• Title/Summary/Keyword: Server network bandwidth

Search Result 188, Processing Time 0.025 seconds

An Efficient Smoothing Algorithm for VOD Server With Limited Network Bandwidth (제한된 네트워크 대역폭을 갖는 VOD 서버를 위한 효율적인 스무딩 알고리즘)

  • Lee Myoun-Jae;Park Do-Soon
    • Proceedings of the Korean Information Science Society Conference
    • /
    • 2006.06d
    • /
    • pp.61-63
    • /
    • 2006
  • 스무딩은 가변 비트율로 저장된 비디오 데이터를 클라이언트로 전송할 때 일련의 고정 비트율로 전송할 수 있도록 전송 계획을 세우는 것이다. 이러한 스무딩 알고리즘들에는 CBA, MCBA, MVBA, PCRTT, e-PCRTT 알고리즘 등이 있다. 그러나 이 알고리즘들에서는 전송률 감소량의 최대화를 고려하지 않아서 새로운 구간에서 전송률의 감소가 요구되는 경우에 서버에서 제공할 수 있는 여분의 대역폭에 대한 효율적인 사용을 어렵게 할 수 있다. 따라서, 본 논문에서는 제한된 네트워크 대역폭을 갖는 VOD 서버에서 보다 많은 비디오 데이터를 서비스 할 수 있게 하기 위하여 새로운 구간의 전송률의 감소가 요구되는 경우에는 전송률 감소량을 최대로 하고 새로운 구간의 전송률의 증가가 요구되는 경우에는 전송률 증가량을 최소로 하는 스무딩 알고리즘을 제안하고 성능을 평가한다. 제안 알고리즘의 성능은 Final Four 비디오 소스를 이용하여 첨두 전송률, 전송률 변화 횟수, 평균 전송률 증가량, 평균 전송률 감소량을 비교 분석하였다.

  • PDF

Client-Side Deduplication to Enhance Security and Reduce Communication Costs

  • Kim, Keonwoo;Youn, Taek-Young;Jho, Nam-Su;Chang, Ku-Young
    • ETRI Journal
    • /
    • v.39 no.1
    • /
    • pp.116-123
    • /
    • 2017
  • Message-locked encryption (MLE) is a widespread cryptographic primitive that enables the deduplication of encrypted data stored within the cloud. Practical client-side contributions of MLE, however, are vulnerable to a poison attack, and server-side MLE schemes require large bandwidth consumption. In this paper, we propose a new client-side secure deduplication method that prevents a poison attack, reduces the amount of traffic to be transmitted over a network, and requires fewer cryptographic operations to execute the protocol. The proposed primitive was analyzed in terms of security, communication costs, and computational requirements. We also compared our proposal with existing MLE schemes.

Caching Strategies of Content-Centric Networking for the Performance of TICN Backbone (TICN 백본망 성능향상을 위한 CCN 적용 캐싱전략)

  • Park, Heungsoon;Kwon, Taewook
    • Journal of the Korea Institute of Military Science and Technology
    • /
    • v.17 no.5
    • /
    • pp.637-642
    • /
    • 2014
  • Military communications have limited resources such as poor bandwidth and insufficient link capacity. Numerous hosts in conventional IP networks, including the Tactical Information Communication Network (TICN), the tactical communication system of the Army, focus heavy traffic on a single server via end-to-end connections and in which data delivery latency is caused by exchanging routing information. Content-Centric Networking (CCN), one framework of Information-Centric Networking (ICN), is a novel Internet paradigm in which data can be accessed by the name of the content instead of the location information. This paper provides content caching strategies for applying CCN to military tactical IP networks based on TICN environment.

DDoS Defense Using the Exhaustiveness of Attackers (공격자의 자원소진특성을 이용한 분산서비스불능화 (DDoS) 공격에 대한 방어)

  • Jeong, Choong-Kyo
    • Journal of Industrial Technology
    • /
    • v.27 no.B
    • /
    • pp.77-82
    • /
    • 2007
  • A novel DDoS (Distributed Denial-of-Service) defense technique, Exaustiveness-Based Detection, is proposed in this work. It dispenses with the network congestion and the unfairness between users of the Defense-by-Offense technique by incorporating a kind of simple Detect-and-Block scheme (user identification), still improving the effectiveness of the defense in comparison to the original Defense-by-Offense technique. It uses SYN cookies to identify users in the granularity of ip address and to prevent ip address spoofing by the attacker. There can be, however, some probability of false negative (denying service to good clients), if the attacker wisely adapt to the new technique by saving some portion of its bandwidth resource and later mimicking good clients. Quantitative analysis the requirement for the good clients to be safe from the false negative is provided and a procedure to design the server capacity is explained.

  • PDF

Machine Learning based Bandwidth Prediction for Dynamic Adaptive Streaming over HTTP

  • Yoo, Soyoung;Kim, Gyeongryeong;Kim, Minji;Kim, Yeonjin;Park, Soeun;Kim, Dongho
    • Journal of Advanced Information Technology and Convergence
    • /
    • v.10 no.2
    • /
    • pp.33-48
    • /
    • 2020
  • By Digital Transformation, new technologies like ML (Machine Learning), Big Data, Cloud, VR/AR are being used to video streaming technology. We choose ML to provide optimal QoE (Quality of Experience) in various network conditions. In other words, ML helps DASH in providing non-stopping video streaming. In DASH, the source video is segmented into short duration chunks of 2-10 seconds, each of which is encoded at several different bitrate levels and resolutions. We built and compared the performances of five prototypes after applying five different machine learning algorithms to DASH. The prototype consists of a dash.js, a video processing server, web servers, data sets, and five machine learning models.

