Emanuele Carlini

Cloud federations in contrail

Cloud computing infrastructures support dynamical and flexible access to computational, network and storage resources. To date, several disjoint industrial and academic technologies provide infrastructure level access to Clouds. Especially for industrial platforms, the evolution of de-facto standards goes together with worries about user lock-in to a platform. The Contrail project [6] proposes a federated and integrated approach to Clouds. In this work we present and motivate the architecture of Contrail federations. Contrails goal is to minimize the burden on the user and increase the efficiency in using Cloud platforms by performing both a vertical and a horizontal integration. To this end, Contrail federations play a key role, allowing users to exploit resources belonging to different cloud providers, regardless of the kind of technology of the providers and with a homogeneous, secure interface. Vertical integration is achieved by developing both the Infrastructure- and the Platform-as-a-Service levels within the project. A third key point is the adoption of a fully open-source approach toward technology and standards. Beside supporting user authentication and applications deployment, Contrail federations aim at providing extended SLA management functionalities, by integrating the SLA management approach of SLA@SOI project in the federation architecture.

QBROKAGE: A Genetic Approach for QoS Cloud Brokering

The broad diffusion of Cloud Computing has fostered the proliferation of a large number of cloud computing providers. The need of Cloud Brokers arises for helping consumers in discovering, considering and comparing services with different capabilities and offered by different providers. Also, consuming services exposed by different providers, when possible, may alleviate the vendor lock-in. While it can be straightforward to choose the best provider when deploying small and homogeneous applications, things get harder if the size and complexity of applications grow up. In this paper we propose a genetic approach for Cloud Brokering, focusing on finding Infrastructure-as-a-Service (IaaS) resources for satisfying Quality of Service (QoS) requirements of applications. We performed a set of experiments with an implementation of such broker. Results show that our broker can find near-optimal solutions even when dealing with hundreds of providers, trying at the same time to mitigate the vendor lock-in.

Integration of P2P and Clouds to support Massively Multiuser Virtual Environments

Massively Multiuser Virtual Environments (MMVEs) are attracting millions of players from all over the world. Currently used Client/Server infrastructures and technologies are reaching their limits of flexibility and scalability. We propose an approach that combines the technological advantages of two different paradigms, namely P2P networking and Cloud Computing. Our proposal leverages known P2P techniques like Virtual Nodes and consistent hashing, as well as separate overlays for different purposes, e.g. interest and object management. We propose combining these techniques with the definition of a specific role in the overlay for Cloud-based, trusted resources. This enables the distribution of the MMVE on top of a mix of Cloud and user resources. The solution outlined allows building key-value distributed storage systems that can resize at run-time (elasticity) and provide scalability and load balancing features to the MMVE platform.

Service and Resource Discovery supports over P2P overlays

We describe the main architecture and the design principles of the Service/Resource Discovery System (SRDS), a component of the XtreemOS Operating System. XtreemOS is a Linux extension that enables management and exploitation as single platform of computational resources provided by federated Virtual Organizations. The SRDS provides scalable and fault-tolerant directory services supporting many of the platform functionalities, ranging from dynamic resource location and job control to system and application-oriented directory services. The key challenge of the SRDS design is to provide the common metaphor of the directory service, meeting the scalability requirements of a Grid-aware Operating system, and at the same time enjoy extendability and configurability, especially with respect to the quality of service provided. The SRDS design combines different peer to peer structured overlay networks, exploiting their peculiar strengths. We describe the implementation and our design of the namespace abstraction as implemented on top of multiple overlay networks. Finally, we show test results of the SRDS on top of a subset of the Grid5000 platform.

