One or more embodiments of the present invention relate to virtual machines and, more specifically, to methods for supporting virtual machines with respective units of storage.
In general, state data for a virtual machine may be encoded in some form of computer readable media. In some cases, storage local to an underlying hardware platform can be used. In some cases, storage array technology may be used to share pools of underlying storage amongst multiple computational systems. Regardless of the storage technologies used, file system constructs are typically employed to mediate access to stored information. Methods are desired whereby at least certain aspects of the virtual machine state may be encoded and accessed without much of the complexity and overhead usually associated with an intermediary file system.
Some embodiments of the present invention address one or more of the above-identified needs. In particular, one embodiment of the present invention includes a method comprising: accessing units of network storage that encode state data of respective virtual machines, wherein the state data for respective ones of the virtual machines are stored in distinct ones of the network storage units such that the state data for more than one virtual machine are not commingled in any one of the network storage units.
The use of the same reference symbols in different figures indicates similar or identical items.
In accordance with one or more embodiments of the present invention, methods for encoding data used by virtual machines allow for certain types or portions of virtual machine data to be encoded, for each such virtual machine, in a distinct unit (or set of units) of network storage. In some embodiments, these methods allow a virtualization system to expose data that encodes a virtual disk, while securing non-disk data that encodes virtual machine state, virtual hardware configuration and/or snapshot or checkpoint states. Typically, an encoding of virtual machine state includes backing state data corresponding to internal states of devices, memory and other system components virtualized by or for a given virtual machine. As such, read access to such state date (e.g., by a guest) may leak or compromise sensitive information, while write access may afford malicious code or users with an attractive vector for attack.
It has been discovered that by provisioning distinct units of storage for respective virtual machines, a virtualization system can provide management and operation of virtual machines without use of an intermediary file system to access virtual machine state data. Such an approach can greatly improve performance in some embodiments, and can provide functional benefits as well. One or more units of storage are allocated to one and only one virtual machine. Typically, an indication of the associated virtual machine is indicated in the storage units and an indication of the allocated units is indicated to a virtualization layer that supports the virtual machine. By allocating distinct shared storage units (or sets thereof) for individual virtual machines, or for particular aspects of individual virtual machines, some embodiments in accordance with the present invention facilitate management and operation of virtual machines at a granularity native to the storage architecture.
In some embodiments, a unit of storage can be prepared or presented in a manner that secures virtual machine state data encoded therein from undesired exposure to a guest application or operating system executing in coordination with the virtual machine, while still facilitating access (e.g., by a guest computation) to other data (e.g., virtual primary disk data) also encoded in the unit of storage. In general, undesired exposure of virtual machine state data to a guest of a virtual machine can risk corruption of the state data and, perhaps, other non-disk data. Whether such corruption is accidental or intentional, the corruption can be avoided by limiting guest access to only an appropriate subset of data encoded in the storage unit.
In some virtualization system embodiments, an appropriate subset includes data that is exposed to the guest computation as the information content of a disk virtualized by the virtual machine. In some embodiments, mutually exclusive units of storage are employed for each virtual machine supported in a virtualization system. Typically, in such embodiments, the storage allocated for a particular virtual machine is partitioned in such a way that the virtualization system may expose virtual disk data, while restricting access to non-disk data (e.g., virtual machine state data corresponding to internal states of devices, memory and other system components virtualized). Various techniques, such as encryption and/or use of offsets, may also be implemented to prevent circumvention of the restricted access. In some embodiments, partitions are organized (and partition information is manipulated) such that a reduced apparent size of storage that encodes both disk and non-disk data is presented to the guest and conceals non-disk data beyond an apparent extent of the storage.
In some embodiments, functionality of storage arrays can be leveraged in the implementation of virtual machine operations that manipulate an encoding of virtual machine state. For example, suitable coding techniques for virtual machine state can facilitate the use of snapshot, replicate, discovery, and other storage array functionality by a virtualization layer on a virtualization system or by a virtual machine manager to effectuate operations on, and/or manipulations of, virtual machine state. For example, by provisioning and managing virtual machine storage using interfaces (e.g., the block-level I/O typical of SAN technology) and/or granularity native to the storage environment (e.g., storage volume, logical unit number, etc.), some embodiments in accordance with the present invention can directly exploit storage system functionality in support of operations such as virtual machine migration, movement, cloning. In some embodiments, check pointing, rollback and even failover support can be efficiently supported using functionality of a storage array.
Based on the description herein, it will be apparent to persons of ordinary skill in the art that the term “non-disk data” refers to data that encodes information other than that exposed as a disk by a virtualization system or other similar software system, rather than to any exclusion of underlying storage technology. For concreteness of description, an example of “non-disk data” is backing state data corresponding to internal states of devices, memory and other system components virtualized by or for a given virtual machine. Similarly, an example of “disk data” is information exposed to a guest application or operating system by a virtualization system as a virtual primary (secondary, tertiary . . . ) disk. For avoidance of doubt, underlying encodings of both disk and non-disk data may reside in media that include or constitute disks.
The following describes embodiments in which virtual machine data is encoded in units of storage allocated from pooled network storage shared amongst virtualization systems. Although network storage units provide a useful descriptive context in which to illustrate embodiments, network storage is not essential. Rather, based on the description herein, persons of ordinary skill in the art will appreciate that embodiments of the present invention may be used in the context of other storage technologies and configurations to selectively expose certain portions of virtual machine state while securing and/or isolating other portions.
As used herein, the term network storage refers generally to storage systems and storage array technology, including storage area network (SAN) implementations, network attached storage (NAS) implementations, and other storage architectures that provide a level of virtualization for underlying physical units of storage. In general, such storage architectures provide a useful mechanism for sharing storage resources amongst computational systems. In some cases, computational systems that share storage resources may be organized as a coordinated system (e.g., as a cluster or cooperatively managed pool of computational resources or virtualization systems). For example, in a failover cluster it may be desirable to share (or at least failover) virtual machine access to some storage units. Similarly, in a managed collection of virtualization systems, it may be desirable to migrate or otherwise transition virtual machine computations from one virtualization system to another. In some cases, at least some computational systems may operate independently of each other, e.g., employing independent and exclusive units of storage allocated from a storage pool (or pools) provided and/or managed using shared network storage.
Generally, either or both of the underlying computer systems and storage systems may be organizationally and/or geographically distributed. For example, some shared storage (particularly storage for data replication, fault tolerance, backup and disaster recovery) may reside remotely from a computational system that uses it. Of course, as will be appreciated by persons of ordinary skill in the art, remoteness of shared storage is a matter of degree. For example, depending on the configuration, network storage may reside across the globe, across the building, across the data center or across the rack or enclosure.
