Search

In this paper, we present the architecture and implementation of a virtual network computers' (VNC) optical storage virtualization scheme called VOSV. Its task is to manage the mapping of virtual optical storage to physical optical storage, a technique known as optical storage virtualization. The design of VOSV aims at the optical storage resources of different clients and servers that have high read-sharing patterns. VOSV uses several schemes such as a two-level Cache mechanism, a VNC server embedded module and the iSCSI protocols to improve the performance. The results measured on the prototype are encouraging, and indicating that VOSV provides the high I/O performance.

Previous full virtualization techniques are implemented in software without any hardware assist. In this paper, for the first time, we propose an advanced hardware assisted full virtualization architecture-Dynamic Binary Translation in DIMM (DBTIM). We integrate a reconfigurable dynamic binary translation chip into a DIMM. DBTIM could be inserted to the mainboard as a normal DIMM, hence easily making the host CPU virtualizable. We also implement a DBTIM prototype on FPGA. Furthermore, we propose an EDBTIM technique-Enhanced Dynamic Binary Translator in Memory, which integrates dynamic binary translator and memory on one SDRAM chip. We analyze our new approaches and compare them with other modern virtualization techniques. We demonstrate that the new techniques in this paper could provide full virtualization to all kinds of CPU architectures, including x86, with better virtualization capability, better software and hardware compatibility, better performance and lower overheads.

The paper presents a novel network RAID storage system based on the out-of-band virtualization architecture and the backend centralized redundant management. The application servers can inquire the mapping information of virtual disk from the out-of-band virtualization server and directly access the storage nodes. The read request can fetch the data from special storage node, while the write request is not only stored into the storage node, and also mirrored into the redundant server by the storage node. The redundant server can cache updated data in local disks with log-structured mode, and calculate the parity of RAID5 in the background process when the system is idle. It relieves the bottle problem of I/O performance and low reliability danger of single controller in the front-end centralized management system. And the layout of RAID1/RAID5 on the data block has acquired the trade-off among performance, reliability and cost.

Virtualization in grid environments is a recent way to improve platform usage. ViSaGe is a middleware designed to provide set of functionalities needed for storage virtualizaion: transparent reliable remote access to data and distributed data management. ViSaGe aggregates distributed physical storage resources. However, ensuring the performances of data access in grid environment is a major issue, as large amount of data are stored and constantly accessed, and directly involved into tasks execution time. Especially, the placement and selection of replicated data are made particularly difficult because of the dynamic nature of grid environments - grid nodes workload variations. The workload variations represent the state of the system resources (CPU, disks and networks). These variations are mainly perceived by a monitoring system. Several monitoring systems exist in the literature. They monitor system resources consumption and applications but none of these systems presents the whole of the pertinent characteristics for ViSaGe. ViSaGe needs a system that analyzes nodes workload during runtime execution for improving data storage management. Therefore, ViSaGe Administration and monitoring service, namely Admon, is proposed. In this paper, we present Admon and its workload prediction model. Admon allows to dynamically placing data according to resources usage, especially disk I/O workload, ensuring the best performances while limiting the monitoring overhead. Experiments show the relevance of Admon decisions.

Virtual machine technology is rapidly emerging as an important component of future computer systems. Full virtualization with dynamic binary translation outperforms paravirtualization and hardware assisted virtualization in x86 CPU virtualization. Previous full virtualization techniques are implemented in software without any hardware assist. In this paper, for the first time, we propose a novel hardware assisted full virtualization technique-dynamic binary translation in DIMM (DBTIM). We integrate a reconfigurable dynamic binary translation chip into a DIMM. DBTIM could be inserted to the mainboard as a normal DIMM, hence easily making the host CPU virtualizable. We also implement a DBTIM prototype on FPGA. We analyze our new approach and compare it with other modern virtualization techniques. We demonstrate that the new technique in this paper could provide full virtualization to all kinds of CPU architectures, including x86, with better virtualization capability, better software and hardware compatibility, better performance and lower overheads.

