Peer-to-Peer (P2P) networks are one way to create large-scale distributed systems. A single peer has only a limited view on other peers. Thus, efficient searching for other peers or their content is a key performance indicator. In this paper, we investigate the search efficiency in an m-ary tree-structured P2P overlay. While previous work aimed for balancing the maximum height of a node's sub-trees, we show that keeping the height balanced throughout the overall network -a property called null-balance -will increase search performance considerably. Simulations using the ns-3 discrete-event simulator show 50% better performance w.r.t. required routing hops in these nullbalanced trees. Therefore, we develop algorithms that keep a tree null-balanced if a node joins or departures. I.e., we prevent the need for restructuring. As we show, the cost of our efficient structure-preserving algorithms is easily set off by a relatively small number of search operations.

No abstract

The digitalization of the world proceeds continually. As part of this, we are facing a smart digital future where everything is connected and linked while we are transforming big data into knowledge. The FIWARE platform, which is the result of a series of EU funded projects, supports this transformation. Developers, integrators and end users are able to rapidly create new applications, by using components provided by this platform. We have analyzed central building blocks of the FIWARE platform from different aspects and -based on experiments and code reviews -like the need for a multi-level role-based security or a caching mechanism when it comes to querying data. Within this paper we also provide further recommendations for the current code base that can help to improve those FIWARE components under investigation. In addition to that we also suggest how to simplify a typical setup by extending FIWARE's core component using a Websocket-Port.

No abstract

Peer-to-Peer (P2P) network consists of a large number of nodes, where each node may have different capabilities and properties. Finding peers with specific capabilities and properties is challenging. Thus, we propose a practical solution to the problem of peer discovery, which is finding peers in the network according to a specified query. We contribute a peer discovery for an m-ary tree-structured P2P network by utilizing a connected dominating set (CDS), a technique that is typically used in unstructured networks. Our approach of constructing the CDS requires no additional communication cost, while nodes can insert, update and remove data within O(1). Each node of the CDS -a dominating set node -maintains only a limited number of nodes. We confirm the properties of our proposed solution by using the ns-3 discrete-event simulator. This includes, besides the degree of decentralism of the peer discovery, also the heterogeneity of peers.