A Novel Approach of Using Data Flipping for Efficient Energy on the Internet of Things

  • Ziyad Almudayni;Ben Soh;Alice Li
    • Journal of information and communication convergence engineering
    • /
    • v.21 no.3
    • /
    • pp.185-191
    • /
    • 2023
  • The Internet of Things (IoT) can be defined as the connection of devices, sensors, and actors via the Internet to a single network to provide services to end-users. Owing to the flexibility and simplicity of IoT devices, which impart convenience to end-users, the demand for these devices has increased significantly in the last decade. To make these systems more scalable, achieve a larger number of connected devices, and achieve greater economic success, it is vital to develop them by considering parameters such as security, cost, bandwidth, data rate, and power consumption. This study aims to improve energy efficiency and prolong the lifetime of IoT networks by proposing a new approach called the constrained application protocol CoAP45. This approach reduces the number of updates to the CoAP server using a centralized resource. The simulation results show that the proposed approach outperforms all existing protocols.

Periodic I/O Scheduling for the Storage of MPEG-DASH Video Servers

  • Seong Chae Lim
    • International journal of advanced smart convergence
    • /
    • v.13 no.3
    • /
    • pp.31-40
    • /
    • 2024
  • The proliferation of video streaming services has led to a need for flexible networking protocols. As a result, the Dynamic Adaptive Streaming over HTTP (MPEG-DASH) protocol has emerged as a dominant streaming protocol due to its ability to dynamically adjust playback bitrates according to the end-user's network conditions. In this paper, we propose a novel I/O scheduling scheme tailored for the storage of MPEG-DASH-enabled video servers. Using the renowned rate-reservation (RR) algorithm and bulk-SCAN mechanism, our proposed scheme improves storage bandwidth utilization while ensuring seamless playback of streams with varying bitrates. In addition, we provide a mechanism for reclaiming the idle I/O time typically incurred while retrieving video segments from storage. Consequently, our scheme offers practical solutions for reducing the storage costs of MPEG-DASH video servers. With a simple cost model, we evaluate the performance enhancements achieved by our proposed I/O scheduling scheme.

Spatial and Temporal Resolution Selection for Bit Stream Extraction in H.264 Scalable Video Coding (H.264 SVC에서 비트 스트림 추출을 위한 공간과 시간 해상도 선택 기법)

  • Kim, Nam-Yun;Hwang, Ho-Young
    • Journal of Korea Multimedia Society
    • /
    • v.13 no.1
    • /
    • pp.102-110
    • /
    • 2010
  • H.264 SVC(Scalable Video Coding) provides the advantages of low disk storage requirement and high scalability. However, a streaming server or a user terminal has to extract a bit stream from SVC file. This paper proposes a bit stream extraction method which can get the maximum PSNR value while date bit rate does not exceed the available network bandwidth. To do this, this paper obtains the information about extraction points which can get the maximum PSNR value offline and decides the spatial/temporal resolution of a bit stream at run-time. This resolution information along with available network bandwidth is used as the parameters to a bit stream extractor. Through experiment with JSVM reference software, we proved that proposed bit stream extraction method can get a higher PSNR value.

The Push Framework for UMPC Tactical Data Link(TDL) Based on The Legacy Radio (레거시 라디오 기반의 UMPC 전술 데이터 링크 Push 프레임워크)

  • Sim, Dong-Sub;Shin, Ung-Hee;Kim, Ki-Hyung
    • Journal of the Korea Institute of Military Science and Technology
    • /
    • v.13 no.3
    • /
    • pp.397-404
    • /
    • 2010
  • Recently, there is on-going research about tactical data link system based on the legacy radio. Tactical data link based on legacy radio is operated mostly in narrow bandwidth under 25khz. Communicating traffics in nodes participated at network need to be minimized for distributing tactical data in narrow bandwidth. In addition, the data distributing structure is necessary for distributing tactical informations such as a situation awareness and so on to war fighters. However, conventional server-client system wastes a lot of time to obtain information for war fighters as user uses pull system to gather necessary information by seeking it one by one. Especially, the fighter pilot is supposed to dedicate into a situation awareness and fight mission in every seconds but seeking information of a user terminal while aircraft maneuvering affects as obstacle to concentrate engaging hostiles. therefore, push technology, the tactical data distributing system, is necessary for war fighters to receive fixed tactical data automatically without putting attention to it. This paper propose the UMPC tactical data link push framework. the UMPC tactical data link is a tactical data link system based on the legacy radio. Proposed push framework is verified by composing experiment environment and testing.

Attack Scenarios and Countermeasures using CoAP in IoT Environment (IoT기기에서 SSDP 증폭 공격을 이용한 공격기법 및 대응 방안)

  • Oh, Ju-Hye;Lee, Keun-Ho
    • Journal of the Korea Convergence Society
    • /
    • v.7 no.4
    • /
    • pp.33-38
    • /
    • 2016
  • DDoS attack has been continuously utilized that caused the excessively large amount of traffic that network bandwidth or server was unable to deal with paralyzing the service. Most of the people regard NTP as the biggest cause of DDoS. However, according to recently executed DDoS attack, there have been many SSDP attack in the use of amplified technique. According to characteristics of SSDP, there is no connection for making a forgery of source IP address and amplified resources feasible. Therefore, it is frequently used for attack. Especially, as it is mostly used as a protocol for causing DDoS attack on IoT devices that constitute smart home including a wireless router, media server, webcam, smart TV, and network printer. Hereupon, it is anticipated for servers of attacks to gradually increase. This might cause a serious threat to major information of human lives, major government bodies, and company system as well as on IoT devices. This study is intended to identify DDoS attack techniques in the use of weakness of SSDP protocol occurring in IoT devices and attacking scenario and counter-measures on them.