Virtualization technology has attracted much attention in recent years. This paper describes the vision and mission of ChinaV, which is the national fundamental research program for virtualization technology in China. Furthermore, related topics about single host virtualization, multiple VM management schemes and desktop virtualization will be introduced. We first describe a remote memory virtualization scheme and a VCPU management scheme for efficient use of physical resource. Then we describe a novel live VM migration approach based on deterministic replay with execution trace. Multiple VM management schemes are also introduced for multi-VM virtualization. In desktop virtualization field, we present the LVD, a system that combines the virtualization technology and inexpensive personal computers to realize a lightweight virtual desktop system. All of those schemes and systems are good practices of virtualization solution and they have become a strong foundation of our future work.

We have developed a highly reliable InfiniBand host attached block storage management and virtualization system that supports several off-the-shelf Fibre Channel RAID controllers on the back end. The system is based on the existing IBM TotalStorage SAN Volume Controller (SVC) product, and therefore offers performance, a wide array of storage virtualization features, and support for many existing storage controllers. We provide an overview of the driver design as well as performance results. Large read performance from SVC cache exceeds 3 GB/s in a minimal two-node cluster configuration.

Virtualization is used increasingly in data centers to achieve server consolidation. More and more products (commercial and open source) are available; virtualization is even natively included in Linux distributions like in the upcoming Microsoft Windows server. In this paper we present the virtualization tests carried out by EDIT R&D since 2005 to improve the durability of industrial applications running on obsolete hardware platforms. The applications based on the Intelpsilas x86 architecture, and requiring no specific hardware, can be easily virtualized. Thus their lifetime can be extended as long as the virtualization software is adapted to the new hardware of the servers. For applications requiring specific hardware, virtualization is not easy. In this paper we present how to implement the support of a new peripheral, using the open source virtualization software QEMU. We also discuss how the IOs of the drivers are forwarded from the emulated hardware to another hardware that has the same features.

Multi-dimensional storage virtualization (MDSV) technology allows multiple virtual disks, each with a distinct combination of capacity, latency and bandwidth requirements, to be multiplexed on a physical disk storage system with performance isolation. This paper presents novel design and implementation techniques that solve the availability guarantee and fairness assurance problems in multi-dimensional storage virtualization. First, we show that a measurement-based admission control algorithm can reduce the effective resource requirement of a virtual disk with availability guarantee by accurately estimating its resource needs without prior knowledge of its input workload characteristics. Moreover, to accurately factor disk access overhead into real-time disk request scheduling algorithm, we propose a virtual disk switching overhead extraction and distribution algorithm that can derive the intrinsic disk access overhead associated with each virtual disk so as to achieve perfect performance isolation. Finally, we develop an adaptive server time leap-forward algorithm to effectively address the short-term unfairness problem of virtual clock-based disk scheduler, the only known proportional-share scheduler that is based on wall-clock time and thus enables disk utilization efficiency optimization while delivering disk QoS guarantees.

This work is part of the ViSaGe project. This project concerns the storage virtualization applied in the grid environment. Its goal is to create and manage virtual storage resources by aggregating geographically distributed physical storage resources shared on a Grid. To these shared resources, a quality of service will be associated and related to the data storage performance. In a Grid environment, sharing resources may improve performances if these resources are well managed and if the management software obtains sufficient knowledge about the grid resources workload (computing nodes, storage nodes and links). The grid resources workload is mainly perceived by a monitoring system. Several existing monitoring systems are available for monitoring the grid resources and applications. Each one provides informations according to its aim. For example, Network Weather Service [6] is designed to monitor grid computing nodes and links. On the other hand, Netlogger [8] monitors grid resources and applications in order to detect application's bottleneck. These systems are useful for a post mortem analysis. However, in ViSaGe, we need a system that analyzes the necessary nodes state during execution time. This system is represented by the ViSaGe Administration and monitoring service "Admon". In this paper, we present our scalable distributed system: Admon. Admon traces applications, and calculates the system resources consumption (CPU, Disks, Links). Its originality consists in providing a new opportunity to improve virtual data storage performance by using workload's constraints (e.g., maximum CPU usage percentage). Many constraints will be discussed (such as time's constraint). These constraints allow performance improvement by assigning nodes to the ViSaGe's jobs in an effective manner.

