SDN‐based architecture for providing quality of service to high‐performance distributed applications

Author(s):  
Alexandre T. Oliveira ◽  
Bruno José C. A. Martins ◽  
Marcelo F. Moreno ◽  
Antônio Tadeu A. Gomes ◽  
Artur Ziviani ◽  
...  

Author(s):  
Ankur Gupta ◽  
Lalit K. Awasthi

P2P networks have caught the imagination of the research community and application developers with their sheer scalability and fault-tolerance characteristics. However, only content-sharing applications based on the P2P concept have reached the desired level of maturity. The potential of the P2P concept for designing the next-generation of real-world distributed applications can be realized only if a comprehensive framework quantifying the performance related aspects of all classes of P2P applications is available. Researchers have proposed some QoS (Quality-of-Service) parameters for content-sharing P2P applications based on response time and delay, but these do not cover the gamut of application domains that the P2P concept is applicable to. Hence, this research paper proposes an early QoS framework covering various classes of P2P applications; content distribution, distributed computing and communication and collaboration. Early results from the prototype implementation of the Peer Enterprises framework (a cross-organizational P2P collaborative application) are used as a basis for formulation of the QoS parameters. The individual performance measures which comprise the QoS framework are also discussed in detail along with some thoughts on how these can be complied with. The proposed framework would hopefully lead to quantifiable Service-Level Agreements for a variety of peer-to-peer services and applications.



Author(s):  
Ankur Gupta ◽  
Lalit K. Awasthi

P2P networks have caught the imagination of the research community and application developers with their sheer scalability and fault-tolerance characteristics. However, only content-sharing applications based on the P2P concept have reached the desired level of maturity. The potential of the P2P concept for designing the next-generation of real-world distributed applications can be realized only if a comprehensive framework quantifying the performance related aspects of all classes of P2P applications is available. Researchers have proposed some QoS (Quality-of-Service) parameters for content-sharing P2P applications based on response time and delay, but these do not cover the gamut of application domains that the P2P concept is applicable to. Hence, this research paper proposes an early QoS framework covering various classes of P2P applications; content distribution, distributed computing and communication and collaboration. Early results from the prototype implementation of the Peer Enterprises framework (a cross-organizational P2P collaborative application) are used as a basis for formulation of the QoS parameters. The individual performance measures which comprise the QoS framework are also discussed in detail along with some thoughts on how these can be complied with. The proposed framework would hopefully lead to quantifiable Service-Level Agreements for a variety of peer-to-peer services and applications.



2020 ◽  
Vol 13 (6) ◽  
pp. 2214-2223 ◽  
Author(s):  
Zhou Zhou ◽  
Hongmin Wang ◽  
Huailing Shao ◽  
Lifeng Dong ◽  
Junyang Yu


2012 ◽  
Vol 263-266 ◽  
pp. 1858-1863
Author(s):  
Jin He Zhou ◽  
Guo Min Xia

Diffserv-aware and Traffic Engineering combine the advantages of MPLS, Traffic Engineering (TE) and Differentiated Services (Diffserv, DS) to provide high performance and Quality of Service(QoS) in networks. We have designed three scenarios on Juniper Networks platforms to analyze the packet loss rate and delay for video, voice and data. The results show that MPLS DS-TE can improve the QoS for differentiated service effectively. The research has practical value for the development of DS-TE based on MPLS.





2021 ◽  
Author(s):  
Allen Yen-Cheng Yu

Many large-scale online applications enable thousands of users to access their services simultaneously. However, the overall service quality of an online application usually degrades when the number of users increases because, traditionally, centralized server architecture does not scale well. In order to provide better Quality of Service (QoS), service architecture such as Grid computing can be used. This type of architecture offers service scalability by utilizing heterogeneous hardware resources. In this thesis, a novel design of Grid computing middleware, Massively Multi-user Online Platform (MMOP), which integrates the Peer-to-Peer (P2P) structured overlays, is proposed. The objectives of this proposed design are to offer scalability and system design flexibility, simplify development processes of distributed applications, and improve QoS by following specified policy rules. A Massively Multiplayer Online Game (MMOG) has been created to validate the functionality and performance of MMOP. The simulation results have demonstrated that MMOP is a high performance and scalable servicing and computing middleware.



2020 ◽  
Vol 19 (2) ◽  
pp. 412-445 ◽  
Author(s):  
Sergej Andreev ◽  
Roman Tregubov ◽  
Alexander Mironov

The paper proposes a solution to the problem of selecting the bandwidth capabilities of digital communication channels of a transport communication network taking into account the imbalance of data traffic by priorities. The algorithm for selecting bandwidth guarantees the minimum costs associated with renting digital communication channels with optimal bandwidth, provided that the requirements for quality of service of protocol data blocks of the first, second, and k-th priority in an unbalanced in terms of priorities transport communication network are met. At the first stage of solving the problem, using the method of Lagrange multipliers, an algorithm for selecting the capacities of digital communication channels for a balanced in terms of priorities transport network was developed. High performance of this algorithm was ensured by applying algebraic operations on matrices (addition, multiplication, etc.). At the second stage of solving the problem, using the generalized Lagrange multipliers method, we compared the conditional extrema of the cost function for renting digital communication channels for single active quality of service requirements for protocol data blocks, for all possible pairs of active quality of service requirements for protocol data blocks, for all possible triples of active requirements for the quality of service of protocol data units, and so on up to the case when all the requirements for quality of service maintenance of protocol data units are active simultaniously. At the third stage of solving the problem, an example of selecting the bandwidth capabilities of digital communication channels of the unbalanced by priorities transport network consisting of eight routers serving protocol data blocks of three priorities was considered. At the fourth stage of the solution of the problem of the choice of carrying capacities the estimation of efficiency of the developed algorithm by a method of simulation modeling was carried out. To this end, in the environment of the network simulator OMNet ++, the unbalanced in terms of priority transport communication network consisting of eight routers connected by twelve digital communication channels with optimal bandwidth was investigated.



2021 ◽  
Author(s):  
Allen Yen-Cheng Yu

Many large-scale online applications enable thousands of users to access their services simultaneously. However, the overall service quality of an online application usually degrades when the number of users increases because, traditionally, centralized server architecture does not scale well. In order to provide better Quality of Service (QoS), service architecture such as Grid computing can be used. This type of architecture offers service scalability by utilizing heterogeneous hardware resources. In this thesis, a novel design of Grid computing middleware, Massively Multi-user Online Platform (MMOP), which integrates the Peer-to-Peer (P2P) structured overlays, is proposed. The objectives of this proposed design are to offer scalability and system design flexibility, simplify development processes of distributed applications, and improve QoS by following specified policy rules. A Massively Multiplayer Online Game (MMOG) has been created to validate the functionality and performance of MMOP. The simulation results have demonstrated that MMOP is a high performance and scalable servicing and computing middleware.



Sign in / Sign up

Export Citation Format

Share Document