While embodiments of the present invention, particularly cluster-organized and/or enterprise scale systems, may build upon or exploit data distribution, replication and management features of modern network storage technology, further embodiments may be used in more modest computational systems that employ network storage technology. For example, even a single computer system may employ SAN-type storage facilities in its storage architecture. Thus, while some embodiments utilize network storage that can be shared and while at least some underlying elements thereof may be remote, persons of ordinary skill in the art will understand that for at least some embodiments, network storage need not be shared or remote.
In some embodiments of the present invention, particularly those that use SAN-type storage arrays, block-level I/O access to virtual machine state data can afford performance advantages. Similarly, encapsulation and/or isolation techniques are described which may be employed in some encodings of virtual machine state data to limit access (e.g., by a guest application or operating system) to underlying data. Accordingly, certain embodiments are described in which non-commingled, encapsulated representations of virtual machine state are maintained in distinct storage volumes (or LUNs) of a SAN. Nonetheless, other embodiments, including those that use NAS-type or filesystem-mediated access mechanisms may still allow a virtualization system to leverage storage system functionality in support of operations such as virtual machine migration, movement, cloning, check pointing, rollback and/or failover using suitable codings of virtual machine state data.
For concreteness, embodiments are described which are based on facilities, terminology and operations typical of certain processor architectures and systems, and based on terminology typical of certain operating systems, virtualization systems, storage systems and network protocols and/or services. That said, the embodiments are general to a wide variety of processor and system architectures (including both single and multi-processor architectures based on any of a variety of instruction set architectures), to numerous operating system implementations and to systems in which both conventional and virtualized hardware may be provided. As described herein, the embodiments are also general to a variety of storage architectures, including storage virtualization systems such as those based on storage area network (SAN) or network attached storage (NAS) technologies.
Accordingly, in view of the foregoing and without limitation on the range of underlying processor, hardware or system architectures, operating systems, storages architectures or virtualization techniques that may be used in embodiments of the present invention are described. Based on these descriptions, and on the claims that follow, persons of ordinary skill in the art will appreciate a broad range of suitable embodiments.
Computational Systems, Generally
In the illustrated collection, storage area network (SAN) technology is used for at least some storage needs of computational systems participating in the collection. In general, network storage systems (including SAN-based system 2030) provide a level of virtualization for underlying physical storage elements (e.g., individual disks, tapes and/or other media), where the characteristics and/or configuration of particular storage elements may be hidden from the systems that employ the storage. SAN-based systems typically provide an abstraction of storage pools from which individual storage units or volumes may be allocated or provisioned for block level I/O access. In the illustrated collection, a switched fabric topology consistent with Fibre Channel SAN technology is shown in which switches 2031A, 2031B, 2031C and/or directors are used to mediate high bandwidth access (typically using a SCSI, Small Computer System Interface, command set) to an extensible and potentially heterogeneous set of storage resources 2032A, 2032B, 2032C, 2032D, 2032E, 2032F, 2032G, e.g., SATA (Serial ATA) and/or SCSI disks, tape drives, as well as arrays thereof (e.g., RAID, i.e., Redundant Array of Inexpensive Disks). Such resources may be distributed and (if desirable) may provide data replication and/or off-site storage elements. Fibre Channel is a gigabit-speed network technology standardized in the T11 Technical Committee of the InterNational Committee for Information Technology Standards (INCITS).
In general, a variety of different types of interconnect entities, including, without limitation, directors, switches, hubs, routers, gateways, and bridges may be used in topologies (or sub-topologies) that include point-to-point, arbitrated loop, switched fabric portions. Fibre Channel and non-Fibre Channel technologies including those based on iSCSI protocols (i.e., SCSI command set over TCP/IP) or ATA-over-ethernet (AoE) protocols may be used in embodiments of the present invention. Similarly, any of a variety of media including copper pair, optical fiber, etc. may be used in a network storage system such as SAN 2030.
Although not specifically illustrated in
Embodiments of the present invention may be understood in the context of virtual machines 2013 (or virtual computers) that are presented or emulated within a virtualization system such as virtualization system 2012 executing on underlying hardware facilities 2015. However, in addition, migration from (or to) a computational system embodied as a conventional hardware-oriented system may be supported in some systems configured in accordance with the present invention. Nonetheless, for simplicity of description and ease of understanding, embodiments are described in which individual computational systems are embodied as virtualization systems that support one or more virtual machines.
Although certain virtualization strategies/designs are described herein, virtualization system 2012 is representative of a wide variety of designs and implementations in which underlying hardware resources are presented to software (typically to operating system software and/or applications) as virtualized instances of computational systems that may or may not precisely correspond to the underlying physical hardware.
Overview of Virtualization Systems
The term virtualization system as used herein refers to any one of an individual computer system with virtual machine management functionality, a virtual machine host, an aggregation of an individual computer system with virtual machine management functionality and one or more virtual machine hosts communicatively coupled with the individual computer system, etc. Examples of virtualization systems include commercial implementations, such as, for example and without limitation, VMware® ESX Server™ (VMware and ESX Server are trademarks of VMware, Inc.), VMware® Server, and VMware® Workstation, available from VMware, Inc., Palo Alto, Calif.; operating systems with virtualization support, such as Microsoft® Virtual Server 2005; and open-source implementations such as, for example and without limitation, available from XenSource, Inc.
As is well known in the field of computer science, a virtual machine (VM) is a software abstraction—a “virtualization”— of an actual physical computer system. Some interface is generally provided between the guest software within a VM and the various hardware components and devices in the underlying hardware platform. This interface—which can generally be termed “virtualization layer”—may include one or more software components and/or layers, possibly including one or more of the software components known in the field of virtual machine technology as “virtual machine monitors” (VMMs), “hypervisors,” or virtualization “kernels.”
Because virtualization terminology has evolved over time and has not yet become fully standardized, these terms (when used in the art) do not always provide clear distinctions between the software layers and components to which they refer. For example, the term “hypervisor” is often used to describe both a VMM and a kernel together, either as separate but cooperating components or with one or more VMMs incorporated wholly or partially into the kernel itself. However, the term “hypervisor” is sometimes used instead to mean some variant of a VMM alone, which interfaces with some other software layer(s) or component(s) to support the virtualization. Moreover, in some systems, some virtualization code is included in at least one “superior” VM to facilitate the operations of other VMs. Furthermore, specific software support for VMs is sometimes included in the host OS itself.