Distributed Virtual Environments: From client server to cloud and P2P architectures

Distributed Virtual Environments (DVEs), like Massively Multiplayer Online Games (MMOG) are attracting millions of users from all over the world. However, as the number of simultaneous users keeps growing, the current used Client/Server infrastructure technologies are showing their limits on the side of the practicality and applicability. As a consequence, novel solutions based on Cloud and P2P architectures have been recently proposed. While Cloud Architectures permit to overcome the over provisioning problem of C/S architectures, the advent of home xDSL connection capability permits to explore the capability of P2P approaches to design the infrastructure of next-generation DVEs. The use of P2P distributed infrastructures permits to exploit the inherent scalability of DVE, their robustness and self-organizing properties. However, this solution has to face with additional related challenges such as consistency, persistence and limited hardware capabilities of the peers. This paper discusses the main research challenges and reviews the main DVE architectures proposed in the last years.

Cracker: Crumbling large graphs into connected components

The problem of finding connected components in a graph is common to several applications dealing with graph analytics, such as social network analysis, web graph mining and image processing. The exponentially growing size of graphs requires the definition of appropriated computational models and algorithms for their processing on high throughput distributed architectures. In this paper we present cracker, an efficient iterative algorithm to detect connected components in large graphs. The strategy of cracker is to iteratively grow a spanning tree for each connected component of the graph. Nodes added to such trees are discarded from the computation in the subsequent iterations. We provide an extensive experimental evaluation considering a wide variety of synthetic and real-world graphs. The experimental evaluation shows that cracker consistently outperforms state-of-the-art approaches both in terms of total computation time and volume of messages exchanged.

Balanced Graph Partitioning with Apache Spark

A significant part of the data produced every day by online services is structured as a graph. Therefore, there is the need for efficient processing and analysis solutions for large scale graphs. Among the others, the balanced graph partitioning is a well known NP-complete problem with a wide range of applications. Several solutions have been proposed so far, however most of the existing state-of-the-art algorithms are not directly applicable in very large-scale distributed scenarios. A recently proposed promising alternative exploits a vertex-center heuristics to solve the balance graph partitioning problem. Their algorithm is massively parallel: there is no central coordination, and each node is processed independently. Unfortunately, we found such algorithm to be not directly exploitable in current BSP-like distributed programming frameworks. In this paper we present the adaptations we applied to the original algorithm while implementing it on Spark, a state-of-the-art distributed framework for data processing.

Flexible load distribution for hybrid distributed virtual environments

This paper proposes an architecture for Distributed Virtual Environments (DVEs) integrating cloud and peer nodes. We define the overall structure of the architecture and propose a flexible strategy to distribute the load due to the management of the entities of the DVE. The proposed approach takes into account the utilisation of the nodes, the economical cost due to their use of bandwidth, and their probability of failure. A greedy heuristics is exploited to reduce the computational cost of the algorithm in order to maintain a fair interactivity level to the end user of the DVE. A mobility model generating realistic Second Life traces is exploited to evaluate our algorithm. The experimental results show the effectiveness of our approach.

Reducing Server Load in MMOG via P2P Gossip

A fundamental problem for the development of Massively Multiplayer Online Games (MMOGs) is the definition of a mechanism supporting interest management, i.e. determining all the entities of the virtual world that are relevant for a given peer. To this end, we propose a hybrid architecture combining centralized and P2P solutions. Our proposal exploits a P2P gossip-based approach supporting a best-effort resolution of interest management, so that it can be mostly supported through the P2P overlay, with minimal intervention of a centralized entity.

Smart cloud federation simulations with CloudSim

In the last few years the broad diffusion of Cloud Computing has encouraged the proliferation of cloud computing providers. However, providers often propose their services using proprietary management software, interfaces and virtualization technologies. This strongly hinders the applications interoperability and migration across providers boundaries. Organizing providers in federations seems promising for addressing such issues, but it introduces other challenges to be faced, often requiring innovative approaches. Unfortunately, the evaluation of new solutions in a repeatable manner and under several configurations is a hard task to achieve using real Clouds platforms. For these reasons we propose SmartFed, a simulator for cloud federations that is able to model the richness typical of an environment with multiple cloud providers. We show the capability of SmartFed by simulating a sample mapping process for assigning applications to providers.