Cloud computing makes extensive use of virtual machines because they permit workloads to be isolated from one another and for the resource usage to be somewhat controlled. In this paper, we explore the performance of traditional virtual machine (VM) deployments, and contrast them with the use of Linux containers. We use KVM as a representative hypervisor and Docker as a container manager. Our results show that containers result in equal or better performance than VMs in almost all cases. Both VMs and containers require tuning to support I/Ointensive applications. We also discuss the implications of our performance results for future cloud architectures.

/pdf/an-updated-performance-comparison-of-virtual-machines-and-51dkcxz9au.pdf

An updated performance comparison of virtual machines and Linux containers

Presentacio sobre l'Oficina de Proteccio de Dades Personals de la UAB i la politica Open Science. Va formar part de la conferencia "Les politiques d'Open Data / Open Acces: Implicacions a la recerca" orientada a investigadors i gestors de projectes europeus que va tenir lloc el 20 de setembre de 2018 a la Universitat Autonoma de Barcelona

/pdf/general-data-protection-regulation-2z11cloyp7.pdf

General data protection regulation

The system and method described herein may identify one or more virtual desktop extensions available in a cloud computing environment and launch virtual machine instances to host the available virtual desktop extensions in the cloud. For example, a virtual desktop extension manager may receive a virtual desktop extension request from a client desktop and determine whether authentication credentials for the client desktop indicate that the client desktop has access to the requested virtual desktop extension. In response to authenticating the client desktop, the virtual desktop extension manager may then launch a virtual machine instance to host the virtual desktop extension in the cloud and provide the client desktop with information for locally controlling the virtual desktop extension remotely hosted in the cloud.

System and method for providing virtual desktop extensions on a client desktop

Virtualization of operating systems provides a common way to run different services in the cloud. Recently, the lightweight virtualization technologies claim to offer superior performance. In this paper, we present a detailed performance comparison of traditional hypervisor based virtualization and new lightweight solutions. In our measurements, we use several benchmarks tools in order to understand the strengths, weaknesses, and anomalies introduced by these different platforms in terms of processing, storage, memory and network. Our results show that containers achieve generally better performance when compared with traditional virtual machines and other recent solutions. Albeit containers offer clearly more dense deployment of virtual machines, the performance difference with other technologies is in many cases relatively small.

Hypervisors vs. Lightweight Virtualization: A Performance Comparison

The system and method for intelligent workload management described herein may include a computing environment having a model-driven, service-oriented architecture for creating collaborative threads to manage workloads, wherein the management threads may converge information for managing identities and access credentials, enforcing policies, providing compliance assurances, managing provisioned and requested services, and managing physical and virtual infrastructure resources In one implementation, an authentication server may generate authentication tokens defining access credentials for managed entities across a plurality of authentication domains, wherein the authentication tokens may control access to resources in an information technology infrastructure For example, a management infrastructure may create service distributions for the managed entities, which may include virtual machine images hosted on physical resources Further, the authentication tokens may be embedded in the service distributions, whereby the embedded authentication tokens may control access to the resources in the information technology infrastructure

System and method for intelligent workload management

Virtualization has become a popular way to make more efficient use of server resources within both private data centers and public cloud platforms. While recent advances in CPU architectures and new virtualization techniques have reduced the performance cost of using virtualization, overheads still exist, particularly when multiple virtual machines are competing for resources. We have performed an extensive performance comparison under hardware-assisted virtualization settings considering four popular virtualization platforms, Hyper-V, KVM, vSphere and Xen, and find that the overheads incurred by each hypervisor can vary significantly depending on the type of application and the resources assigned to it. We also find dramatic differences in the performance isolation provided by different hypervisors. However, we find no single hypervisor always outperforms the others. This suggests that effectively managing hypervisor diversity in order to match applications to the best platform is an important, yet unstudied, challenge.

A component-based performance comparison of four hypervisors

The complexity of modern product realization processes requires collaborative work of engineering teams from different disciplines. We apply principles from game theory to model the relationships between engineering teams and facilitate collaborative decision making. In order to maintain design freedom in the early stages of product realization so that engineering teams in the later stages can adjust their decisions while still maintaining feasibility, we postulate the use of design capability indices to facilitate the teams making a ranged set of decisions, instead of specific ones. The effect of game theoretic principles and design capability indices on design freedom and therefore on the design solutions is also investigated. An electronic package design and analysis scenario is used to demonstrate the efficacy of the proposed method.

Collaborative multidisciplinary decision making using game theory and design capability indices

Endpoint management, including patching, health checking, configuration etc., is a key function for data center and cloud management. Managing multiple nodes through automation tools or scripts significantly increases efficiency. However, the risks of adverse impact due to excessive privilege or human error may propagate to a large pool of endpoints and lead to massive service disruptions and SLA (Service Level Agreement) violations. In this paper, we present a system that proactively and systematically manages the risk throughout the lifecycle phases of automation. We present a prototype implementation consisting of an authorization mechanism that guarantees the right level of eligibility and privilege of accessing the automation content (during the deployment stage), and an execution validator that controls the risk of human error which may cause massive damage to the infrastructure (during execution of the automation content). Our current implementation has been deployed to more than a dozen customer environments and achieved an efficiency gain of 58% with high execution accuracy.

Managing risk in multi-node automation of endpoint management

A plurality of artifacts, in a plurality of repositories, are maintained under revision control. A plurality of artifact tuples, in another repository, are maintained under revision control. At least one first artifact tuple comprise first information providing access to at least one of the artifacts. At least one second artifact tuple comprises second information providing access to at least one other artifact tuple. Each of the artifact tuples are associated with at least one of a plurality of product development processes. An apparatus comprises a plurality of repositories, each maintaining under revision control a plurality of artifacts and another repository that maintains under revision control a plurality of artifact tuples. A workflow process is adapted to allow users to associate each of the artifact tuples with at least one of a plurality of product development processes.

Techniques supporting collaborative product development

It is commonly agreed that accounts receivable (AR) can be a source of financial difficulty for firms when they are not efficiently managed and are underperforming. Experience across multiple industries shows that effective management of AR and overall financial performance of firms are positively correlated. In this paper we address the problem of reducing outstanding receivables through improvements in the collections strategy. Specifically, we demonstrate how supervised learning can be used to build models for predicting the payment outcomes of newly-created invoices, thus enabling customized collection actions tailored for each invoice or customer. Our models can predict with high accuracy if an invoice will be paid on time or not and can provide estimates of the magnitude of the delay. We illustrate our techniques in the context of real-world transaction data from multiple firms. Finally, simulation results show that our approach can reduce collection time up to a factor of four compared to a baseline that is not model-driven.

/pdf/using-predictive-analysis-to-improve-invoice-to-cash-3e0uo8l76u.pdf

Sai Zeng

Papers

A component-based performance comparison of four hypervisors

Collaborative multidisciplinary decision making using game theory and design capability indices

Managing risk in multi-node automation of endpoint management

Techniques supporting collaborative product development

Using predictive analysis to improve invoice-to-cash collection