Showing papers on "Temporal isolation among virtual machines published in 2012"

Systems and methods for analyzing performance of virtual environments

Abstract: Intelligent monitoring systems and methods for virtual environments are disclosed that understand various components of a virtual infrastructure and how the components interact to provide improved performance analysis to users. In certain examples, a monitoring system assesses the performance of virtual machine(s) in the context of the overall performance of the physical server(s) and the environment in which the virtual machine(s) are running. For instance, the monitoring system can track performance metrics over a determined period of time to view changes to the allocation of resources to virtual machines and their location(s) on physical platforms. Moreover, monitoring systems can utilize past performance information from separate virtual environments to project a performance impact resulting from the migration of a virtual machine from one physical platform to another.

Performance isolation and fairness for multi-tenant cloud storage

Abstract: Shared storage services enjoy wide adoption in commercial clouds. But most systems today provide weak performance isolation and fairness between tenants, if at all. Misbehaving or high-demand tenants can overload the shared service and disrupt other well-behaved tenants, leading to unpredictable performance and violating SLAs.This paper presents Pisces, a system for achieving datacenter-wide per-tenant performance isolation and fairness in shared key-value storage. Today's approaches for multi-tenant resource allocation are based either on per-VM allocations or hard rate limits that assume uniform workloads to achieve high utilization. Pisces achieves per-tenant weighted fair shares (or minimal rates) of the aggregate resources of the shared service, even when different tenants' partitions are co-located and when demand for different partitions is skewed, time-varying, or bottlenecked by different server resources. Pisces does so by decomposing the fair sharing problem into a combination of four complementary mechanisms--partition placement, weight allocation, replica selection, and weighted fair queuing--that operate on different time-scales and combine to provide system-wide max-min fairness.An evaluation of our Pisces storage prototype achieves nearly ideal (0.99 Min-Max Ratio) weighted fair sharing, strong performance isolation, and robustness to skew and shifts in tenant demand. These properties are achieved with minimal overhead (

Dynamic resource management using virtual machine migrations

Abstract: Virtualization is a key concept in enabling the "computing-as-a-service" vision of cloud-based solutions. Virtual machine related features such as flexible resource provisioning, and isolation and migration of machine state have improved efficiency of resource usage and dynamic resource provisioning capabilities. Live virtual machine migration transfers "state" of a virtual machine from one physical machine to another, and can mitigate overload conditions and enables uninterrupted maintenance activities. The focus of this article is to present the details of virtual machine migration techniques and their usage toward dynamic resource management in virtualized environments. We outline the components required to use virtual machine migration for dynamic resource management in the virtualized cloud environment. We present categorization and details of migration heuristics aimed at reducing server sprawl, minimizing power consumption, balancing load across physical machines, and so on. We conclude with a discussion of open research problems in the area.

Systems and methods for sr-iov pass-thru via an intermediary device

Abstract: This disclosure describes a system for Single Root I/O Virtualization (SR-IOV) pass-thru for network packet processing via a virtualized environment of a device. The system includes a device comprising a virtualized environment and a plurality of virtual machines having a virtual network interface for receiving and transmitting network packets. A driver for the physical network interface of the device creates a plurality of virtual devices corresponding to the physical network interface, which appear as a Peripheral Component Interconnect (PCI) device to the virtualized environment. A virtual device of the plurality of virtual devices is assigned via the virtualized environment to each virtual machine of the plurality of virtual machines. The virtual machine uses the virtual device assigned to the virtual machine, to receive and transmit network packets via the physical network interface of the device.

Improving consolidation of virtual machines with risk-aware bandwidth oversubscription in compute clouds

Abstract: Current trends in virtualization, green computing, and cloud computing require ever increasing efficiency in consolidating virtual machines without degrading quality of service. In this work, we consider consolidating virtual machines on the minimum number of physical containers (e.g., hosts or racks) in a cloud where the physical network (e.g., network interface or top of the rack switch link) may become a bottleneck. Since virtual machines do not simultaneously use maximum of their nominal bandwidth, the capacity of the physical container can be multiplexed. We assume that each virtual machine has a probabilistic guarantee on realizing its bandwidth Requirements-as derived from its Service Level Agreement with the cloud provider. Therefore, the problem of consolidating virtual machines on the minimum number of physical containers, while preserving these bandwidth allocation guarantees, can be modeled as a Stochastic Bin Packing (SBP) problem, where each virtual machine's bandwidth demand is treated as a random variable. We consider both offline and online versions of SBP. Under the assumption that the virtual machines' bandwidth consumption obeys normal distribution, we show a 2-approximation algorithm for the offline version and improve the previously reported results by presenting a (2 +∈)-competitive algorithm for the online version. We also observe that a dual polynomial-time approximation scheme (PTAS) for SBP can be obtained via reduction to the two-dimensional vector bin packing problem. Finally, we perform a thorough performance evaluation study using both synthetic and real data to evaluate the behavior of our proposed algorithms, showing their practical applicability.