Embodiments are described and illustrated herein primarily as including one or more virtual machine monitors that appear as separate entities from other components of the virtualization software. This paradigm for illustrating virtual machine monitors is only for the sake of simplicity and clarity and by way of illustration. Differing functional boundaries may be appropriate for differing implementations. In general, functionality and software components/structures described herein can be implemented in any of a variety of appropriate places within the overall structure of the virtualization software (or overall software environment that includes the virtualization software).
Virtual Machine Monitor (VMM)
In view of the above, and without limitation, an interface usually exists between a VM and an underlying platform which is responsible for executing VM-issued instructions and transferring data to and from memory and storage devices or underlying hardware. A VMM is usually a thin piece of software that runs directly on top of a host, or directly on the hardware, and virtualizes at least some of the resources of the physical host machine. The interface exported to the VM is then the same as the hardware interface of a physical machine. In some cases, the interface largely corresponds to the architecture, resources and device complements of the underlying physical hardware; however, in other cases it need not.
The VMM usually tracks and either forwards to some form of operating system, or itself schedules and handles, all requests by its VM for machine resources, as well as various faults and interrupts. An interrupt handling mechanism is therefore included in the VMM. As is well known, in the Intel IA-32 (“x86”) architecture, such an interrupt/exception handling mechanism normally includes an interrupt descriptor table (IDT), or some similar table, which is typically a data structure that uses information in the interrupt signal to point to an entry address for a set of instructions that are to be executed whenever the interrupt/exception occurs. In the Intel IA-64 architecture, the interrupt table itself contains interrupt handling code and instead of looking up a target address from the interrupt table, it starts execution from an offset from the start of the interrupt when a fault or interrupt occurs. Analogous mechanisms are found in other architectures. Based on the description herein, interrupt handlers may be adapted to correspond to any appropriate interrupt/exception handling mechanism.
Although the VM (and thus applications executing in the VM and their users) cannot usually detect the presence of the VMM, the VMM and the VM may be viewed as together forming a single virtual computer. They are shown and described herein as separate components for the sake of clarity and to emphasize the virtual machine abstraction achieved. However, the boundary between VM and VMM is somewhat arbitrary. For example, while various virtualized hardware components such as virtual CPU(s), virtual memory, virtual disks, and virtual device(s) including virtual timers are presented as part of a VM for the sake of conceptual simplicity, in some virtualization system implementations, these “components” are at least partially implemented as constructs or emulations exposed to the VM by the VMM. One advantage of such an arrangement is that the VMM may be set up to expose “generic” devices, which facilitate VM migration and hardware platform-independence. In general, such functionality may be said to exist in the VM or the VMM.
It is noted that while VMMs have been illustrated as executing on underlying system hardware, many implementations based on the basic abstraction may be implemented. In particular, some implementations of VMMs (and associated virtual machines) execute in coordination with a kernel that itself executes on underlying system hardware, while other implementations are hosted by an operating system executing on the underlying system hardware and VMMs (and associated virtual machines) executed in coordination with the host operating system. Such configurations, sometimes described as “hosted” and “non-hosted” configurations, are illustrated in
OS Hosted Virtual Computers
Non-OS Hosted Virtual Computers
Different systems may implement virtualization to different degrees—“virtualization” generally relates to a spectrum of definitions rather than to a bright line, and often reflects a design choice in respect to a trade-off between speed and efficiency and isolation and universality. For example, “full virtualization” is sometimes used to denote a system in which no software components of any form are included in the guest other than those that would be found in a non-virtualized computer; thus, the OS guest could be an off-the-shelf, commercially available OS with no components included specifically to support use in a virtualized environment.
Para-Virtualization
Another term, which has yet to achieve a universally accepted definition, is “para-virtualization.” As the term implies, a “para-virtualized” system is not “fully” virtualized, but rather a guest is configured in some way to provide certain features that facilitate virtualization. For example, the guest in some para-virtualized systems is designed to avoid hard-to-virtualize operations and configurations, such as by avoiding certain privileged instructions, certain memory address ranges, etc. As another example, many para-virtualized systems include an interface within the guest that enables explicit calls to other components of the virtualization software. For some, the term para-virtualization implies that the OS guest (in particular, its kernel) is specifically designed to support such an interface. According to this definition, having, for example, an off-the-shelf version of Microsoft Windows XP as the OS guest would not be consistent with the notion of para-virtualization. Others define the term para-virtualization more broadly to include any OS guest with any code that is specifically intended to provide information directly to the other virtualization software. According to this definition, loading a module such as a driver designed to communicate with other virtualization components renders the system para-virtualized, even if the OS guest as such is an off-the-shelf, commercially available OS not specifically designed to support a virtualized computer system.
Unless otherwise indicated or apparent, virtualized systems herein are not restricted to use in systems with any particular “degree” of virtualization and are not to be limited to any particular notion of full or partial (“para-”) virtualization.
Provisioning in a Virtualization System Supported with Network Storage
Provisioning Integration
Network storage systems provide virtualized storage where the physical structure of the storage (e.g., individual disks, tapes or other media) is hidden from the user. Network storage systems provide abstractions of storage pools and/or storage units to manage the physical storage. For example, physical storage is often organized into storage pools, possibly in the form of RAID groups/sets. Storage pools can then be subdivided into storage units, which are then exposed to computer systems (e.g., a SCSI LUN on a SAN communicating via Fibre Channel, iSCSI, etc.). In some environments, storage pools may be nested in a hierarchy, where pools are divided into sub-pools.
In the illustrated configuration, virtualization system 105 supports virtual machines 102A-102B, and virtualization system 103 supports virtual machines 104A-104B. Although virtual machines 102A and 104A are depicted in
Storage abstractions provided by network storage systems can be directly used by the virtual machine management system to integrate virtual machine provisioning and storage provisioning. For example, one or more storage pools may be assigned to a virtual machine manager. A virtual machine provisioning operation will automatically provision storage units from the assigned storage pools. The degree of exposure of a network storage system to virtual machine management may vary based on needs/structure of a particular entity. The degree of integration of facilities of a network storage system and a virtualization system will vary in accordance with the degree of exposure. A small entity, for example, may wish to task a single group with management of both virtualization systems and storage systems. Full exposure of the raw storage devices can allow for full integration of virtualization management and storage management. Full exposure of raw storage devices to a virtualization management system allows the virtualization management system to organize the raw storage devices into logical units. In such a scenario, the virtualization management system may create the abstraction of storage units directly upon the raw storage devices, or create additional layers of abstraction, such as storage pools.
