Patrick Poullie

A Survey of the State-of-the-Art in Fair Multi-Resource Allocations for Data Centers

Multi-resource allocation in data centers determines a network and service management task of crucial importance. While, traditionally computing systems are shared based on a single resource, it was shown that in data centers this simplification often impedes overall efficiency and fairness. Therefore, newer approaches consider data center resource allocations as a multi-resource allocation problem. However, the capability of these approaches to reach overall fairness or efficiency is limited due to theoretical assumptions they make or due to practical means they deploy to control resources. This survey: 1) details all steps necessary to allocate data center resources and puts these steps in relation to each other; 2) highly relevant concepts in support of fair data center resource allocations, such as utility functions and allocation characteristics, are discussed and compared; and 3) in turn, major approaches to allocate multiple data center resources in a fair manner are outlined, mapped to practical steps and economically driven-targets, and compared with respect to their suitability of being applied in today’s data centers.

The Design and Evaluation of a Heaviness Metric for Cloud Fairness and Correct Virtual Machine Configurations

Fairness problems in data centers have been pointed out frequently over the last years. To enforce fairness in data centers, the application of job/Virtual Machine (VM) scheduling impels the traditional solution. Scheduling determines the order in which VMs/jobs are started. However, it is insufficient to enforce fairness, when jobs/VMs run over long periods and/or their PR utilization is highly fluctuant. Clouds form a special case of data centers in which this can be observed.

Traffic Management for Cloud Federation

The chapter summarizes activities of COST IC1304 ACROSS European Project corresponding to traffic management for Cloud Federation (CF). In particular, we provide a survey of CF architectures and standardization activities. We present comprehensive multi-level model for traffic management in CF that consists of five levels: Level 5 - Strategies for building CF, Level 4 - Network for CF, Level 3 - Service specification and provision, Level 2 - Service composition and orchestration, and Level 1 - Task service in cloud resources. For each level we propose specific methods and algorithms. The effectiveness of these solutions were verified by simulation and analytical methods. Finally, we also describe specialized simulator for testing CF solution in IoT environment.

Virtual machine priority adaption to enforce fairness among cloud users

In recent years fairness problems in data centers have been pointed out and job/Virtual Machine (VM) scheduling has been chosen as a solution approach. Clouds are a special case of data centers, where resources are deployed by VMs in a highly dynamic manner during VM runtime. However, scheduling only allows influencing resource allocations, when VMs are instantiated, i.e., before runtime. Thus, runtime prioritization bears a great potential to manage cloud resources and promote fairness in clouds, especially, when VMs run over long periods. Nevertheless, runtime prioritization is not leveraged accordingly. This paper defines fairness as handicapping VMs of heavy users during runtime to allocate more resources to VMs of light users. Thereby, the need to make assumptions on users utility functions is avoided, while different fairness notions can be captured by adapting the definition of heaviness. Guidelines for this definition are provided to ensure incentives to configure and utilize VMs adequately. Finally, OpenStack is extended in its implementation by a decentralized fairness service to enforce fairness according to this definition. The fairness services functionality is certified by experiments in terms of overhead and fairness promotion.

Fair allocation of multiple resources using a non-monetary allocation mechanism

The fair allocation of scarce resources is relevant to a wide field of applications. For example, cloud resources, such as CPU, RAM, disk space, and bandwidth, have to be shared. This paper presents a mechanism to find fair allocations of multiple divisible resources, which, contrary to other mechanisms, is applicable to but not limited to the example above. Wide applicability of the mechanism is achieved by designing it (1) to scale with the number of consumers and resources, (2) to allow for arbitrary preference functions of consumers, and (3) to not rely on monetary compensation. The mechanism uses a mathematical definition of greediness to balance resources consumers receive and thereby to compute a fair allocation.

Attacks on internet names

The Domain Name System (DNS) determines the major component in todays Internet, as it maps memorable names, such as www.uzh.ch into routable Internet Protocol addresses, such as 136.105.200.244. Since the early days of trusted hosts in the Internet have passed, the potential of severe attacks on DNS has reached a level of higher risk, e.g. DNS Spoofing or Cache Poisoning, such that work on DNS Security Extensions (DNSSEC) did commence. However, the deployment of DNSSEC has not reached that large attention needed to safeguard fully future Internet communications for all services.