vSlicer: latency-aware virtual machine scheduling via differentiated-frequency CPU slicing

Abstract: Recent advances in virtualization technologies have made it feasible to host multiple virtual machines (VMs) in the same physical host and even the same CPU core, with fair share of the physical resources among the VMs. However, as more VMs share the same core/CPU, the CPU access latency experienced by each VM increases substantially, which translates into longer I/O processing latency perceived by I/O-bound applications. To mitigate such impact while retaining the benefit of CPU sharing, we introduce a new class of VMs called latency-sensitive VMs (LSVMs), which achieve better performance for I/O-bound applications while maintaining the same resource share (and thus cost) as other CPU-sharing VMs. LSVMs are enabled by vSlicer, a hypervisor-level technique that schedules each LSVM more frequently but with a smaller micro time slice. vSlicer enables more timely processing of I/O events by LSVMs, without violating the CPU share fairness among all sharing VMs. Our evaluation of a vSlicer prototype in Xen shows that vSlicer substantially reduces network packet round-trip times and jitter and improves application-level performance. For example, vSlicer doubles both the connection rate and request processing throughput of an Apache web server; reduces a VoIP server's upstream jitter by 62%; and shortens the execution times of Intel MPI benchmark programs by half or more.

Cloud-based middlebox management system

Abstract: A virtual network virtual machine may be implemented on a cloud computing facility to control communication among virtual machines executing applications and virtual machines executing middlebox functions. This virtual network virtual machine may provide for automatic scaling of middleboxes according to a heuristic algorithm that monitors the effectiveness of each middlebox on the network performance as application virtual machines are scaled. The virtual machine virtual network may also locate virtual machines in actual hardware to further optimize performance.

Virtual machine snapshotting and analysis

Abstract: Described are techniques for capturing and analyzing snapshots of virtual machines. One or more computers may automatically obtain snapshots of virtual machines as they are executing to form a pool of virtual machine snapshots. The virtual machine snapshots are then read to obtain a set of features properties of the virtual machine snapshots, including information about a running guest operating system, software installed on the virtual machine, metadata about the virtual machine itself, and others. The features or properties are analyzed, in one embodiment using a machine learning algorithm, to automatically compute and store information about the virtual machines.

Method and architecture for virtual desktop service

Abstract: The present invention relates to a method and an architecture capable of efficiently providing a virtual desktop service. The service architecture for the virtual desktop service includes a connection broker for performing the management of virtual machines, a server monitoring function, and a protocol coordination function. A resource pool is configured to manage software resources that are transferred to a specific virtual machine in a streaming form at a predetermined time and that are executed on the specific virtual machine and to provide provision information about the managed software resources at the request of the connection broker, in order to provide an on-demand virtual desktop service. A virtual machine infrastructure is configured to support hardware resources, generate virtual machines in which the software of the user terminal is operated, and provide the virtual machines as virtual desktops.

Method and apparatus for maintaining a workload service level on a converged platform

Abstract: A management server is provided in a computer system having one or more hosts, one or more storage systems and one or more switches, the hosts having a plurality of virtual machines, each virtual machine being defined according to a service level agreement. The management server is operable to manage the virtual machines and resources associated with the virtual machines; receive a notification of an event from a node in the computer system; determine if the event affects a service level agreement for any of the virtual machines defined in the computer system, the service level agreements listing required attributes for the corresponding virtual machines; allocate a new resource for a virtual machine whose service level agreement is affected by the event; and move the virtual machine whose service level agreement is affected by the event to the newly allocated resource.

Energy-Efficient virtual machine placement in data centers by genetic algorithm

Abstract: Server consolidation using virtualization technology has become an important technology to improve the energy efficiency of data centers. Virtual machine placement is the key in the server consolidation. In the past few years, many approaches to the virtual machine placement have been proposed. However, existing virtual machine placement approaches to the virtual machine placement problem consider the energy consumption by physical machines in a data center only, but do not consider the energy consumption in communication network in the data center. However, the energy consumption in the communication network in a data center is not trivial, and therefore should be considered in the virtual machine placement in order to make the data center more energy-efficient. In this paper, we propose a genetic algorithm for a new virtual machine placement problem that considers the energy consumption in both the servers and the communication network in the data center. Experimental results show that the genetic algorithm performs well when tackling test problems of different kinds, and scales up well when the problem size increases.