Wireless virtualization aims to support concurrent experiments on shared wireless network resources. This research is based on efforts to enhance the control structure of the wireless test bed 'ORBIT' through different forms of virtualization. The focus of this paper is virtualization in frequency or frequency division multiplexing (FDM), and more specifically using multiple wireless interfaces per node and splitting the wireless medium by assigning a different channel frequency to each interface. A wireless node is virtualized through User Mode Linux (UML), with each sliver or experiment using a different interface. Extensive experimental results are presented that analyze the effectiveness of this method of virtualization in terms of different types of traffic. We examine cross-coupling effects of coexisting virtualized experiments and investigate proof-of-concept virtualization for one typical usage scenarios. This research provides practical guidance for the use of this type of virtualization for simultaneous wireless experiments.

Virtualization, a well established technology in the desktop and server domain, is currently investigated and analyzed with respect to its potential within mobile devices. The paper argues that mobile devices, which are targeting a completely open setup (e.g. Linux based), are facing severe security challenges. Thus, virtualization will be discussed as an enabler for security in mobile devices. Virtualization approaches and elements are discussed in detail taking this use case and existing limitations of mobile embedded devices into account.

This chapter explores the methods for the design of home storage networks and devices that connect to these networks and also looks at future developments for home network storage including how storage networks can be used to simplify the management of consumer devices and overall storage content in and around the home. In addition to sharing the internet, network hardware, and entertainment, there is another factor that may drive home networking and also home network storage: growing quantities of personal digital content such as digital photos, home digital movies, Quicken files, documents, letters, e-mails, and so on. Many options are available and being commercially developed for home networking including use of wired and wireless Ethernet as well as using existing home networks such as coaxial cable, phone, and power lines. The largest general uses of home networks are for media sharing and home reference data backup. A home storage network enables all the digital devices with storage that come into association with the home network to be part of a virtualized storage pool with automatic data replication, backup, remote access and sharing of content, organization and indexing of the content, and data synchronization between devices and locations. The growth of networks in the home drives the growth of network storage devices in the home.

Recent advances in virtualization technologies have allowed telecommunications carriers to apply virtualization-based operation support systems. By separating logical view of computing resources from physical view, virtualization is expected to help consolidate/simplify existing complex telecommunication systems, improve utilization rate and reduce Total Cost of Ownership. While the benefits of virtualization are compelling, potential drawbacks also must be identified before actual deployment. Without proper evaluation, virtualization cannot benefit operation support systems. From the viewpoint of operation support systems, this paper (1) analyses performance issues (CPU, network throughput and Disk I/O) associated with Xen virtualization. Given the current technology level and development progress, this paper (2) identifies applicable areas for virtualization-based operation support systems and addresses suggested system architecture.

This direct attach storage (DAS) paradigm dates back to the early days of mainframe computing, when disk drives were located close to the CPU and cabled directly to the computer system via some control circuits. Computing evolved from being computation-centric to storage-centric. The decentralization of departmental computing into networked individual workstations requires efficient sharing of data. Storage subsystems, basically a collection of disk drives, and perhaps some other backup storage devices such as tape and optical disks that can be managed together, evolved out of necessity. There are two orthogonal aspects of storage subsystems that are discussed. One aspect has to do with how multiple drives within a subsystem can be organized together, cooperatively, for better reliability and performance. A second aspect deals with how a storage subsystem is connected to its clients and accessed. Some form of networking is usually involved. The storage subsystem software understands what organization is being used and knows how to map the logical addresses of the virtual disk to the addresses of the underlying physical devices. This concept is described as virtualization in the storage community. The technique of using data striping as an organization is purely for improving performance.

