Jouni Mäenpää

Study on maintenance operations in a chord-based Peer-to-Peer session initiation protocol overlay network

Peer-to-Peer Session Initiation Protocol (P2PSIP) is a new technology being standardized in the Internet Engineering Task Force. A P2PSIP network consists of a collection of nodes organized in a peer-to-peer fashion for the purpose of enabling real-time communication using the Session Initiation Protocol (SIP). In this paper, we present experimental results obtained by running a P2PSIP prototype in PlanetLab. Our prototype uses the Chord Distributed Hash Table (DHT) to organize the P2PSIP overlay and Peer-to-Peer Protocol (P2PP) as the protocol spoken between the peers. In the experiments, the performance of the system is studied under different churn rates and using different DHT maintenance intervals.

Using RELOAD and CoAP for wide area sensor and actuator networking

In this article, we propose a new architecture for wide area sensor and actuator networking. The architecture is based on combining two protocols being standardized by the Internet Engineering Task Force, REsource LOcation And Discovery (RELOAD) and Constrained Application Protocol (CoAP). To integrate CoAP and RELOAD, we introduce a CoAP application usage for RELOAD. The architecture provides a decentralized peer-to-peer rendezvous service for CoAP nodes in Wireless Sensor Networks (WSNs). Our architecture also enables a peer-to-peer federation of geographically distributed WSNs. This is supported by proxy nodes that are part of the WSN but also connect to a RELOAD overlay network via cellular Internet access. Due to the use of RELOAD, the system does not need to rely on centralized services such as DNS service discovery (DNS-SD) or central resource directories to discover sensors and resources. Other features of the architecture include integration to web, self-organization, scalability, and robustness. We evaluate the proposed architecture through simulations and real-life measurements, and compare its performance to a traditional client/server architecture.

Impact of Network Address Translator Traversal on Delays in Peer-to-Peer Session Initiation Protocol

Peer-to-Peer Session Initiation Protocol (P2PSIP) is a distributed communication system being standardized in the Internet Engineering Task Force (IETF). Since it uses the peer-to-peer paradigm, P2PSIP faces the problems created by Network Address Translators (NATs); even peers located behind NATs need to be able to not only initiate connections to other peers but also accept connections initiated by other peers. In this paper, we study the impact of standardized NAT traversal solutions, namely Session Traversal Utilities for NAT (STUN), Traversal Using Relays around NAT (TURN), and Interactive Connectivity Establishment (ICE), on delays in P2PSIP overlay networks. These delays are studied from the viewpoint of wireless and wired nodes acting as clients in a P2PSIP overlay network running in the PlanetLab. The delays are also compared to those of the traditional client/server Session Initiation Protocol (SIP).

Analysis of Delays in a Peer-to-Peer Session Initiation Protocol Overlay Network

Peer-to-Peer Session Initiation Protocol (P2PJSIP) is a new decentralized person-to-person communication system that is currently being standardized in the Internet Engineering Task Force (IETF). P2PSIP uses the Session Initiation Protocol (SIP) to enable real-time communication in a peer-to-peer environment. The underlying lookup mechanism is implemented using a Distributed Hash Table (DHT). In this paper, we study delays associated with joining, leaving, and initiating calls in a P2PSIP system through experiments in PlanetLab. We also compare the performance of P2PSIP and traditional client-server SIP.

Performance of REsource LOcation and Discovery (RELOAD) on Mobile Phones

REsource LOcation And Discovery (RELOAD) is a peer-to-peer signaling protocol that can be used to maintain an overlay network, and to store data in and retrieve data from the overlay. RELOAD is currently being standardized in the Internet Engineering Task Force (IETF). The main application using RELOAD is Peer-to-Peer Session Initiation Protocol (P2PSIP). In this paper, we study the performance of RELOAD on a mobile phone participating in a P2PSIP overlay. We focus on memory consumption, CPU load, battery consumption, and bandwidth usage. The goal is to find out whether mobile phones can act as full peers in a P2PSIP overlay.