In another scenario, an entity may outsource management of storage devices or utilize a third-party vendor that provides access to raw storage devices while managed by the third party vendor allowing the entity to retain some degree of control of the raw storage devices. In this third-party scenario, the entity responsible for management of the raw storage devices may expose storage pools to the entity managing virtual machines. Hence, the raw source devices are hidden from the virtual machine management entity, but are provided some latitude in management of the storage. This allows the virtual machine management entity greater ability to tailor the storage to their needs.
A large entity may have an established division for management of their storage devices separate from a division for virtualization. Such a large entity may wish to maintain distinct separation between the divisions. Hence, the virtualization division would only have exposure to units of the network storage as allocated by the storage device division, perhaps as requested by the virtualization division. However, as already stated, some degree of integration of network storage management and virtual machine management satisfies some calls for efficiency.
At block 202, the storage pool is selected. At block 203, network storage units are created in the selected storage pool. The network storage units are created in accordance with configuration settings that indicate, for example, a given size for each network storage unit. The size may be a default size, a discrete size selected from multiple options, user defined size, etc. At block 205, the created network storage units are designated for virtual machine data. A virtual machine manager may mark each of the created network storage units with a value identifying the network storage units as supporting virtual machines. Marking may be implemented by writing a partition type into a root partition for a network storage unit that has been associated with virtual machines. However, it is possible that a partition type may be overloaded with multiple associations. In addition to, or instead of, partition type marking, a unique value may be written into the network storage unit. For example, a checksum may be coded into the virtual machine data stored in a network storage unit. A discovery operation would then search for both the partition type and the checksum to verify that the network storage unit has been designated for virtual machines. Indication of whether a network storage unit supports a virtual machine may be maintained in a data structure separate and distinct from the network storage units in addition to or instead of marking the created network storage units.
To help illustrate, the following example configuration is provided, which refers to an industry standard Storage Management Initiative Specification (SMI-S). Storage array vendors typically provide APIs for third party management. One example is the industry standard SMI-S, which describes vendor neutral interfaces to manage storage arrays. A virtual machine manager may configure storage (i.e., provision logical representations of raw storage) for virtual machine provisioning automatically or manually. If automatic configuration is implemented, then the virtual machine manager performs Service Location Protocol (SLP) discovery for SNIA SMI-S registered profiles. The virtual machine manager displays a list of discovered network storage in a user-interface and prompts a user to enter a user/password. After the user enters a valid name/password for the discovered network storage, the virtual machine manager authenticates. If manual configuration is implemented, a user enters an IP address (optional port number) and name/password for the network storage management interface. With the entered information, the virtual machine manager authenticates. After authenticating, the virtual machine manager connects to the network storage unit and discovers available storage pools. From the available discovered storage pools, the virtual machine manager divides the pools into primordial pools and data store pools, from which network storage units are created. The virtual machine manager enumerates units (e.g., logical unit numbers (LUNs)) of the network storage from the storage pools, and compares the enumerated units against a list of in-band discovered LUNs across applicable hosts. The virtual machine manager displays root/boot LUNs, known virtual machine LUNs, and all other LUNs separately.
Referring again to
Of course, creation and designation of network storage units for virtual machines does not necessitate immediate or contemporaneous creation of a virtual machine or communication of a network storage unit to the virtualization system. In general, the division of labor between a virtual machine manager and a virtualization layer of a virtualization system represents an implementation choice and implementations can vary greatly. For example, a virtual machine manager may only perform storage pool discovery and communicate discovered storage pools to the virtualization layer of a virtualization system. The virtualization layer may be responsible for creation of network storage units, designation of the created network storage units for virtual machines, and eventual provisioning of a virtual machine. In another example, a virtual machine manager may perform all operations from storage unit discovery (perhaps even storage unit creation), to provisioning and activation of a virtual machine. In this embodiment, a virtualization layer is tasked with the responsibility of operating and exposing resources to supported virtual machines executing in coordination therewith. Regardless of the division of labor in a particular embodiment, provisioning of a virtual machine involves preparation of a corresponding network storage unit. Preparation involves formatting of a network storage unit, and possibly initialization of the network storage unit with certain data for a virtual machine. Preparation of network storage unit may be entirely performed by a virtual machine manager or a virtualization layer, collectively performed by a virtual machine manager and a virtualization layer, alternately performed, etc.
Mutually Exclusive Network Storage Units for Individual Virtual Machines
Although the above depicts allocation and preparation of a single unit of network storage for each virtual machine, multiple units of network storage may be allocated and prepared for a given virtual machine. Virtual machine provisioning can range from completely encapsulating a single virtual machine in a single network storage unit (i.e., storing state data, virtual primary disk data, virtual hardware configuration data, snapshot data, virtual secondary disk data, etc.) to allocating individual network storage units for each aspect of a virtual machine (e.g., a network storage unit for each of virtual hardware configuration data, state data, virtual primary disk data, boot sector, etc.). In light of these variations, the network storage unit that backs state data of a virtual machine is referred to herein as the base unit, whether or not the unit also encodes additional data.
At block 311, one or more additional units of network storage are allocated for one or more secondary virtual disks. At block 313, the additional network storage units for the secondary disks are indicated. Various techniques can be utilized to indicate the additional units of network storage allocated for the virtual secondary disks of the virtual machine. For example, indications may be written into the base unit to identify the additionally allocated units of network storage as virtual secondary disks (e.g., the base unit will identify a globally unique ID, such as SCSI Inquiry VPD page 0x83, for a secondary virtual disk). The globally unique ID will allow the virtual secondary disks to be automatically detected and attached to the virtual machine. The indications of the additional network storage units as virtual secondary disks may alternatively, or in addition, be communicated to the virtualization system. The secondary virtual disk network storage units may also be marked to distinguish them from base units. Control flows from block 313 to block 317. At block 317, the base unit is indicated to the virtualization system at least for backing state data of the virtual machine.
For a given embodiment, the degree of separation of virtual machines across network storage units can range from only separating state data for virtual machines to separating all data for virtual machines. For example, a first network storage unit may encode a library of possible virtual hardware configuration. While the state data of two distinct virtual machines would be backed on separate network storage units, these two distinct virtual machines may point to the same or different virtual hardware configuration data in the library. In another example, two distinct virtual machines may share a virtual secondary disk on a network storage unit while their respective virtual primary disks and state data are stored on mutually exclusive network storage units.