Publisher Summary This chapter provides information on key storage terminology and concepts related to Microsoft Exchange 2007. The goal of storage in any kind of computing environment is to have a place where data can be kept at rest. Data storing can be done randomly or sequentially. The mass media (disks and tapes) on which data is stored varies depending on the density, speed, and frequency of access. The chapter defines several terms related to data storage including capacity measurement unit Gigabyte (GB), data rate measurement unit Megabyte per second (MB/s), request rate measurement unit I/O per second (IOPS), and response time. For Exchange 2007, one needs to provide online storage that is reliable and fast. The storage speed for Microsoft Exchange depends on the activity one performs on the server. For regular online access to email, one needs to have a good I/O throughput (request rate). The size of the I/O is typically that of a database page, set to 8 KB with Exchange 2007 (with previous version of Microsoft Exchange, this was 4 KB). It is required to connect CPU (or CPUs) to disks by the means of storage controller, because a CPU accesses data directly from RAM, but certainly not directly to disk. RAID, used for redundant array of independent disks, is a technique that combines several independent disk drives into a single logical volume. RAID comes in many levels, with the base levels ranging from 0 to 5. Some storage controllers implement virtualization, which can be implemented at various levels including disk level, controller level, and network. Storage networks are inevitable in large enterprises to share the benefits of a strong storage environment among different servers (or storage consumers). Storage networks typically come in two modes: file-level access and block-level access.

iConnect is an abstraction that encapsulates all of a virtual machine's interactions with outside entities. Its intent is to exploit semantic information to better support the end-to-end requirements of such VM communications. Focusing on the I/O performed by VMs and leveraging the fact that modern systems already have to virtualize the physical devices used by VMs, this paper shows that the iConnect abstraction can be implemented with the extension of existing virtual device interfaces. Specifically, by devising enhanced virtual devices, we can (1) efficiently implement the communication paths between virtual machines (VMs) and the virtualized platforms (VPs) on which they run, and (2) capture semantic information about VM-device interactions, which can then be used to implement additional functionality and efficient sharing of physical devices. The paper presents three concrete realizations of the iConnect abstraction: a multimedia device virtualization solution (VMedia) which utilizes semantic information to implement efficient sharing and enhanced functionality, a network virtualization solution that provides virtual NICs with QoS-support where the VM communicates its QoS requirements to the VP, and a storage virtualization solution which permits a VM to access a block device regardless of whether such a device is physically located locally or must be accessed at a remote location. A Xen-based implementation of the iConnect concept demonstrates substantial performance improvements and additional functionality derived from their use at a minimal cost to VMs, in part because iConnect utilizes additional computational resources of the VP and can take better advantage of certain underlying platform capabilities.

In this paper we describe specific challenges in virtualizing a wireless network and multiple strategies to address them. Among different possible wireless virtualization strategies, our current work in this domain is focussed on a Time-Division Multiplexing (TDM) approach. Hence, we we present our experiences in the design and implementation of such TDM-based wireless virtualization. Our wireless virtualization system is specifically targeted for multiplexing experiments on a large-scale 802.11 wireless testbed facility.

We present a packet forwarding method for the implementation of the iSCSI virtualization switch used in the SANs. Using the existing virtualization software plus network protocol stacks, first we build up a network-based storage virtualization switch model in the architecture of the symmetric approach. In compliance with the same functionality, our method can reduce the overheads of protocol processing by using a packet forwarding method based on caching the structure ID of the iSCSI session. With the Address Translator and Duplicating Handler, our design can also achieve the feature of storage virtualization management. From the result of our performance experiment, our forwarding system can achieve a higher performance on the READ/WRITE operations by 25% ~30% and reduce the CPU utilization by 10% ~15% compared to the conventional virtualization switch.