Estimating operating conditions in a Peer-to-Peer Session Initiation Protocol overlay network

Distributed Hash Table (DHT) based peer-to-peer overlays are decentralized, scalable, and fault tolerant. However, due to their decentralized nature, it is very hard to know the state and prevailing operating conditions of a running overlay. If the system could figure out the operating conditions, it would be easier to monitor the system and re-configure it in response to changing conditions. Many DHT-based system such as the Peer-to-Peer Session Initiation Protocol (P2PSIP) would benefit from the ability to accurately estimate the prevailing operating conditions of the overlay. In this paper, we evaluate mechanisms that can be used to do this. We focus on network size, join rate, and leave rate. We start from existing mechanisms and show that their accuracy is not sufficient. Next, we show how the mechanisms can be improved to achieve a higher level of accuracy. The improvements we study include various mechanisms improving the accuracy of leave rate estimation, use of a secondary network size estimate, sharing of estimates between peers, and statistical mechanisms to process shared estimates.

Reducing delays related to NAT traversal in P2PSIP session establishments

This paper focuses on reducing the Network Address Translator (NAT) traversal-related components of the session establishment delay in peer-to-peer Session Initiation Protocol (P2PSIP) overlays. To reduce the delay, we propose to group the management of different connections so that the (time-consuming) NAT traversal procedures performed for one connection can be reused when establishing other connections. To do this, we propose to use the Host Identity Protocol (HIP) to perform connection management in P2PSIP overlays. In order to evaluate the performance gains resulting from this approach, we have implemented a P2PSIP system whose modular design allows us to build overlay networks with and without HIP and measure their differences in performance. Our experiments show that grouping the management of different connections results in a significant reduction in the session establishment delay in the presence of NATs.

Load balancing for structured P2P networks using the advanced finger selection algorithm (AFSA)

Structured Peer-to-Peer (P2P) networks, such as networks based on Distributed Hash Tables (DHTs), can be enhanced by using load balancing mechanisms. Current load balancing mechanisms are either trying to achieve even distribution of objects among nodes, make the address space as evenly populated as possible, or both. However, we have taken a different approach to load balancing in this paper and we have defined the Advanced Finger Selection Algorithm (AFSA) which is not focused on balancing the objects among the nodes and does not require evenly populated address space. AFSA is an algorithm which changes the way how nodes are selected as fingers to the overlay routing tables in structured P2P networks. We implemented the AFSA algorithm for both Chord and Bamboo and we evaluated it with simulations.

Performance evaluation of Recursive Distributed Rendezvous based service discovery for Peer-to-Peer Session Initiation Protocol

Recursive Distributed Rendezvous (ReDiR) is a service discovery mechanism for Distributed Hash Table (DHT) based Peer-to-Peer (P2P) overlay networks. One of the major P2P systems that has adopted ReDiR is Peer-to-Peer Session Initiation Protocol (P2PSIP), which is a distributed communication system being standardized in the P2PSIP working group of the Internet Engineering Task Force (IETF). In a P2PSIP overlay, ReDiR can be used for instance to discover Traversal Using Relays around NAT (TURN) relay servers needed by P2PSIP nodes located behind a Network Address Translator (NAT). In this paper, we study the performance of ReDiR in a P2PSIP overlay network. We focus on metrics such as service lookup and registration delays, failure rate, traffic load, and ReDiRs ability to balance load between service providers and between nodes storing information about service providers.

Reducing P2PSIP session setup delays

It has been shown in previous work that session setup delays in a Peer-to-Peer Session Initiation Protocol (P2PSIP) based Voice over IP (VoIP) system can be unacceptably high. This is due to the need to route several messages across the P2P overlay and also due to the fact that P2PSIP needs to establish separate connections for SIP signaling and Real-Time Protocol (RTP) media between the communicating parties across Network Address Translators (NATs). In this paper, we will develop optimizations to reduce P2PSIP session setup delays. We will also analyze the performance of these optimizations. Using the proposed optimizations, it is possible to achieve delays that are up to three times lower than those of unoptimized P2PSIP.