In general, prevention of commingling of state data of the state virtual machines may be enforced implicitly or explicitly. For example, an embodiment of a virtual machine manager may not present a network storage unit already encoding state data of a virtual machine during virtual machine provisioning. However, in some cases, a given embodiment of a virtual machine manager may preserve the availability for provisioning multiple virtual machines onto a single network storage unit, perhaps for performance comparison purposes, dwindling resources, etc. Of course, the preserved option would involve mounting a file system as intermediary between a virtual machine and a network storage system, thus losing integration of the virtualization system in the network storage system and manageability of virtual machines at network storage unit granularity.
As mentioned above, data that encodes a virtual machine typically includes multiple data components: state data, virtual primary disk (sometimes referred to as virtual primary disk data), virtual hardware configuration data (e.g., type of processor, type of virtual network card, type of virtual storage host bus adapter (HBA), amount of memory, etc.), snapshot data, and zero or more virtual secondary disks (sometimes referred to as virtual secondary disk data). Those of ordinary skill in the art should appreciate that in various embodiments these multiple data components may be aggregated together, separated differently across different embodiments, further divided, etc. The state data of a virtual machine indicates execution state of a virtual machine at a particular time whether suspended or not suspended. For example, state data indicates current data in all or a portion of a memory of a virtual machine (e.g., instruction and/or value data in the virtual machine's RAM, cache, registers, etc.). A boot disk (e.g., a boot sector and OS disk image) may reside on the primary virtual disk, virtual secondary disks, or not be present at all for “network boot” of virtual machines. Virtual hardware configuration data indicates a configuration of a virtual machine. For example, virtual hardware configuration data indicates a type of virtual processor, type of virtual network card, type of virtual storage HBA, amount of virtual RAM, virtual chipset, type and size of a virtual primary disk, etc.
The virtual machine manager or virtualization layer of the corresponding virtualization system should allocate enough space in the state data portion of a network storage unit to support the configured memory size of the virtual machine. To support memory growth in the future, the virtual machine manager or virtualization layer (provisioning functionality) may choose to over-allocate space to accommodate the growth. The provisioning functionality may later change a virtual disk size or the supported virtual RAM size by extending the unit of network storage, updating a top-level partition data structure, and moving the data around (e.g., to the end of the network storage unit in the data structure outlined above). If the virtual disk size is being expanded, the virtual disk exposed to a guest will now report a larger size. If RAM is being added, the state data region will be increased. The state data region and other regions are typically small compared to virtual disks, so this reconfiguration should be a relatively fast operation. The provisioning computation may also choose to use a fragmented model without relocating regions when growing/shrinking the network storage unit.
Regardless of the particular embodiment, allocating mutually exclusive network storage units for distinct virtual machines dispenses with a traditional file system to store data components of the virtual machine. Dispensing with the traditional file system allows, for example, a storage virtualization engine to communicate directly with a network storage unit manager (assuming a virtual machine stack that comprises a virtual machine, a storage virtualization engine (i.e., instantiated code that presents virtual storage to a virtual machine), a network storage manager (e.g., a SAN agent), and a communication medium interface). The intermediary file system that would reside between the storage virtualization engine and the network storage manager has been obviated. With at least mutually exclusive network storage units backing state data of a virtual machine, the entire virtual machine state is encapsulated in a single base unit.
Securing Virtual Machine Data
Since a virtual machine has complete access to an entire network storage unit, a guest may also have access to the entire network storage unit. It would be prudent to restrict access to non-disk data. Exposing the state data, virtual hardware configuration data, or snapshot data can potentially leak sensitive information to a guest of the virtual machine. Actions by a guest, whether accidental or intentional, could result in the deletion or corruption of this data. Restricting guest access to non-disk data provides security and reliability. Furthermore, the potential harmful and/or undesirable exposure of non-disk data to a guest of a virtual machine is not limited to virtual machines with data stored in remote shared storage. For example, risks may also exist for a virtual machine with data stored in remote unshared storage, storage local to the hardware system that supports a given virtual machine (e.g., flash memory plugged into a USB port, a local disk, etc.).
To address this concern, non-disk data is isolated from disk data. A method for isolating disk data from non-disk data would be to limit guest access to a partition of a network storage unit that encodes a virtual primary disk, and zero or more partitions that encode virtual secondary disks. For example, the virtualization layer on a virtualization system, perhaps a virtual machine monitor in the virtualization layer, provides a virtualized SCSI disk to the virtual machine that only includes a virtual disk partition. To the virtual machine, this looks like a regular virtual disk with its own partition table and partitions. A guest of the virtual machine is no longer able to access the non-disk data.
Referring again to
Although the embodiment(s) described with reference to
Another method for restricting access to non-disk data hides the non-disk data from a guest while exposing a global partition table. For example, the non-disk data, whether or not organized into different partitions, are located at a location in a partition or storage known to the virtualization layer. The size of the storage, folder, or partition, is truncated when reported to a guest to hide the region occupied by non-disk data. Hiding the region occupied by non-disk data obviates modification of I/O requests with offsets.
Methods that align the beginning of a virtual disk with the beginning of a physical storage for a virtual machine can also be used to conceal non-disk data from a guest. A partition table for the virtual disk is stored in a non-disk data partition. When a guest requests access to the beginning of the storage, the second partition table is fetched and exposed to the guest. Attempted access by the guest to other than the beginning of the storage is allowed if not off the truncated end.
Operations of Virtual Machine Directly Supported by Network Storage
Access Requests from a Virtual Machine Guest
As described above, in some embodiments of the present invention, a virtual machine directly accesses data encoded in a unit of network storage rather than via an intermediate file system interface. To enforce security and restrict access by a guest of a virtual machine to at least state data, if not all non-disk data, the virtualization layer intercepts requests from a guest of a virtual machine.
For a response to an access request, the reverse of the operations depicted in
As previously stated, other techniques for concealing non-disk data from a guest may be implemented that avoid modifying requests with an offset. For example, storage size may be truncated when reported to a guest to conceal a region occupied by non-disk data. The previously described alignment technique also avoids modifying access requests from guests with offsets. Avoiding offset modification can be implemented with a less performance intensive set of operations (e.g., introducing a compare operation and a branch predict operation to the I/O path instead of a set of operations to compute and apply an offset).