Direct-attached storage has historically had the reputation of being less capable than equivalently sized SAN installations. Here, we empirically demonstrate the performance achievable in multiple-terabyte, direct-attached disk subsystems. A number of parameters are explored, including file system, number of logical drives, and RAID configuration.

Storage virtualization can be implemented at three layers: server-layer, storage-device-layer and network-layer. iSCSI-based storage virtualization is to implement storage virtualization at network-layer, and the implementation includes three sides: host, manager, and store. Manager manages some stores and generates a unified, logical store pool to provide store services to host. According the method of forwarding at the manager side, iSCSI-based store virtualization was implemented in two ways: in-band and out-of-band. Manager must transmits data stream and command stream to host and store in in-band way. After the two ways of implementation, the performance test results indicated that the performance of out-of-band was better than in-band.

Advances in server, network, and storage virtualization are enabling the creation of resource pools of servers that permit multiple application workloads to share each server in the pool. This paper proposes and evaluates aspects of a capacity management process for automating the efficient use of such pools when hosting large numbers of services. We use a trace based approach to capacity management that relies on i) a definition for required capacity, ii) the characterization of workload demand patterns, iii) the generation of synthetic workloads that predict future demands based on the patterns, and iv) a workload placement recommendation service. A case study with 6 months of data representing the resource usage of 139 workloads in an enterprise data center demonstrates the effectiveness of the proposed capacity management process. Our results show that when consolidating to 8 processor systems, we predicted future per-server required capacity to within one processor 95% of the time. The approach enabled a 35% reduction in processor usage as compared to today's current best practice for workload placement.

Distributed storage systems are expected to serve a broad spectrum of applications, satisfying various requirements with respect to capacity, speed, reliability, security at low cost. Virtualization techniques allow flexible configuration of storage systems in order to meet resource constraints and application requirements. Violin provides block level virtualization that enables the extension of storage with new mechanisms and combining them to create modular hierarchies. Creating and maintaining such virtualization hierarchies however, is a complex task where a human system administrator is the most expensive and less efficient element. We introduced Conductor, an automated support system that tries to grasp human expertise with declarative rules that are applied to storage management. So far the initial, static configuration capabilities of Conductor have been elaborated. Static features however, are not sufficient for practical purposes as the storage system evolves, i.e. requirements, workloads, access patterns may change in time. This paper presents work in progress that is aimed at extending Conductor with supporting dynamic features. We introduce the concepts of global and directed reconfigurations and discuss their potential strengths and weaknesses.

Virtualized storage systems are required to service concurrently executing workloads, with potentially diverse data delivery requirements, that are running under multiple operating systems. Although a number of algorithms have been developed for I/O performance virtualization among operating system (OS) instances and their applications, none results in absolute performance virtualization. By absolute performance virtualization we mean that the performance experienced by applications of one operating system does not suffer due to variations in the I/O request stream characteristics of applications of other operating systems. Key requirements of I/O performance virtualization are fairness and performance isolation. In this paper, we present a novel virtual I/O scheduler (VIOS) that provides absolute performance virtualization by being fair in sharing I/O system resources among operating systems and their applications, and provides performance isolation in the face of variations in the characteristics of I/O streams. The VIOS controls the coarse grain allocation of disk time to the different operating system instances and is OS independent; optionally, a set of OS-dependent schedulers may determine the fine-grain interleaving of requests from the corresponding operating systems to the storage system.

In this paper, we present the design of a performance isolation benchmark that quantifies the degree to which a virtualization system limits the impact of a misbehaving virtual machine on other well-behaving virtual machines running on the same physical machine. Our test suite includes six different stress tests - a CPU intensive test, a memory intensive test, a disk intensive test, two network intensive tests (send and receive) and a fork bomb. We describe the design of our benchmark suite and present results of testing three flavors of virtualization systems -- an example of full virtualization (VMware Workstation), an example of paravirtualization (Xen) and two examples of operating system level virtualization (Solaris Containers and OpenVZ). We find that the full virtualization system offers complete isolation in all cases and that the paravirtualization system offers nearly the same benefits -- no degradation in many cases with at most 1.7% degradation in the disk intensive test. The results for operating system level virtualization systems are varied -- illustrating the complexity of achieving isolation of all resources in a tightly coupled system. Our results highlight the difference between these classes of virtualization systems as well as the importance of considering multiple categories of resource consumption when evaluating the performance isolation properties of a virtualization system.

