Energy Savings on a Cloud-based Opportunistic Infrastructure

in Service Oriented Computing ICSOC 2013 Workshops (2014)

In this paper, we address energy savings on a Cloud-based opportunistic infrastructure. The infrastructure implements opportunis- tic design concepts to provide basic services, such as virtual CPUs, RAM ... [more ▼]

In this paper, we address energy savings on a Cloud-based opportunistic infrastructure. The infrastructure implements opportunis- tic design concepts to provide basic services, such as virtual CPUs, RAM and Disk while profiting from unused capabilities of desktop computer laboratories in a non-intrusive way. We consider the problem of virtual machines consolidation on the oppor- tunistic cloud computing resources. We investigate four workload packing algorithms that place a set of virtual machines on the least number of physical machines to increase resource utilization and to transition parts of the unused resources into a lower power states or switching off. We em- pirically evaluate these heuristics on real workload traces collected from our experimental opportunistic cloud, called UnaCloud. The final aim is to implement the best strategy on UnaCoud. The results show that a consolidation algorithm implementing a policy taking into account fea- tures and constraints of the opportunistic cloud saves energy more than 40% than related consolidation heuristics, over the percentage earned by the opportunistic environment. [less ▲]

Performance Evaluation of an IaaS Opportunistic Cloud Computing

Poster (2014)

This poster shows the performance evaluation of UnaCloud Opportunistic Computing IaaS. We analyze from an HPC perspective, two virtualization frameworks Virtual Box and VMware ESXi and compare them over ... [more ▼]

This poster shows the performance evaluation of UnaCloud Opportunistic Computing IaaS. We analyze from an HPC perspective, two virtualization frameworks Virtual Box and VMware ESXi and compare them over this particular opportunistic cloud environment. The benchmarks consist of two set of tests, High Performance Linpack and IOzone, that examine the performance and the Input/Output response. The purpose of the experiments is to evaluate the behavior of the different virtual environments over an opportunistic cloud environment and investigate how these are affected by different percentage of end-users. The results show a better performance for Virtual Box than VMware and the other way around for I/O response. Nevertheless, the experiments shows that VBox have more robustness than VMware. [less ▲]

Evolutionary Multi-Objective Optimisation with Quantile Constraint for the Protein Structure Similarity Problem

Scientific Conference (2014)

Internet shopping with price sensitive discounts

in 4OR: a quarterly journal of operations research (2014), 12(1), 35-48

A customer would like to buy a given set of products in a given set of Internet shops. For each Internet shop, standard prices for the products are known as well as a concave increasing discounting ... [more ▼]

A customer would like to buy a given set of products in a given set of Internet shops. For each Internet shop, standard prices for the products are known as well as a concave increasing discounting function of total standard and delivery price. The problem is to buy all the required products at the minimum total discounted price. Computational complexity of various special cases is established. Properties of optimal solutions are proved and polynomial time and exponential time solution algorithms based on these properties are designed. Two heuristic algorithms are suggested and computationally tested. [less ▲]

Solving the three dimensional quadratic assignment problem on a computational grid

in Cluster Computing (2014), 17(2), 205-217

The exact resolution of large instances of combinatorial optimization problems, such as three dimensional quadratic assignment problem (Q3AP), is a real challenge for grid computing. Indeed, it is ... [more ▼]

The exact resolution of large instances of combinatorial optimization problems, such as three dimensional quadratic assignment problem (Q3AP), is a real challenge for grid computing. Indeed, it is necessary to reconsider the resolution algorithms and take into account the characteristics of such environments, especially large scale and dynamic availability of resources, and their multi-domain administration. In this paper, we revisit the design and implementation of the branch and bound algorithm for solving large combinatorial optimization problems such as Q3AP on the computational grids. Such gridification is based on new ways to effi- ciently deal with some crucial issues, mainly dynamic adaptive load balancing and fault tolerance. Our new approach allowed the exact resolution on a nation-wide grid of a dif- ficult Q3AP instance. To solve this instance, an average of 1,123 computing cores were used for less than 12 days with a peak of around 3,427 computing cores. [less ▲]

NC-CELL: Network Coding-based Content Distribution in Cellular Networks for Cloud Applications

in IEEE Global Communications Conference, Austin, TX, USA, 2014 (2014)

The popularity of cloud applications surged in the last years. Billions of mobile devices remain always connected. Location services, online games, social networking and navigation are a just few examples ... [more ▼]

The popularity of cloud applications surged in the last years. Billions of mobile devices remain always connected. Location services, online games, social networking and navigation are a just few examples of “always on” cloud applications in which the same or partially overlapping content is delivered to multiple users. In this paper, we propose a technique, called NCCELL, which uses network coding to foster content distribution in mobile cellular networks. Specifically, NC-CELL implements a software module at mobile base stations (or eNodeBs) which cans in transit traffic and looks for opportunities to code packets destined to different mobile users together. The proposed approach can significantly improve cell throughput and is particularly relevant for delay tolerant content distribution. [less ▲]

Adaptive Energy Efficient Distributed VoIP Load Balancing in Federated Cloud Infrastructure

in IEEE International Conference on Cloud Networking (CLOUDNET), Luxembourg City, 2014. (2014)

Design of an Energy Efficiency Model and Architecture for Cloud Management using Prediction Models

Scientific Conference (2013, December 18)

In this paper, we present a new energy efficiency model and architecture for cloud management based on a prediction model with Gaussian Mixture Models. The methodology relies on a distributed agent model ... [more ▼]

In this paper, we present a new energy efficiency model and architecture for cloud management based on a prediction model with Gaussian Mixture Models. The methodology relies on a distributed agent model and the validation will be performed on OpenStack. This paper intends to be a position paper, the implementation and experimental run will be conducted in future work. The design concept leverages the prediction model by providing a full architecture binding the resource demands, the predictions and the actual cloud environment (Openstack). The prediction analysis feeds the power-aware agents that run on the compute nodes in order to turn the nodes into sleep mode when the load state is low to reduce the energy consumption of the data center. [less ▲]

Comparing the Performance and Power Usage of GPU and ARM Clusters for Map-Reduce

in Proc. of the 3rd Intl. Conf. on Cloud and Green Computing (CGC'13) (2013, October)

Hopfield neural network for simultaneous job scheduling and data replication in grids

in Future Generation Computer Systems (2013), 29(8), 1885-1900

This paper presents a novel heuristic approach, named JDS-HNN, to simultaneously schedule jobs and replicate data files to different entities of a grid system so that the overall makespan of executing all ... [more ▼]

This paper presents a novel heuristic approach, named JDS-HNN, to simultaneously schedule jobs and replicate data files to different entities of a grid system so that the overall makespan of executing all jobs as well as the overall delivery time of all data files to their dependent jobs is concurrently minimized. JDS-HNN is inspired by a natural distribution of a variety of stones among different jars and utilizes a Hopfield Neural Network in one of its optimization stages to achieve its goals. The performance of JDS-HNN has been measured by using several benchmarks varying from medium- to very-large-sized systems. JDS-HNN’s results are compared against the performance of other algorithms to show its superiority under different working conditions. These results also provide invaluable insights into scheduling and replicating dependent jobs and data files as well as their performance related issues for various grid environments. [less ▲]