Exploring local service allocation in Community Networks Davide Vega, Roc Meseguer Computer Architecture Department Universitat Politècnica de Catalunya - BarcelonaTech C/ Jordi Girona, 1-3 08034 Barcelona, Spain Email: {dvega, meseguer}@ac.upc.edu Guillem Cabrera, Joan M. Marquès Internet Interdisciplinary Institute Universitat Oberta de Catalunya C/ Roc Boronat, 117 7th floor 08018 Barcelona, Spain Email: {gcabreraa, jmarquesp}@uoc.edu Abstract—Community Cloud computing is a new trend on cloud computing that aims to build service infrastructures upon Wireless Community Networks taking advantage of underused community physical resources. Service allocation protocols are a key design challenge that all cloud systems must properly address to optimize resource utilization. They are specially important when cloud services requires a Quality of Service (QoS) and network stability or performance (delay, jitter, minimum bandwidth) cannot be guaranteed a-priory. This work presents a study that tries to understand how to address cloud service deployments in such scenario. In particular, we start proposing an allocation algorithm to find optimal solutions when there is a central authority that coordinates the process. These solutions optimize the communication cost in two ways: (1) minimizing the service overlay diameter and, (2) minimizing the coordination cost along the network. Based on the study of the algorithm and the experimental simulations, we study the variables that outcome optimal service allocations in detriment to other solutions. We verify these findings using data mining techniques. Researchers can take advantage of the simulation results and our observations to design more reliable distributed algorithms able to dynamically self-adapt to network changes. Keywords—Service allocation, community cloud, wireless net- work I. I NTRODUCTION AND MOTIVATION Wireless Community Networks (WCN) showed up as a cutting-edge model of open communication infrastructure, as they offer low-cost but participatory connectivity to cit- izens. These novel infrastructures are constructed, operated, maintained and owned by the citizens themselves and bring closer information technologies to underdeveloped countries or isolated areas often out of the plans of traditional telecommu- nication operators. Many of them succeeded in the recent years across the world as non-profit actions, thanks to the progress of ad-hoc wireless technologies. Guifi.net [1], Athens Wireless Metropolitan Network [2], FunkFeuer [3], Seattle Wireless [4] and Consume [5] are the most fruitful exponents of community Internet access providers. Through their openess, neutrality and the effort of their own members, Guifi.net [1][1] has become the world’s bigger independent wireless community network. Mainly deployed in Catalonia (Spain), it has more than 17,000 operative nodes and up to 30,000 km of links. It uses a growing pattern mostly based on planed radio-link deployment adapted to each region. As a result, the network topology contains many recognizable topology patterns also present in other community networks. Beyond participants’ initiatives to improve the Internet access or increase the network quality and security, most community members also devote efforts to build and deploy distributed and scalable applications. This useful applications (e.g. web or FTP servers, monitoring systems) are typically provided to other community members as an open-access service, often without any other form of recognition than pro- moting the community network usage. As a natural evolution of this model, some communities started to provide mecha- nisms that regulate and normalize how their members deploy services or contribute with their computational resources. In [6] authors adopted the term Community Cloud Computing to describe such contributory systems that provide community- based services upon Wireless Community Networks. Several initiatives can be found in the literature of community-based clouds, like Cloud@Home [7], Nebulas [8], Contributory Communities [9] or Clommunity [10]. They all set the trend on the aggregation of user-donated computing resources in such a way they can be regarded as a platform to provide long-lived services, as happens in traditional cloud computing. A common particularity of these scenarios is that cloud services coordination and performance computing are highly dependent on the diversity caused by the networking charac- teristics of their host being in a community network. Some examples of such diversity are: (a) network deployment based on geographic singularities rather than network QoS, (b) pow- erless and heterogeneous network – and computation – equip- ment or (c) wireless radio links with asymmetric quality of services. This paper aims to explore the consequences of a local service allocation in a particular wireless community cloud, called Clommunity, such that minimizes the coordination cost of such services. In our proposal, service locality is defined as a service deployment which guarantees a solution overlay with minimum diameter. A typical service with this requirements is a redundant information backup. The coordination cost, which in Clommunity is associated to the use of nodes from different administrative zones, has been also minimized looking for placement solutions that traverses a minimum number of zones. In the previous example, is has also sense for the clients