We investigate methodologies for placement and migration of logical data stores in virtualized storage systems leading to optimum system configuration in a dynamic workload scenario. The aim is to optimize the tradeoff between the performance or operational cost improvement resulting from changes in store placement, and the cost imposed by the involved data migration step. We propose a unified economic utility based framework in which the tradeoff can be formulated as a utility maximization problem where the utility of a configuration is defined as the difference between the benefit of a configuration and the cost of moving to the configuration. We present a storage management middleware framework and architecture Compass that allows systems designers to plug-in different placement as well as migration techniques for estimation of utilities associated with different configurations. The biggest obstacle in optimizing the placement benefit and migration cost tradeoff is the exponential number of possible configurations that one may have to evaluate. We present algorithms that explore the configuration space efficiently and compute a candidate set of configurations that optimize this cost-benefit tradeoff. Our algorithms have many desirable properties including local optimality. Comprehensive experimental studies demonstrate the efficacy of the proposed framework and exploration algorithms, as our algorithms outperform migration cost-oblivious placement strategies by up to 40% on real OLTP traces for many settings.

Virtualization has seen a lot of adoption, as it allows better usage of the data center resources. Although we have virtualized systems, we need a better means to visualize the virtualized services on the system and underlying physical and logical dependencies. In the absence of such visualization tools it is clear that the total cost of ownership is expected to be higher because of the time it would take to troubleshoot a problem and analyze impacts of failures. Unfortunately, today there is no standard way to obtain information about these relationships. Entity MIB(RFC-2737) is a standard SNMP MIB that is well understood and is supported by most devices. In this paper, a methodology is proposed, to use Entity MLB to provide information about the virtualized entities.

Snapshot technology is becoming prevalent to perform data protection and other tasks such as data mining and data cloning. To improve the performance and reliability of the traditional Linux LVM snapshot, we propose a novel cached dependent snapshot system, ESnap. ESnap decreases the total amount of data copy effectively using the data dependency among snapshot volumes. A new snapshot metadata organization scheme is designed to support massive snapshot volumes and the corresponding read/write algorithms are put forward. Also the automatic extending of snapshot volumes is implemented in ESnap to avoid the failure of the whole dependent snapshot chain due to space overflow of one snapshot. experimental results show that ESnap has higher performance, reliability and the resource utilization rate than traditional Linux LVM snapshot system.

We present a randomized block-level storage virtualization for arbitrary heterogeneous storage systems that can distribute data in a fair and redundant way and can adapt this distribution in an efficient way as storage devices enter or leave the system. More precisely, our virtualization strategies can distribute a set of data blocks among a set of storage devices of arbitrary non-uniform capacities so that a storage device representing x% of the capacity in the system will get x% of the data (as long as this is in principle possible) and the different copies of each data block are stored so that no two copies of a data block are located in the same device. Achieving these two properties is not easy, and no virtualization strategy has been presented so far that has been formally shown to satisfy fairness and redundancy while being time- and space-efficient and allowing an efficient adaptation to a changing set of devices.

This tutorial presents the concepts that lay behind the most successfully applied ideas to manage heterogeneous storage devices in cluster environments. Among these ideas we can point out storage virtualization that can either be coupled with deterministic mechanisms to optimize heterogeneous disk usage or with randomization techniques. Focusing on the concepts instead on the real implementations and/or systems (like it is normally done) has the advantage that the attendees get a much clearer view of the ideas presented and enables the attendees to transfer the ideas to their own fields.

