/pdf/a-break-in-the-clouds-towards-a-cloud-definition-svovsw1r5a.pdf

A Break in the Clouds: Towards a Cloud Definition

Data centers are critical, energy-hungry infrastructures that run large-scale Internet-based services. Energy consumption models are pivotal in designing and optimizing energy-efficient operations to curb excessive energy consumption in data centers. In this paper, we survey the state-of-the-art techniques used for energy consumption modeling and prediction for data centers and their components. We conduct an in-depth study of the existing literature on data center power modeling, covering more than 200 models. We organize these models in a hierarchical structure with two main branches focusing on hardware-centric and software-centric power models. Under hardware-centric approaches we start from the digital circuit level and move on to describe higher-level energy consumption models at the hardware component level, server level, data center level, and finally systems of systems level. Under the software-centric approaches we investigate power models developed for operating systems, virtual machines and software applications. This systematic approach allows us to identify multiple issues prevalent in power modeling of different levels of data center systems, including: i) few modeling efforts targeted at power consumption of the entire data center ii) many state-of-the-art power models are based on a few CPU or server metrics, and iii) the effectiveness and accuracy of these power models remain open questions. Based on these observations, we conclude the survey by describing key challenges for future research on constructing effective and accurate data center power models.

Data Center Energy Consumption Modeling: A Survey

Cloud computing environments allow customers to dynamically scale their applications. The key problem is how to lease the right amount of resources, on a pay-as-you-go basis. Application re-dimensioning can be implemented effortlessly, adapting the resources assigned to the application to the incoming user demand. However, the identification of the right amount of resources to lease in order to meet the required Service Level Agreement, while keeping the overall cost low, is not an easy task. Many techniques have been proposed for automating application scaling. We propose a classification of these techniques into five main categories: static threshold-based rules, control theory, reinforcement learning, queuing theory and time series analysis. Then we use this classification to carry out a literature review of proposals for auto-scaling in the cloud.

A Review of Auto-scaling Techniques for Elastic Applications in Cloud Environments

The recently introduced series of description logics under the common moniker 'DL-Lite' has attracted attention of the description logic and semantic web communities due to the low computational complexity of inference, on the one hand, and the ability to represent conceptual modeling formalisms, on the other. The main aim of this article is to carry out a thorough and systematic investigation of inference in extensions of the original DL-Lite logics along five axes: by (i) adding the Boolean connectives and (ii) number restrictions to concept constructs, (iii) allowing role hierarchies, (iv) allowing role disjointness, symmetry, asymmetry, reflexivity, irreflexivity and transitivity constraints, and (v) adopting or dropping the unique name assumption. We analyze the combined complexity of satisfiability for the resulting logics, as well as the data complexity of instance checking and answering positive existential queries. Our approach is based on embedding DL-Lite logics in suitable fragments of the one-variable first-order logic, which provides useful insights into their properties and, in particular, computational behavior.

/pdf/the-dl-lite-family-and-relations-4dq237sy63.pdf

The DL-lite family and relations

Cloud research to date has lacked data on the characteristics of the production virtual machine (VM) workloads of large cloud providers. A thorough understanding of these characteristics can inform the providers' resource management systems, e.g. VM scheduler, power manager, server health manager. In this paper, we first introduce an extensive characterization of Microsoft Azure's VM workload, including distributions of the VMs' lifetime, deployment size, and resource consumption. We then show that certain VM behaviors are fairly consistent over multiple lifetimes, i.e. history is an accurate predictor of future behavior. Based on this observation, we next introduce Resource Central (RC), a system that collects VM telemetry, learns these behaviors offline, and provides predictions online to various resource managers via a general client-side library. As an example of RC's online use, we modify Azure's VM scheduler to leverage predictions in oversubscribing servers (with oversubscribable VM types), while retaining high VM performance. Using real VM traces, we then show that the prediction-informed schedules increase utilization and prevent physical resource exhaustion. We conclude that providers can exploit their workloads' characteristics and machine learning to improve resource management substantially.

/pdf/resource-central-understanding-and-predicting-workloads-for-3etxuzxef2.pdf

Resource Central: Understanding and Predicting Workloads for Improved Resource Management in Large Cloud Platforms

Empirical prediction models for adaptive resource provisioning in the cloud

A new class of data storage systems, called NoSQL (Not Only SQL), have emerged to complement traditional database systems, with rejection of general ACID transactions as one common feature. Different platforms, and indeed different primitives within one NoSQL platform, can offer various consistency properties, from Eventual Consistency to single-entity ACID. For the platform provider, weaker consistency should allow better availability, lower latency, and other benefits. This paper investigates what consumers observe of the consistency and performance properties of various offerings. We find that many platforms seem in practice to offer more consistency than they promise; we also find cases where the platform offers consumers a choice between stronger and weaker consistency, but there is no observed benefit from accepting weaker consistency properties.

/pdf/data-consistency-properties-and-the-trade-offs-in-commercial-1x843en9ax.pdf

Data Consistency Properties and the Trade-offs in Commercial Cloud Storages: the Consumers' Perspective

One major benefit claimed for cloud computing is elasticity: the cost to a consumer of computation can grow or shrink with the workload. This paper offers improved ways to quantify the elasticity concept, using data available to the consumer. We define a measure that reflects the financial penalty to a particular consumer, from under-provisioning (leading to unacceptable latency or unmet demand) or over-provisioning (paying more than necessary for the resources needed to support a workload). We have applied several workloads to a public cloud; from our experiments we extract insights into the characteristics of a platform that influence its elasticity. We explore the impact of the rules used to increase or decrease capacity.

How a consumer can measure elasticity for cloud platforms

For ontologies represented as Description Logic Tboxes, optimised DL reasoners are able to detect logical errors, but there is comparatively limited support for resolving such problems. One possible remedy is to weaken the available information to the extent that the errors disappear, but to limit the weakening process as much as possible. The most obvious way to do so is to remove just enough Tbox sentences to eliminate the errors. In this paper we propose a tableau-like procedure for finding maximally concept-satisfiable terminologies represented in the description logic ALC. We discuss some optimisation techniques, and report on preliminary, but encouraging, experimental results.

https://www.aaai.org/Papers/AAAI/2006/AAAI06-043.pdf

Finding maximally satisfiable terminologies for the description logic ALC

Description logic reasoners are able to detect incoherences (such as logical inconsistency and concept unsatisfiability) in knowledge bases, but provide little support for resolving them. We propose to recast techniques for propositional inconsistency management into the description logic setting. We show that the additional structure afforded by description logic statements can be used to refine these techniques. Our focus in this paper is on the formal semantics for such techniques, although we do provide high-level decision procedures for the knowledge integration strategies discussed.

/pdf/knowledge-integration-for-description-logics-2spfz8j7fx.pdf

Kevin Lee

Papers

Empirical prediction models for adaptive resource provisioning in the cloud

Data Consistency Properties and the Trade-offs in Commercial Cloud Storages: the Consumers' Perspective

How a consumer can measure elasticity for cloud platforms

Finding maximally satisfiable terminologies for the description logic ALC

Knowledge integration for description logics