System and method for enforcement of security controls on virtual machines throughout life cycle state changes

Abstract: Systems and methods associated with virtual machine security are described herein. One example method includes instantiating a guest virtual machine in a virtual computing environment. The method also includes installing a life cycle agent on the guest virtual machine, assigning an identifying certificate, a set of policies, and an encryption key to the guest virtual machine, and providing the certificate, policies, and encryption key to the guest virtual machine. The certificate, policies, and encryption key may then be used by the guest virtual machine to authenticate itself within the virtual computing environment and to protect data stored on the guest virtual machine.

Management of inter-dependent configurations of virtual machines in a cloud

Abstract: A server computer system determines that configuring a first virtual machine in a cloud depends on a configuration result of configuring a second virtual machine. The server computer system configures the second virtual machine in the cloud and configures the first virtual machine in the cloud using the configuration result of the second virtual machine.

Online traffic-aware virtual machine placement in data center networks

Abstract: Data centers rely on virtualization to provide different services over a shared infrastructure. The placement of the different services and tasks in the physical machines is crucial for the performance of the whole system. A misplaced service can overload some network links, lead to congestion, or even connection disruptions. On the other hand, virtual machine migration allows reallocating services and changing the traffic matrix, leading to more efficient use of bandwidth. In this paper, we propose a Virtual Machine Placement (VMP) algorithm to (re)allocate virtual machines in the data center servers, based on the current traffic matrix, CPU, and memory usage. Analyzing the formation of community patterns in terms of traffic using graph theory, we are able to find virtual machines that are correlated because they exchange high amount of data. Those virtual machines are aggregated and allocated to servers as close as possible to each other, reducing traffic congestion. Our simulation results show that VMP was able to improve the traffic distribution. In some specific cases we were able to reduce 80% of the core traffic, concentrating it at the edge of the network.

System and method for migrating virtual machines

Abstract: A method, computer program product, and computing system for receiving an indication that a target virtual machine is going to be migrated from a first operating environment to a second operating environment. The target cache system is associated with target virtual machine. An auxiliary virtual machine is generated within the second operating environment. An auxiliary cache system is associated with the auxiliary virtual machine. The target virtual machine and the auxiliary virtual machine are connected. IO requests for the target virtual machine are mirrored to the auxiliary virtual machine. At least a portion of cache data included within a target memory system associated with the target cache system is copied to an auxiliary memory system associated with the auxiliary cache system. The target virtual machine is migrated from the first operating environment to the second operating environment.

Metrics and techniques for quantifying performance isolation in cloud environments

Abstract: The cloud computing paradigm enables the provision of costefficient IT-services by leveraging economies of scale and sharing data center resources efficiently among multiple independent applications and customers. However, the sharing of resources leads to possible interference between users and performance problems are one of the major obstacles for potential cloud customers. Consequently, it is one of the primary goals of cloud service providers to have different customers and their hosted applications isolated as much as possible in terms of the performance they observe.To make different offerings, comparable with regards to their performance isolation capabilities, a representative metric is needed to quantify the level of performance isolation in cloud environments. Such a metric should allow to measure externally by running benchmarks from the outside treating the cloud as a black box. In this paper, we propose three different types of novel metrics for quantifying the performance isolation of cloud-based systems and a simulation-based case study applying these metrics in the context of a Softwareas-a-Service (SaaS) scenario where different customers (tenants) share one single application instance. We consider four different approaches to achieve performance isolation and evaluate them based on the proposed metrics. The results demonstrate the effectiveness and practical usability of the proposed metrics in quantifying the performance isolation of cloud environments.

Application-Level CPU Consumption Estimation: Towards Performance Isolation of Multi-tenancy Web Applications

Abstract: Performance isolation is a key requirement for application-level multi-tenant sharing hosting environments. It requires knowledge of the resource consumption of the various tenants. It is of great importance not only to be aware of the resource consumption of a tenant's given kind of transaction mix, but also to be able to be aware of the resource consumption of a given transaction type. However, direct measurement of CPU resource consumption requires instrumentation and incurs overhead. Recently, regression analysis has been applied to indirectly approximate resource consumption, but challenges still remain for cases with non-determinism and multicollinearity. In this work, we adapts Kalman filter to estimate CPU consumptions from easily observed data. We also propose techniques to deal with the non-determinism and the multicollinearity issues. Experimental results show that estimation results are in agreement with the corresponding measurements with acceptable estimation errors, especially with appropriately tuned filter settings taken into account. Experiments also demonstrate the utility of the approach in avoiding performance interference and CPU overloading.