This chapter introduces the core concepts of server, storage, and network virtualization as a foundation for learning more about Xen. The chapter illustrates the potential benefits of virtualization to any organization. Virtualization is an abstraction layer that breaks the standard paradigm of computer architecture, decoupling the operating system from the physical hardware platform and the applications that run on it. Virtualization allows multiple virtual machines to run in isolation side-by-side on the same physical machine. Each virtual machine has its own set of virtual hardware (CPU, memory, network interfaces, and disk storage) upon which an operating system and applications are loaded. Virtualization technology and its core components, such as the Virtual Machine Monitor manage the interaction with the operating system calls to the virtual hardware and the actual execution that takes place on the underlying physical hardware. Today, virtualization is growing as a core technology in the forefront of data center management. The technology is helping businesses, both large and small to solve their problems with scalability, security, and management of their global IT infrastructure.

Publisher Summary Virtualization software that presents a virtual set of hardware to a guest operating system makes up the majority of server virtualization products available. These virtualization products provide a VMM that either partially or fully virtualizes the underlying hardware, allowing both modified and unmodified guests to run in a safe and isolated fashion. The most popular of these products, especially for the enterprise, are VMware, Microsoft, and Xen, in order of commercial market share. Operating system-level virtualization software is either included as part of an operating system, such as Solaris containers, or is installed on top of an operating system, such as Virtuozzo and OpenVZ. In a few ways, operating system-level virtualization software products provide some advantages over hardware virtualization software. First, they can usually run the same operating system that the host is running without having to duplicate every layer and subsystem. Both the open source Xen hypervisor and the commercial XenServer product family from XenSource have also been included in the list of List of Hardware Virtualization Software.

International audience; Virtualization tools are becoming popular in the context of Grid Computing because they allow running multiple operating systems on a single host and provide a confined execution environment. In several Grid projects, virtualization tools are envisioned to run many virtual machines per host. This immediately raises the issue of virtualization scalability. In this paper, we compare the scalability merits of Four virtualization tools. First, from a simple experiment, we motivate the need for simple microbenchmarks. Second, we present a set of metrics and related methodologies. We propose four microbenchmarks to measure the different scalability parameters for the different machine resources (CPU, memory disk and network) on three scalability metrics (overhead, linearity and isolation). Third, we compare four virtual machine technologies (Vserver, Xen, UML and VMware). The results of this study demonstrate that all the compared tools present different behaviors with respect to scalability, in terms of overhead, resource occupation and isolation. Thus this work will help user to select tools according to their application characteristics.

A great deal of popular software is not designed for mobility (Griffiths, 2004). This is peculiar because many mobile users expect to have easy access to an information infrastructure that links up their mobile phone, laptop, personal digital assistant (PDA), and other devices, while the backend systems of organizations need to be agile, especially as the number, range, and diversity of services and associated technologies grow. Enter virtualization, a technology that has been part of computing for many years, but only fairly recently become mainstream (Intel, 2006; Singh, 2004).

Virtualization is defined as a framework or methodology of dividing the resources of computer hardware into multiple execution environments, by applying one or more concepts or technologies such as hardware and software partitioning, time-sharing, partial or complete machine simulation, emulation, quality of service, and many others. The objectives of any virtualization technology include adding a layer of abstraction between the applications and the hardware, enabling a reduction in costs and complexity, providing the isolation of computer resources for improved reliability and security, and improve service levels and the quality of service, etc. Software, referred to as a virtual machine monitor (VMM), controls use and access to the CPU, memory, storage, and network resources underneath. Server virtualization is the more dominant form of virtualization in use today. Implementations of server virtualization exist on, and for all, CPU platforms and architectures, the most popular being the IA-32 or x86. In its most basic form, storage virtualization exists in the assembly of multiple physical disk drives, or spindles, into a single entity that is presented to the host server and operating system, such as with RAID implementations. The true storage tier and its components were further masked by the introduction and adoption of storage area network (SAN) technologies. The most popular forms of network virtualization are: virtual LAN (VLAN), virtual IP (VIP), and virtual private network (VPN). Application virtualization, or software virtualization, is the newest member of the ever-growing virtualization family. It is a new approach to software management.