Exposing Non-Disk Data Locations in a Network Storage Unit for Backup and Snapshot
Although non-disk data locations of the network storage unit are hidden from guests of a virtual machine, non-disk data locations are exposed to a corresponding virtualization system for certain management operations, such as backup operations and snapshot operations. A component of a virtualization layer (e.g., a network storage management agent) has direct access to a network storage unit. Direct access to a network storage unit allows state data to be written directly to a backing network storage unit for efficient backup of the state data. It is not necessary, however, for functionality that performs a write for backing state data to be resident in the virtualization layer or embodied as a separate agent. Embodiments may implement the backup writing functionality in the virtual system exposed to a virtual machine, in an agent that is separate from the virtual machine and the virtualization layer, etc.
Similarly, snapshot functionality may be implemented in the exposed virtual system, in the virtualization layer, in a separate agent, as an agent in the virtualization layer, etc. Regardless of the particular implementation, by exploiting direct access to a network storage system and an n:1 relationship between network storage units and virtual machines, a virtualization system can leverage facilities of a network storage system. In general, a network storage system application programming interface allows direct access by a virtualization system to snapshot facilities of a network storage management system. Allocation of mutually exclusive network storage units for distinct virtual machines allows one to use network storage management system facilities without the overhead from an intermediary file system.
Modern storage arrays typically provide the ability to snapshot a storage volume. A snapshot is typically implemented as copy-on-write implemented at the storage array layer where a second volume is created and linked to the first volume. To the client of the array, a snapshot results in a second volume that has the exact state of the original volume at the point the snapshot was taken. Arrays provide further capabilities such as refreshing a snapshot to bring it back into synchronization with the base volume, as well as severing a link between the two volumes to make clones. Since state data for a single virtual machine is completely stored in a base network storage unit, taking a snapshot for a virtual machine can be implemented with a write operation to the base unit and a network storage management system snapshot operation of the base unit, and, perhaps, any supplemental network storage units for the virtual machine.
Once this point is reached where a consistent state can be conveyed for the virtual machine, the virtualization layer triggers a snapshot of the base network storage unit to the linked snapshot network storage unit. This causes the underlying facilities of the network storage (e.g., network storage management system) to copy the data of the base network storage unit and store it as “snapshot 1” in the snapshot network storage unit. The virtualization also begins routing I/O to the snapshot network storage unit, which is written as current state data for the virtual machine. This process repeats as long as snapshot activation continues, storing a given number of snapshots in a different network storage unit.
In the context of arrays and volumes, the VM snapshot is now stored in a separate volume (or volumes if multiple disks are in use by the VM). Multiple VM snapshots can be supported by performing multiple snapshots on the array. In the state data portion of the base volume, references can be stored to the snapshot parent base volume, so the virtual machine manager can discover the parent child relationship between snapshots. In this approach, the copy-on-write is not implemented in the virtual machine or in a file system layer, but in the storage array layer.
Typically, storage arrays change the unique ID of the snapped volume to distinguish it from the base volume. Many operating systems, especially those with advanced multipath capabilities, use the unique ID of a volume as the ID for their disks instead of relying on the SCSI target and unit numbers. In a virtualization environment, the same ID for the virtual disk can be exposed to a guest(s) of a virtual machine; otherwise, drivers within an OS guest may become confused and may crash the OS. Within the state data portion of the base disk, or even a separate location, the “initial” unique ID for the volume(s) is recorded. When using snapped volumes, the virtualization layer will interpose and replace the snapped ID with the initial ID so that the OS guest is unaware that it is operating on a snapshot. Of course, this functionality is not limited to snapshotting the state data. A snapshot may be taken of the entire base unit, a very active portion of the state data (assuming there is a very active and less active state data portions), the primary virtual disk and state data, all disks and state data, etc.
Although an initial snapshot includes all data (e.g., all state data, all data in a base unit, etc.), subsequent snapshots may be taken in accordance with a variety of snapshotting techniques. For example, each snapshot in a series of snapshots may include all data at the particular moment of the snapshot, or only the difference between the current snapshot and a predecessor snapshot, which is not necessarily the root snapshot or immediately preceding snapshot. For example, a snapshot x may only include differences between itself and snapshot x−3. The snapshot may include a reference to the snapshot x−3, which may include another reference to the base snapshot. Another approach may track snapshot relationships in a table, linked list, etc. Those of ordinary skill in the art should appreciate that a variety of data structures may be utilized to encode snapshots and/or snapshot relationships.
At block 1305, the virtual machine is stopped. Although the virtual machine is stopped in the illustrated embodiment, embodiments in accordance with the present invention may suspend the virtual machine, thus preserving running state if so desired. In either case, the virtual machine state (in network storage) will be consistent and suitable for snapshot. At block 1307, the selected snapshot data is loaded from the corresponding network storage unit. At block 1309, it is determined whether additional corresponding snapshot data resides on other network storage units (e.g., snapshots of virtual secondary disks that correspond to the selected snapshot). If additional snapshots exist, then control flows to block 1311. If there are no other corresponding snapshots for the virtual machine, then control flows to block 1313. At block 1311, the corresponding snapshots in the other network storage units are bound to the selected snapshot. At block 1313, the virtual machine is resumed with the selected snapshot.
Maintaining snapshots of virtual machines allows activation of a virtual machine(s) at different points in time, which can be used for a myriad of purposes such as testing, agent-less back-up, offline virus scanning and other integrity checks, etc. When reverting to a previous snapshot, it may be desirable to discard an intermediate snapshot and maintain a snapshot that reflects a most recent state. For example, a snapshot to be deleted may correspond to state results from a failed test or experiment. This can also be used for applications that reset and erase any changes made by users upon every reboot, such as a kiosk application. Discarding or deleting the intermediate snapshot may reduce complexity and possibly improve performance. In a snapshot hierarchy that maintains children snapshots with deltas, a to-be-deleted snapshot may be rolled back into the predecessor intermediate snapshot, thus effectively deleting the intermediate snapshot. Of course, the snapshot to be discarded is not necessarily an intermediate snapshot, and may be the most recent snapshot. Those of ordinary skill in the art should appreciate the availability of methods for deleting a child snapshot.
Virtual Machine Management with Network Storage Facilities
In addition to the ability to snapshot, virtual machine management may include functionality to clone and/or migrate virtual machines (running or offline). These functions are typically implemented using redo-logs with copy-on-write semantics. With a network storage unit devoted to a single virtual machine, redo logs have been discarded along with the independent intermediary file system. With access to built-in facilities of a network storage system, virtual machine management can leverage the network storage management operations to efficiently manage virtual machines (e.g., clone, migrate, discovery, replicate for disaster recovery, etc.). The facilities are exposed to a virtualization system with an application programming interface for network storage management (“storage API”) and exposure of network storage units. The functionality that makes calls to the storage API may be implemented as part of a virtual machine manager that resides on a server communicatively coupled to virtualization system, part of a virtual machine manager that resides on a virtualization system, etc.
