14-12-2012, 12:52 PM
Locaqting Equivalent Servants Over P2P Networks
locaqting equivalent servants.doc (Size: 2.11 MB / Downloads: 36)
ABSTRACT
This project focuses on services provided by equivalent servants and models and analyzes the performance of structured and unstructured overlays when used to provide such services. I demonstrate that the architecture chosen for the P2P network has a huge impact on the overall performance of the service. In particular, with the support of some analytical and simulation results, I show how an unstructured network based on epidemic dissemination and built over a scale-free overlay topology is an effective solution to deploy in this context. Then, I present EQUATOR (EQUivalent servAnt locaTOR), a P2P-based architecture deployable in real networks for the provision of services based on equivalent servants. EQUATOR aims at guaranteeing high lookup performance, as well as high robustness to failures and churn phases, when a significant number of peers joins/leaves the network.
Existing works lack in providing adequate support to these emerging distributed systems. In fact, most of them focus on the development of a system supporting specific requests, ranging from a unique specific file to a set of resources characterized by well-defined parameters. While these systems can also support the localization of equivalent servants, they are not optimized for this purpose because of the different requirements they comply with, more stringent in terms of resource constraints, but simpler in terms of timely response. Hence, for example, they might be unable to locate a serving node in a very short time, such as a relay to be used in an incoming VoIP call. Furthermore, they may insert an unnecessary overhead in the servant lookup, due to the features they provide to support complex queries, which are of little help in the context of services based on equivalent servants.
SCOPE OF THE PROJECT
Peer-to-Peer applications, Grid systems, large-scale distributed systems, and overlay networks are open for consideration. We would like to give special encouragement to experience reports with applications, with particular emphasis to those that treat problems related to scalability, robustness, and security.
Topics of interest for the conference include, but are not limited to:
• P2P applications and services
• Performance and robustness of P2P systems
• Self-organization in P2P systems
• P2P information retrieval
• Higher-level query support in P2P systems
• Semantic overlay networks and semantic query routing in P2P systems
• Trust and reputation management in P2P systems
• Overlay architectures and topologies
• P2P overlay interworking with underlying infrastructure
• Overlay monitoring and management
• P2P systems over MANETs
• Delay-tolerant P2P systems
INTRODUCTION
Introduction This project focuses on services provided by equivalent servants and models and analyzes the performance of structured and unstructured overlays when used to provide such services. I demonstrate that the architecture chosen for the P2P network has a huge impact on the overall performance of the service. In particular, with the support of some analytical and simulation results, I show how an unstructured network based on epidemic dissemination and built over a scale-free overlay topology is an effective solution to deploy in this context. Then, I present EQUATOR (EQUivalent servAnt locaTOR), a P2P-based architecture deployable in real networks for the provision of services based on equivalent servants. EQUATOR aims at guaranteeing high lookup performance, as well as high robustness to failures and churn phases, when a significant number of peers joins/leaves the network.
Existing works lack in providing adequate support to these emerging distributed systems. In fact, most of them focus on the development of a system supporting specific requests, ranging from a unique specific file to a set of resources characterized by well-defined parameters. While these systems can also support the localization of equivalent servants, they are not optimized for this purpose because of the different requirements they comply with, more stringent in terms of resource constraints, but simpler in terms of timely response. Hence, for example, they might be unable to locate a serving node in a very short time, such as a relay to be used in an incoming VoIP call. Furthermore, they may insert an unnecessary overhead in the servant lookup, due to the features they provide to support complex queries, which are of little help in the context of services based on equivalent servants.
State Of Problem
While peer-to-peer networks are mainly used to locate unique resources across the Internet, new interesting deployment scenarios are emerging. Particularly, some applications (e.g., VoIP) are proposing the creation of overlays for the localization of services based on equivalent servants (e.g., voice relays). Existing works lack in providing adequate support to these emerging distributed systems In fact, most of them focus on the development of a system supporting specific requests, previous work explores the idea of a service based on it limit equivalent servants.
STUDY OF THE SYSTEM
Existing System
Existing works lack in providing adequate support to these emerging distributed systems. In fact, most of them focus on the development of a system supporting specific requests, ranging from a unique specific file to a set of resources characterized by well-defined parameters. While these systems can also support the localization of equivalent servants, they are not optimized for this purpose because of the different requirements they comply with, more stringent in terms of resource constraints, but simpler in terms of timely response. Hence, for example, they might be unable to locate a serving node in a very short time, such as a relay to be used in an incoming VoIP call. Furthermore, they may insert an unnecessary overhead in the servant lookup, due to the features they provide to support complex queries, which are of little help in the context of services based on equivalent servants.
Disadvantages
• Existing cant implemented feature of scale-free networks
• Existing cant implemented feature Network size grows
Proposed System
This paper focuses on services provided by equivalent servants and models and analyzes the performance of structured and unstructured overlays when used to provide such services. We demonstrate that the architecture chosen for the P2P network has a huge impact on the overall performance of the service. In particular, with the support of some analytical and simulation results, we show how an unstructured network based on epidemic dissemination and built over a scale-free overlay topology is an effective solution to deploy in this context. Then, we present EQUATOR (EQUivalent servAnt locaTOR), a P2P-based architecture deployable in real networks for the provision of services based on equivalent servants. EQUATOR aims at guaranteeing high lookup performance, as well as high robustness to failures and churn phases, when a significant number of peers joins/leaves the network.
DHT equivalent servants
DHT (Dynamic Hash Table) introduce an additional feature to this querying mechanism: during the lookup process, any node encountered along the path is checked for availability and can be selected as a servant for the querying user. Notice that this operating mode makes the approach independent of the adopted DHT. In fact, only the overlay topology (which is a regular graph in existing DHTs) is of interest in our context. In other words, we adopt the topology of a generic DHT, with a fixed number of neighbors for each node, but we use a different routing mechanism. The idea of using a DHT for our scenario of equivalent servants is especially interesting in case a DHT has to be implemented anyway for some other services. For example, P2PSIP already uses a structured overlay to index all possible targets of a multimedia communication, i.e., all the user agents registered in the SIP domain. DHT Equator Simulation equivalent servants
In EQUATOR, we prefer a more flexible approach that relies on multiple
Bootstrap reachable through appropriate DNS records thus guaranteeing redundancy and load balancing. Bootstrap servers globally store information about m0 participating peers; when a peer joins the overlay. it adds we proposed the Equivalent servant locator (EQUATOR) architecture, which overcomes the issues related to the deployment of a scale-free topology for service location in a real network, mainly due to the static nature of the ideal scale-free construction algorithm and the lack of a global knowledge of the participating peers. Simulation results confirmed the effectiveness of EQUATOR, showing how it offers good lookup performance in conjunction with low message overhead and high resiliency to node churn and failures.