The Infona portal uses cookies, i.e. strings of text saved by a browser on the user's device. The portal can access those files and use them to remember the user's data, such as their chosen settings (screen view, interface language, etc.), or their login data. By using the Infona portal the user accepts automatic saving and using this information for portal operation purposes. More information on the subject can be found in the Privacy Policy and Terms of Service. By closing this window the user confirms that they have read the information on cookie usage, and they accept the privacy policy and the way cookies are used by the portal. You can change the cookie settings in your browser.
The previous surveys show that more than 90% of the flows in typical data center networks are smaller than 100KB in size, while most bytes transmitted are from a few large flows. The small flows are usually sensitive to their completion times while the large flows require a high throughput. The previous works usually either achieve low completion times for small flows or high throughput for large...
We generally utilize high-speed transport protocols to transfer and share large-scale data in long-distance broadband networks. However, the difference in performance between protocols when some kinds of protocols share the same link increases. Consequently, we propose a control method of discarding packets corresponding to protocol types that takes into consideration high-speed performance and fairness...
Rapid OS deployment has always been a challenging task in enterprise data center environments. In particular, deploying hundreds of thousands of systems is necessary to automate the installation process. This paper describes BitTorrent-based OS provisioning method to afford large-scale data centers. It can save the time necessary for installing and deploying operating systems and software.
Many research works have been carried out in differentiated Services, to study the fairness issues between aggregates of both TCP and UDP. Due to multimedia demands, the UDP flows can also demand assured service. Hence, there is a need to protect certain UDP flows which require the same fair treatment as TCP. In such cases, the UDP flows are characterized as AS UDP with target rates and are modeled...
This paper introduces a hierarchical trick play mechanism for video streaming. Trick level concept is introduced to support multiple trick rates, reduce delay when switching from high to low rate and minimize network bandwidth usage. Dedicated trick play streams on server side are created to improve user experience and reduces client processing requirement. It is efficient and easy to implementation.
In this paper, we study the performance of active queue management algorithms for packet queueing in Internet routers in the presence of new versions of the TCP congestion control mechanism. In particular, we compare the performance of tail drop, adaptive RED, AVQ, PI, REM and AN-AQM queueing in four TCP cases: (i) the classic New Reno protocol, (ii) the Sack protocol, (iii) the Fack protocol, (iv)...
In high-speed and long-distance networks, TCP NewReno, which is the most popular version of TCP, cannot achieve sufficient throughput because of the essential nature of the congestion control mechanism of TCP. Therefore, Compound TCP has been proposed. Compound TCP can achieve a considerably larger throughput than TCP NewReno in high-speed and long-distance networks. The congestion control mechanism...
TCP is the main and most widely used transport protocol for reliable communication. Because of its widespread need, researchers have been studying and proposing new TCP variants trying to improve its behavior towards congestion to make it use the most available bandwidth while preserving a logical level of fairness towards other protocols. This paper aims at evaluating and comparing the performance...
Per-flow unfairness of TCP throughput in the IEEE 802.11 wireless LAN environment has been reported. Although a number of researchers have proposed various methods for alleviating the unfairness, they evaluated the fairness of their methods separately from the network bandwidth utilization, meaning that they did not consider the trade-off relationships between fairness and bandwidth utilization. In...
Parallel transport control protocol (TCP) has been used to effectively utilize bandwidth for data intensive applications over high bandwidth-delay product (BDP) networks. On the other hand, it has been argued that, a single based TCP connection with proper modification such as HSTCP can emulate and capture the robustness of parallel TCP and can well replace it. In this work a Comparison between single-based...
RED is the only active queue management mechanism recommended by IETF. Though RED is successful in network congestion control and link utilization enhancement, it suffers from two problems, the stability and the fairness. The fairness problem means 1) RED has little affects on unresponsive flows in congestion control - the UDP flow for example; 2) responsive flows with different packet size or round...
Peer-to-peer (P2P) multimedia sharing is widely envisioned as one of the major applications of wireless communication network in the future, and wireless bandwidth constraint is its principal obstacle. Aiming to save bandwidth, a novel P2P lookup protocol named BF-chord is proposed, in which the finger table in chord is changed, shared files' information is compressed into Bloom filters (BFs) before...
Recently, wireless LANs have been used widely because they are flexible and easy to install. However, there are unfairness problems in wireless LANs, in that a small number of stations (STAs) may obtain a large throughput while other STAs hardly get any bandwidth. One of the unfairness problems occurs between upstream and downstream flows because the access point (AP) and each STA have the same transmission...
As the Internet evolves, traffic management has become a high-priority problem. Without a network centric traffic management scheme, severe fairness problems can arise. For instance, unfairness occurs with TCP traffic in the presence of careless or subversive misbehaving flows, or even in the presence of other TCP flavors if we consider inter-protocol unfairness. In this paper we propose a new AQM...
With the best-effort TCP transporting foreground traffic, below-best-effort transport protocols are developed to transport background traffic. However, we argue in this paper that: 1) two fixed levels of priority are not enough, and 2) the background and foreground transfers are subjective concepts which can only be decided by users but not arbitrarily determined by the network according to the application...
The current DiffServ proposal suffers from an unfair distribution of excess network bandwidth among aggregates and also between TCP and UDP flows. To alleviate this problem, in this paper a new time sliding window based marker has been proposed. The improvement in fairness among TCP aggregates is proved to be over a large range of provision level (20% to 90%) compared to the recently proposed Improved...
Throughput unfairness in TCP between upflows over an IEEE 802.11 WLAN is a well known issue. However, this holds true only if the network has no or very little packet loss. The packet loss rate in a wired network is, in fact, not negligible. We quantitatively evaluated the effect of TCP data packet loss in a wired network connected to a wireless LAN. The simulation results show that an increase in...
The important communications or real time applications over IP networks such as the Internet have received considerable attention over recent years. However, it is very difficult for the standard TCP to maintain QoS (i.e. bandwidth) which is demanded by important communications or real time applications, when multiple TCP connections share the bandwidth of a bottleneck link. Thus, it is useful for...
In this paper, we propose an adaptive pacing scheme at the link layer for IEEE 802.11 based multihop wireless networks. Our objective is to improve the performance of higher layer protocols without any modifications to them. Our adaptive pacing scheme estimates the four-hop transmission delay in the network path without incurring any additional overheads, and accordingly paces the packets to reduce...
Set the date range to filter the displayed results. You can set a starting date, ending date or both. You can enter the dates manually or choose them from the calendar.