Discovery
Instead of initiating a communication session with every virtualization system, or accessing a centralized database of instantiated virtual machines, a virtual machine management system can efficiently and automatically discover virtual machines by examining visible network storage units. If a virtual machine management system does not have visibility into storage units of a network storage system or has limited visibility and access, then network storage units may be made visible by modifying the network storage mapping and masking configuration. For example, a partition table on all visible network storage units may be examined. A unique partition type, or other marker as discussed previously, may be used to facilitate automatic detection of these network storage units allocated for virtual machines. Within the context of arrays, existing array-based masking and mapping primitives can be used to control access to virtual machines. Additionally, locking capability may be included within a portion (e.g., partition) of a network storage unit to allow multiple virtualization systems to cooperate when accessing a single VM.
At block 1407, one of the virtual machine marked units is selected from the list. At block 1409, virtual machine data is loaded from the selected network storage unit. At block 1411, the loaded data is examined to identify any virtual secondary disk(s) and their corresponding network storage unit(s). Any identified network storage units are associated with the selected unit, and removed from the list. Of course, embodiments may mark network storage units that encode virtual secondary disks or other non-state data differently. Hence, detect and scan operations, such as those depicted in blocks 1401 and 1403, will yield a list of base network storage units. At block 1413, snapshot data and corresponding network storage units are identified. Any identified corresponding network storage units are associated with the selected unit and removed from the list. At block 1414, an indication of the selected unit and associated units is encoded. Control flows from block 1414 back to block 1405. If there are no remaining unselected network storage units in the list, then control flows to block 1415. At block 1415, the encoding of discovered virtual machines is indicated (e.g., presented via a user interface, stored in a file, etc.).
Moving a Virtual Machine
Instead of streaming pages of memory across an IP network, network storage system facilities can be leveraged to implement local move and remote move of a virtual machine. Moving a virtual machine locally transitions virtual machine computations to a different virtualization system, while maintaining data for the virtual machine in the same network storage unit. Moving a virtual machine remotely transitions virtual machine computations to a different virtualization system and relocates virtual machine data to a different network storage unit.
As previously stated, the movement of a virtual machine may be across a local-area network or a wide-area network. For network storage systems that support wide area replication (e.g., a SAN that supports wide-array replication), the supported facility can be employed to implement movement of a virtual machine across a wide-area network.
Continuous data protection support may also be achieved by running the move command operations 24×7 without the final suspend/resume. Periodic check points may be implemented to ensure consistency of the memory image. The periodic check points may occur coincident with the replica updates. At any point in time, the virtual machine may be resumed at the destination virtualization system based on the last coherent replica at the destination network storage unit. A customer may be presented with tunable parameters to control how frequently these check-points are performed, where the maximum setting requires all pages of memory to be written out constantly. The customer is then presented with a tradeoff of performance versus the time delay and potential for data loss as a result.
Cloning
When a user wishes to create a clone of a virtual machine, a virtual machine manager will trigger a snapshot as described above, and then request the snapshots to be unlinked. With the capabilities of a network storage system, a virtual machine can be cloned expeditiously.
Although the embodiment depicted in
Migrate
In addition to providing facilities for implementing current virtual machine management commands, implementation of virtual machines with network storage introduces additional management commands, such as migrate. In general, it may be necessary, or desirable, to move data (e.g., a virtual disk) of a virtual machine to a different storage pool using a migration facility.
The described embodiments may be provided as a computer program product, or software, that may include a machine-readable medium having stored thereon instructions, which may be used to program a computer system (or other electronic devices) to perform a process according to embodiments of the invention, whether presently described or not, since every conceivable variation is not enumerated herein. A machine readable medium includes any mechanism for storing or transmitting information in a form (e.g., software, processing application) readable by a machine (e.g., a computer). The machine-readable medium may include, but is not limited to, magnetic storage medium (e.g., floppy diskette); optical storage medium (e.g., CD-ROM); magneto-optical storage medium; read only memory (ROM); random access memory (RAM); erasable programmable memory (e.g., EPROM and EEPROM); flash memory; or other types of medium suitable for storing electronic instructions. In addition, embodiments may be embodied in an electrical, optical, acoustical or other form of propagated signal (e.g., carrier waves, infrared signals, digital signals, etc.), or wireline, wireless, or other communications medium.
While the invention(s) is (are) described with reference to various implementations and exploitations, it will be understood that these embodiments are illustrative and that the scope of the invention(s) is not limited to them. In addition, while our description of virtualization techniques has generally assumed that the virtual machines present interfaces consistent with a hardware system, persons of ordinary skill in the art will recognize that the techniques described may be used in conjunction with virtualizations that do not correspond directly to any particular hardware system. Virtualization systems in accordance with the present invention, implemented as hosted embodiments, non-hosted embodiments or as embodiments that tend to blur distinctions between the two, are all envisioned. Furthermore, various operations may be wholly or partially implemented in hardware. For example, a hardware implementation may employ a look-up table for modification of storage access requests to secure non-disk data.
Many variations, modifications, additions, and improvements are possible, regardless the degree of virtualization. For example, while particular example operations were depicted for illustrative purposes, various techniques for exposing a network storage system and various operations for performing moves, migrations, clones, etc. of virtual machines will also be appreciated by persons of ordinary skill in the art. Furthermore, while techniques and mechanisms have been described using particular network configurations, hardware architectures, memory organizations and particular operating system constructs (typically IA-32 based architectures/systems and Windows operations systems) as a descriptive framework, persons of ordinary skill in the art will recognize other implementations can be envisioned for use in systems that support other processor instruction set architectures, other network or memory configurations and/or other operating system constructs.
Plural instances may be provided for components, operations or structures described herein as a single instance. Finally, boundaries between various components, operations and data stores are somewhat arbitrary, and particular operations are illustrated in the context of specific illustrative configurations. Other allocations of functionality are envisioned and may fall within the scope of the invention(s). In general, structures and functionality presented as separate components in the exemplary configurations may be implemented as a combined structure or component. Similarly, structures and functionality presented as a single component may be implemented as separate components. These and other variations, modifications, additions, and improvements may fall within the scope of the invention(s).