Demand based hierarchical QoS using storage resource pools

Abstract: The high degree of storage consolidation in modern virtualized datacenters requires flexible and efficient ways to allocate IO resources among virtual machines (VMs). Existing IO resource management techniques have two main deficiencies: (1) they are restricted in their ability to allocate resources across multiple hosts sharing a storage device, and (2) they do not permit the administrator to set allocations for a group of VMs that are providing a single service or belong to the same application. In this paper we present the design and implementation of a novel software system called Storage Resource Pools (SRP). SRP supports the logical grouping of related VMs into hierarchical pools. SRP allows reservations, limits and proportional shares, at both the VM and pool levels. Spare resources are allocated to VMs in the same pool in preference to other VMs. The VMs may be distributed across multiple physical hosts without consideration of their logical groupings. We have implemented a prototype of storage resource pools in the VMware ESX hypervisor. Our results demonstrate that SRP provides hierarchical performance isolation and sharing among groups of VMs running across multiple hosts, while maintaining high utilization of the storage device.

Automated tuning in a virtual machine computing environment

Abstract: Automatic optimization of application performance parameters is provided. This optimization may be provided by identifying the resource demands of applications using virtual resources, identifying the virtual resources that can be available to the applications, and allocating the virtual resources to the applications to be run on a virtual machine.

Method and apparatus for assignment of virtual resources within a cloud environment

Abstract: A virtual resource assignment capability is disclosed. The virtual re¬ source assignment capability is configured to support provisioning of virtual resources within a cloud environment (110). The provision¬ ing of virtual resources within a cloud environment includes receiving a user virtual resource request requesting provisioning of virtual re¬ sources within the cloud environment, determining virtual resource assignment information specifying assignment of virtual resources within the cloud environment, and provisioning the virtual resources within the cloud environment using the virtual resource assignment information. The assignment of the requested virtual resources with¬ in the cloud environment includes assignment of the virtual resource to datacenters (111) of the cloud environment in which the virtual resources will be hosted and, more specifically, to the physical re¬ sources within the datacenters of the cloud environment in which the virtual resources will be hosted. The virtual resources may include vir¬ tual processor resources, virtual memory resources, and the like. The physical resources may include processor resources (115), storage re¬ sources, and the like (e.g., physical resources of blade servers (114) of racks (112) of datacenters of the cloud environment).

Towards bandwidth guarantee in multi-tenancy cloud computing networks

Abstract: To efficiently utilize their infrastructure and thus increase their revenue, cloud providers need mechanisms to provide resource allocation and performance isolation for different tenants in the shared platform. In particular, network bandwidth sharing is a critical yet still an open problem to most cloud providers. In this paper, we study the problem of virtual machine (VM) allocation under the consideration of providing bandwidth guarantees. We first propose an online allocation algorithm for tenants with homogeneous bandwidth demand, which improves on the accuracy of existing algorithms. Subsequently, we extend it to handle heterogeneous bandwidth demand. Extensive simulations show that our algorithm makes much more efficient utilization of the network resource than existing algorithms, and performs close to the optimal offline allocation.

Methods and systems for evaluating historical metrics in selecting a physical host for execution of a virtual machine

Abstract: Methods and systems for improved management of power utilization and resource consumption among physical hosts in a cloud computing environment. The management server may provide functionality facilitating the identification and optimized placement of a virtual machine within a cloud computing environment by evaluating historical and heuristic metrics data associated with both the physical hosts and the virtual machines. The management server utilizes the metrics data to generate scores for a plurality of physical host based on physical resources available in a cloud of computing resources. The management server identifies a physical host on which to place a virtual machine using the metrics data, generated scores, and numerous, configurable criteria. The management server responds to the identification of the physical host on which to place a virtual machine by adjusting processor performance and/or operating states for one or more of the physical hosts in the cloud computing environment.

Cloud-Based Virtual Machines and Offices

Abstract: Cloud-based virtual machines and offices are provided herein. Methods may include establishing a cloud-based virtual office, by providing selections, corresponding to backups of servers of a computing network, to a user interface, establishing a cloud gateway for the virtual office, virtualizing a backup for each server using a virtualization program to create the cloud-based virtual office that includes virtual server machines networked with one another via the cloud gateway, and providing a workload to the cloud-based virtual office.

