flow weights and making scheduling decisions, the length of a virtual queue is used instead of the length of a real queue. We theoretically prove that VQB is throughput-optimal. Simulation results show that the VQB algorithm significantly outperforms a classical backpressure scheduling algorithm in heterogeneous multi-hop wireless networks in terms of the packet delivery ratio, packet delivery time, and average sum of the queue lengths of all nodes per timeslot.

• KSII Transactions on Internet and Information Systems (TIIS)
• /
• v.12 no.3
• /
• pp.1002-1015
• /
• 2018

The advent of Internet of Things (IoT) and the evident inadequacy of Cloud networks concerning management of numerous end nodes have brought about a shift of paradigm giving birth to Fog computing. Fog computing is an extension of Cloud computing that extends Cloud resources at the edge of the network, closer to the user. Cloud computing has become one of the essential needs of people over the Internet but with the emerging concept of IoT, traditional Clouds seem inadequate. IoT entails extremely low latency and for that, the Cloud servers that are distant and unknown to the user appear to be unsuitable. With the help of Fog computing, the Fog devices installed would be closer to the user that will provide an immediate storage for the frequently needed data. This paper discusses data migration between different storage types especially between Cloud devices and then presents a mechanism to migrate data between Cloud and Fog Layer. We call this mechanism Adaptive Deadline-Aware Scheme (ADAS) for Data migration between Cloud and Fog. We will demonstrate that we can access and process latency sensitive "hot" data through the proposed ADAS more efficiently than with a traditional Cloud setup.

• Proceedings of the Korean Institute of Information and Commucation Sciences Conference
• /
• 2002.05a
• /
• pp.201-204
• /
• 2002

Although the reuse of the cached data for scrolling the map reduces the amount of passed data between client and server, it needs the conversions of data coordinates, selective deletion of objects and cache compaction at client. The conversion is time intensive operation due to limited resources of mobile phones such as low computing power, small memory. Therefore, for the efficient map control in the vector map service based mobile phone, it is necessary to study the method for reducing wireless network bandwidth and for overwhelming the limited resources of mobile phone as well. This paper proposes the methods for caching pre-received spatial objects in client-server systems for mobile GIS. We also analyze the strengths and drawbacks between the reuse of cached data and transmission of raw data respectively.

• Proceedings of the Korean Society of Broadcast Engineers Conference
• /
• 2016.11a
• /
• pp.45-46
• /
• 2016

본 논문은 CDN(Content Delivery Network)의 동적 캐싱 방식을 기반으로 하는 적응적 TTL(Time-To-Live) 할당기법을 제안한다. 이는 클라이언트가 실시간 개인방송을 시청하는 중 지나간 과거의 특정 장면을 다시 시청할 때 근원(Origin)서버의 부하를 효율적으로 줄일 수 있을 뿐만 아니라 캐시(Cache)서버의 저장 공간도 효율적으로 사용할 수 있는 장점이 있다. 따라서 본 논문에서 제안하는 적응적 TTL 할당기법은 개인방송 시청자들이 지나간 과거의 영상들을 선택적으로 시청할 때 보다 나은 서비스를 제공할 수 있을 것으로 기대된다.

• Journal of KIISE:Computing Practices and Letters
• /
• v.6 no.6
• /
• pp.643-656
• /
• 2000

MPOA proposed and standardized by the ATM Forum is a protocol tllat provides effective bridging and routing for ATM networks in a diverse network environment. Its plimary goal is to transfer unicast data effectively among the subnets. In this paper, MPOA components are implemented using the SDL(Specification and Description Language) which the ITU has standardized for the development of communication systems. In addition, MPOA procedures for various operations such as address translation for packets from upper layers, Ingress/Egress cache management and shortcut configuration, are examined with tlle help of the SDT(SDL Design Too]) simulator.

• Journal of Information Processing Systems
• /
• v.18 no.2
• /
• pp.245-257
• /
• 2022

In the Internet of Things (IoT) era, the types of devices used by one user are becoming more diverse and the number of devices is also increasing. However, a forensic investigator is restricted to exploit or collect all the user's devices; there are legal issues (e.g., privacy, jurisdiction) and technical issues (e.g., computing resources, the increase in storage capacity). Therefore, in the digital forensics field, it has been a challenge to acquire information that remains on the devices that could not be collected, by analyzing the seized devices. In this study, we focus on the fact that multiple devices share data through account synchronization of the online platform. We propose a novel way of identifying the user's interest through analyzing the remnants of targeted advertising which is provided based on the visited websites or search terms of logged-in users. We introduce a detailed methodology to pick out the targeted advertising from cache data and infer the user's interest using deep learning. In this process, an improved learning model considering the unique characteristics of advertisement is implemented. The experimental result demonstrates that the proposed method can effectively identify the user interest even though only one device is examined.

• Journal of Platform Technology
• /
• v.11 no.4
• /
• pp.71-79
• /
• 2023

In this paper, we analyze the characteristics of machine learning workloads and, based on them, propose a distributed in-memory caching technique to improve the performance of machine learning workloads. The core of machine learning workload is model training, and model training is a computationally intensive task. Performing machine learning workloads in a Kubernetes-based cloud environment in which the computing framework and storage are separated can effectively allocate resources, but delays can occur because IO must be performed through network communication. In this paper, we propose a distributed in-memory caching technique to improve the performance of machine learning workloads performed in such an environment. In particular, we propose a new method of precaching data required for machine learning workloads into the distributed in-memory cache by considering Kubflow pipelines, a Kubernetes-based machine learning pipeline management tool.

• Annual Conference of KIPS
• /
• 2011.11a
• /
• pp.116-119
• /
• 2011

다중코어(Many-Core) 시스템은 많은 코어들이 상호연결망을 통해서 연결되어있는 시스템으로, 단일코어나 멀티코어 시스템에 비해 보다 많은 병렬 컴퓨팅 자원을 지원한다. Amdahl 의 법칙에 의하면 병렬화되어 처리하는 부분은 이론적으로 프로세서의 개수에 비례하게 가속화 될 수 있지만, 상호연결망에서의 전송 지연을 비롯한 많은 요인에 의해서 성능의 가속화가 저해된다. 특히 캐시 일관성 규약(Cache Coherence Protocol)을 지원하는 대부분의 다중코어 시스템에서는 병렬화를 함에 있어서 캐시 미스로 인해 발생하는 데이터의 전송 지연이 성능에 많은 영향을 미칠 수 있다. 따라서 효과적인 병렬 프로그램을 위해서는 캐시 구조에 대한 이해를 바탕으로 상호연결망에 대한 연구가 필요하다. 본 논문에서는 메쉬(Mesh) 구조의 64 코어 다중코어 시스템인 TilePro64 를 이용하여 상호연결망의 데이터 전송 지연에 따른 프로그램 성능의 민감도를 측정하였다. 결과적으로 코어간 거리(Hop)가 늘어날수록 작업의 수행시간이 평균적으로 4.27%씩 선형적으로 증가하는 관계가 있는 것으로 나타났다.