This application is a continuation of prior U.S. application Ser. No. 16/189,511, filed Nov. 13, 2018, which is a continuation of U.S. application Ser. No. 15/701,390, filed Sep. 11, 2017, which is a continuation of prior U.S. application Ser. No. 14/937,453, filed Nov. 10, 2015, which is a continuation of prior U.S. application Ser. No. 11/960,460, filed Dec. 19, 2007, which claims the benefit of U.S. Provisional Application No. 60/871,234, filed Dec. 21, 2006, U.S. Provisional Application No. 60/884,568, filed Jan. 11, 2007, and U.S. Provisional Application No. 60/886,072, filed Jan. 22, 2007. This application is related to U.S. patent application Ser. No. 11/960,491, filed Dec. 19, 2007 and issued as U.S. Pat. No. 9,098,347, and to U.S. patent application Ser. No. 11/960,524, filed Dec. 19, 2007 and issued as U.S. Pat. No. 9,354,927 and entitled “SECURING VIRTUAL MACHINE DATA”. The content of each of the above applications is hereby incorporated by reference in its entirety.
Number | Name | Date | Kind |
---|---|---|---|
6321314 | Van Dyke | Nov 2001 | B1 |
6973447 | Aguilar et al. | Dec 2005 | B1 |
7089377 | Chen | Aug 2006 | B1 |
7424601 | Xu | Sep 2008 | B2 |
7865893 | Omelyanchuk et al. | Jan 2011 | B1 |
8234640 | Fitzgerald et al. | Jul 2012 | B1 |
8307359 | Brown et al. | Nov 2012 | B1 |
9098347 | Hiltgen et al. | Aug 2015 | B2 |
9189265 | Hiltgen et al. | Nov 2015 | B2 |
9354927 | Hiltgen et al. | May 2016 | B2 |
9760393 | Hiltgen et al. | Sep 2017 | B2 |
10162668 | Hiltgen et al. | Dec 2018 | B2 |
20020023225 | Lomnes | Feb 2002 | A1 |
20030120856 | Neiger et al. | Jun 2003 | A1 |
20040030822 | Rajan et al. | Feb 2004 | A1 |
20050193195 | Wu et al. | Sep 2005 | A1 |
20050223220 | Campbell et al. | Oct 2005 | A1 |
20060015869 | Neiger et al. | Jan 2006 | A1 |
20060075146 | Schoinas et al. | Apr 2006 | A1 |
20060112212 | Hildner | May 2006 | A1 |
20060174053 | Anderson et al. | Aug 2006 | A1 |
20060195715 | Herington | Aug 2006 | A1 |
20060224816 | Yamada et al. | Oct 2006 | A1 |
20060259734 | Sheu et al. | Nov 2006 | A1 |
20060282618 | Thompson et al. | Dec 2006 | A1 |
20060294519 | Hattori et al. | Dec 2006 | A1 |
20070005919 | van Riel | Jan 2007 | A1 |
20070028244 | Landis et al. | Feb 2007 | A1 |
20070073916 | Rothman et al. | Mar 2007 | A1 |
20070074208 | Ling et al. | Mar 2007 | A1 |
20070079307 | Dhawan et al. | Apr 2007 | A1 |
20070079308 | Chiaramonte et al. | Apr 2007 | A1 |
20070106986 | Worley | May 2007 | A1 |
20070106992 | Kitamura | May 2007 | A1 |
20070156986 | Neiger et al. | Jul 2007 | A1 |
20070174897 | Rothman et al. | Jul 2007 | A1 |
20070220121 | Suwarna | Sep 2007 | A1 |
20070283015 | Jackson et al. | Dec 2007 | A1 |
20080005447 | Schoenberg et al. | Jan 2008 | A1 |
20080059732 | Okada et al. | Mar 2008 | A1 |
20080072223 | Cowperthwaite et al. | Mar 2008 | A1 |
20080104591 | McCrory et al. | May 2008 | A1 |
20080114916 | Hummel et al. | May 2008 | A1 |
20080127182 | Newport et al. | May 2008 | A1 |
20080155169 | Hiltgen et al. | Jun 2008 | A1 |
20080155208 | Hiltgen et al. | Jun 2008 | A1 |
20080155223 | Hiltgen et al. | Jun 2008 | A1 |
20080162849 | Savagaonkar et al. | Jul 2008 | A1 |
20090300605 | Edwards et al. | Dec 2009 | A1 |
20110119748 | Edwards et al. | May 2011 | A1 |
20160062789 | Hiltgen et al. | Mar 2016 | A1 |
20180011731 | Hiltgen et al. | Jan 2018 | A1 |
20190079792 | Hiltgen et al. | Mar 2019 | A1 |
20190079793 | Hiltgen et al. | Mar 2019 | A1 |
Entry |
---|
Khalil, M. & Slovick, B., “ESX Server Storage II Tips and Tricks Raw Disk Mapping,” Presentation, VMworld2005 Virtualize Now, Las Vegas, NV, Oct. 18-20, 2005, © 2005 VMware, Inc., Palo Alto, CA, 49 pages. |
Stockman, M. et al., “Centrally-Stored and Delivered Virtual Machines in the Networking/System Administration Lab”, SIGITE Newsletter vol. 2:2, Jun. 2005, ACM Press, New York, NY, pp. 4-6. |
“Using Raw Device Mapping,” White Paper, © 1998-2005 VMware, Inc., Palo Alto, CA, 16 pages. |
“Using VMware® ESX Server System and VMware Virtual Infrastructure for Backup, Restoration, and Disaster Recovery,” White Paper, © 2005 VMware, Inc., Palo Alto, CA, 18 pages. |
Vaghini, Satyam “PAC267-C ESX Server Storage Internals”, Presentation, VMworld2005 Virtualize Now, Las Vegas, NV, Oct. 19, 2005, © 2005 VMware, Inc., Palo Alto, CA, 28 pages. |
Number | Date | Country | |
---|---|---|---|
20200272502 A1 | Aug 2020 | US |
Number | Date | Country | |
---|---|---|---|
60886072 | Jan 2007 | US | |
60884568 | Jan 2007 | US | |
60871234 | Dec 2006 | US |
Number | Date | Country | |
---|---|---|---|
Parent | 16189511 | Nov 2018 | US |
Child | 16816977 | US | |
Parent | 15701390 | Sep 2017 | US |
Child | 16189511 | US | |
Parent | 14937453 | Nov 2015 | US |
Child | 15701390 | US | |
Parent | 11960460 | Dec 2007 | US |
Child | 14937453 | US |