Secure migration of virtual machines

Abstract: Technologies are generally described for the secure live migration of virtual machines. The migration may take place in the context of, for example, public clouds. In various embodiments, by using a hidden process incorporated in a virtual machine's kernel and a trusted wireless and/or wired positioning service, a cloud provider and/or cloud user may be alerted about possible virtual machine hijacking/theft. The provider or user may also be provided with an approximate physical location of the platform running the compromised virtual machine for further investigation and enforcement measures.

Grids, Clouds and Virtualization

Abstract: Research into grid computing has been driven by the need to solve large-scale, increasingly complex problems for scientific applications. Yet the applications of grid computing for business and casual users did not begin to emerge until the development of the concept of cloud computing, fueled by advances in virtualization techniques, coupled with the increased availability of ever-greater Internet bandwidth. The appeal of this new paradigm is mainly based on its simplicity, and the affordable price for seamless access to both computational and storage resources. This timely text/reference introduces the fundamental principles and techniques underlying grids, clouds and virtualization technologies, as well as reviewing the latest research and expected future developments in the field. Readers are guided through the key topics by internationally recognized experts, enabling them to develop their understanding of an area likely to play an ever more significant role in coming years. Topics and features: presents contributions from an international selection of experts in the field; provides a thorough introduction and overview of existing technologies in grids, clouds and virtualization, including a brief history of the field; examines the basic requirements for performance isolation of virtual machines on multi-core servers, analyzing a selection of system virtualization technologies; examines both business and scientific applications of grids and clouds, including their use in the life sciences and for high-performance computing; explores cloud building technologies, architectures for enhancing grid infrastructures with cloud computing, and cloud performance; discusses energy aware grids and clouds, workflows on grids and clouds, and cloud and grid programming models. This useful text will enable interested readers to familiarize themselves with the key topics of grids, clouds and virtualization, and to contribute to new advances in the field. Researchers, undergraduate and graduate students, system designers and programmers, and IT policy makers will all benefit from the material covered.

Application installation management by selectively reuse or terminate virtual machines based on a process status

Abstract: Example systems and methods of managing installation of applications are described. In one implementation, a repository stores data associated with multiple application instances. An orchestrator analyzes application instances running on multiple provisioned virtual machines and identifies at least one unused virtual machine among the multiple provisioned virtual machines. The orchestrator terminates processes running on the unused virtual machine and clears data associated with the unused virtual machine. The cleared virtual machine is added to a virtual machine pool for use by another application instance.

Providing overlay networks via elastic cloud networking

Abstract: A device receives network infrastructure requirements from a user, and selects, based on the network infrastructure requirements, at least one of a virtual core network, a virtual network resource, and a virtual system from resources provided in a cloud computing environment. The device also enables use of the selected at least one of the virtual core network, the virtual network resource, and the virtual system by the user. The device receives, from the user, traffic destined for the selected at least one of the virtual core network, the virtual network resource, and the virtual system, and provides the traffic to the selected at least one of the virtual core network, the virtual network resource, and the virtual system.

Virtual machine placement to improve memory utilization

Abstract: Virtual machines having a high amount of identical memory pages are grouped under a common hypervisor to enable greater memory savings as the result of transparent page sharing. One method comprises analyzing the software image content of a plurality of virtual machines running on a plurality of hypervisors in a common migration domain, identifying two or more of the virtual machines having greater than a threshold amount of common memory pages, placing the two or more virtual machines under a common hypervisor, and sharing the common memory pages among the two or more virtual machines. Optionally, the identifying of two or more of the virtual machines may include identifying two or more of the virtual machines having the same software image content classifications; thereby, identifying images with a greater affinity for common memory pages.

Integrated virtual infrastructure system

Abstract: A technique is provided for creating virtual units in a computing environment. A virtual system definition is received by a processor that is utilized to create the virtual units for a virtual system. Relationship constraints between the virtual units in the virtual system are received by the processor. The relationship constraints between the virtual units include a communication link requirement between the virtual units and/or a location requirement between the virtual units. The virtual units in the virtual system are deployed by the processor according to the relationship constraints between virtual units.

Augmented allocation of virtual machines for application

Abstract: The starting up of an application involving multiple virtual machines by overallocating virtual machines. In response to a request to allocate a certain number of virtual machines corresponding to the application, an augmented number of virtual machines is determined to be allocated in response to the request. The augmented number includes both the initially requested number of virtual machines in addition to a surplus number of virtual machines. The virtual machines are then initiated to start up if they are not already started up. Before all of the virtual machines are started up, code is bound to the virtual machines. Thus, because more virtual machines were initiated startup than are required for the application, the code may be bound to some of the virtual machines in the application before all of the virtual machine have started up.