In information warfare (IW), information storage is one of important processes to obtaining information dominance. Existing military storage systems are mostly independent, so they can't provide reliable and effective storage service for users. How to integrate existing military storage systems into a network storage system which have stronger compatibility, higher usability and better cooperation? It's the problem urgently need us to solve. The article makes a detailed introduction to technologies of storage virtualization (SV). As to its different realization methods, we analyze both their advantages and disadvantages. In the paper, the conception of integrating military storage resources based on SV is proposed. Moreover, some questions in processes of integration are researched.

This paper implements two technologies in a storage area network (SAN): pipelining scheduling and hierarchical volume management. Firstly, we design a pipelining-scheduling, which overlaps network transfer procedure and disk I/O procedure. Concretely, two kinds of pipelining scheduling schemes have been implemented: (1) Fixed pipelining scheduling (FPS) synchronizing the pipeline stages at fixed time point; (2) Variable pipelining scheduling (VPS) synchronizing the pipeline stages dynamically according to the completion sequence of multiple procedures. Secondly, the hierarchical logical volume management (HLVM) provides a kind of storage virtualization on the different layers. By decomposing the original SCSI address, it builds a virtual layer between the physical storage devices and the applications.

The ever growing storage demands of modern applications cultivate a promising area of storage system management, centered upon metadata servers. Some of the works are done at the file system level, which care little about device level data organization. In this paper we present the design of managing metadata at block-level by using storage virtualization techniques. We propose both static and dynamic block mapping policies to meet the diverse needs of applications, balance disk workload and improve storage system performance. We build our prototype in a Virtual Interface based storage area network, and the evaluation results demonstrate that metadata management at block-level can be efficient and flexible.

As Intel is rolling out its Vanderpool processor virtualization technology and AMD its Secure Virtual Machine technology, we are only seeing the first wave of processor virtualization assists. Over the next few years the x86 space will change dramatically. We will see the introduction of massive multi-core, 64-bit, 2 nd generation processor virtualization capabilities, I/O isolation capabilities, and hardware security assists.Both Intel and AMD are differentiating their processors by providing enhancements that enable you to run multiple virtual machines in such a way that the guest is unaware that it is being virtualized. Ironically, largely because these technologies have been unavailable for so long, Linux and Windows are going into a different direction: paravirtualization. With paravirtualization the guest operating system collaborates closely with the virtual machine monitor through a set of well defined software interfaces. This approach does not require any new hardware features at all and has the potential of performing much better. So, this raises an interesting dilemma: Some of the new virtualization capabilities may already be obsolete before they are brought to market.In this talk I will discuss the new virtualization technologies that will be introduced over the next few years, how they help virtualization, what challenges they pose and how these virtualization technologies will likely consolidate.

The capability of making an immediate point-in-time copy of large data sets has become an essential feature for storage management solutions. These copies, also called snapshots, are required for backup purpose, recovery, and to work on real data for testing applications. Depending on an application’s needs, different capabilities, like one-level read-only snapshots or rewritable ones are required. Besides their functionality, snapshot implementations are also differentiated with respect to their location and how they enable different stages of access to the duplicated data. Significant problems, like preserving consistency, arise when using snapshots in distributed environments with multiple accessing servers, and in which the configuration might change at any time. This paper describes a useful way to integrate a snapshot implementation that provides multiple levels of data access. We introduce a consistency preserving protocol for the storage virtualization solution V:DRIVE and present benchmark results for various snapshot scenarios.

Block-level security provides the lowest level security for storage devices. In this paper, we propose a block-level security based on hierarchical logical volume of FC RAID. Using the hierarchical logical volume, a group of physical disks can be mapped to a uniform logical storage space and two encrypt/decrypt algorithms can be applied to different logic volumes.