Vmware, Inc.

Google Inc.

Cisco Technology, Inc.

Velostrata Ltd.

Leonid Vasetsky

Israel Zvi Ben-Shaul

Tal Zamir

Guy Yogev

Shahar Glixman

Kfir Lev-Ari

Ady Degany

Yaniv Ben-Ari

Ming Jin Chen

Sheng Lu

Zhikai Chen

Kan Liu

Etai Lev-Ran

Daniel Kaminsky

Dan Ilan

Eytan Heidingsfeld

Yehuda Itzhakov

Ophir Romano

Barak Yagour

Or Igelka

Disclosure is made of methods, apparatus and system for migrating virtual machines (VMs) between source and destination in a computing environment and, more specifically, to replication based migration. VMs migration is controlled so as to manage transferral of data associated with one or more VMs from a source location to a destination location to meet certain user definable or system constraints. Dynamic control and adjustment of system parameters associated with the migration is also disclosed.

Optimizing replication-based migration of virtual machines

Techniques are disclosed herein for managing network bandwidth in a distributed system having multiple endpoints that upload data to a central server or server cluster. One embodiment provides a computer-implemented method that includes receiving a logical topology of the network that is determined based on network measurements. The method further includes generating a tree having links of the logical network topology as nodes, and allocating bandwidth during a traversal of the tree to obtain a bandwidth allocation scheme for the endpoints uploading data. In addition, the method includes controlling bandwidth consumption by the endpoints based on the determined bandwidth allocation scheme.

Reverse breadth-first search method for optimizing network traffic in a distributed system with a point of convergence

Methods and systems for n-way cloning and synchronization of a user desktop image are provided. Example embodiments provide a Cloning and Synchronization System ('CSS') which binds a server stored CVD object representing the user's desktop image to one or more endpoint devices. Each endpoint device receives a clone of the CVD object that comprises one or more layers of the server CVD depending upon the suitability of the endpoint device hardware and operating system to the server stored desktop. The cloned CVDs in the endpoint devices are then kept synchronized by synchronization operations. In one embodiment, the CSS allows only one endpoint device to act as a master device and push up changes to the server CVD. These changes are then pushed down to the other devices using different synchronization methods dependent upon the layer.

N-way synchronization of desktop images

Techniques are disclosed herein for managing network bandwidth in a distributed system having multiple endpoints that upload data to a central server or server cluster. One embodiment provides a computer-implemented method that includes collecting network measurements from traffic flows in a network over which the endpoints upload the data to the central server or server cluster. The method further includes determining a logical topology of the network based on the network measurements, and solving an optimization problem to determine a bandwidth allocation scheme for the endpoints uploading data, wherein capacity of links in the logical topology are constraints in the optimization problem. In addition, the method includes controlling bandwidth consumption by the endpoints based on the determined bandwidth allocation scheme.

Method and system for optimizing network traffic in a distributed system with a point of convergence

A method and system for real-time cloud bursting is provided. The method and system are directed to extending a data center with cloud computing resources by decoupling computing resources and storage devices in a virtualized data center, and booting the decoupled computing resources in a staged process while storage devices are divided and prioritized into components. Data and boot instructions are re-routed and cached as needed through a proxy system.

Real time cloud workload streaming

A method and system for workload migration across a hybrid network is provided. The method and system are directed to migrating a workload to a cloud by transferring the workload computing processes to the cloud, streaming workload data as necessary to execute the workload processes in the cloud, transferring the remaining workload data in a background process to cloud storage, and then completing migration by switching the primary data source of the workload computing processes to the cloud storage. In another aspect, a method and system are directed to migrating on-premises storage associated with a virtual machine to the cloud to be stored and executed in association with containerized applications. Additionally, techniques for full detach and partial detach of containerized applications are provided.

Workload migration across a hybrid network

A desktop image management system is described that can efficiently distribute updates to virtual desktops running on host servers in a data center. The system is comprised of a central server and multiple agents, each agent installed on a virtual machine. When a VM receives an update from the central server, the files are stored in a single instance store on the host server. The agent running on the VM then creates a corresponding stub file (empty of content) on the VM for each file in the single instance store. The agent further marks the stub file to indicate that the stub file is mapped to the single instance store. When the guest operating system requests to read the stub file, the virtual disk layer of the host server detects that the requested block has been marked and fetches the content of the file from the single instance store.

Desktop image management for virtual desktops

A method and system for workload migration across a hybrid network is provided. The method and system are directed to migrating a workload to a cloud by transferring the workload computing processes to the cloud, streaming workload data as necessary to execute the workload processes in the cloud, transferring the remaining workload data in a background process to cloud storage, and then completing migration by switching the primary data source of the workload computing processes to the cloud storage.

Real time cloud bursting

Techniques are disclosed herein for managing network bandwidth in a distributed system having multiple endpoints that upload data to a central server or server cluster. One embodiment provides a computer-implemented method that includes receiving a logical topology of the network that is determined based on network measurements. The method further includes resolving a convex optimization problem over the logical topology to determine a bandwidth allocation scheme for the endpoints uploading data. In addition, the method includes controlling bandwidth consumption by the endpoints based on the determined bandwidth allocation scheme.

Resolving a convex optimization problem to optimize network traffic in a distributed system

Methods and systems for n-way cloning and synchronization of a user desktop image are provided. Example embodiments provide a Cloning and Synchronization System ('CSS') which binds a server stored CVD object representing the user's desktop image to one or more endpoint devices. Each endpoint device receives a clone of the CVD object that comprises one or more layers of the server CVD depending upon the suitability of the endpoint device hardware and operating system to the server stored desktop. The cloned CVDs in the endpoint devices are then kept synchronized by synchronization operations. In one embodiment, the CSS allows only one endpoint device to push up changes to the server CVD. User files are synchronized with the all endpoints using a live sync process. System and managed applications may be intelligently merged into an inactive device rather than overwriting data modified on the inactive device.

Synchronization of desktop images with smart image merging

Techniques are provided for de-duplication of data. In one embodiment, a system comprises de-duplication logic that is coupled to a de-duplication repository. The de-duplication logic is operable to receive, from a client device over a network, a request to store a file in the de-duplicated repository using a single storage encoding. The request includes a file identifier and a set of signatures that identify a set of chunks from the file. The de-duplication logic determines whether any chunks in the set are missing from the de-duplicated repository and requests the missing chunks from the client device. Then, for each missing chunk, the de-duplication logic stores in the de-duplicated repository that chunk and a signature representing that chunk. The de-duplication logic also stores, in the de-duplicated repository, a file entry that represents the file and that associates the set of signatures with the file identifier.

Storage-network de-duplication

Independent synchronization of virtual desktop image layers

Fast provisioning of a centralized virtual desktop using linked clones with overlaid centralized virtual desktop layers

A desktop image management system is described that can efficiently distribute updates to virtual desktops running on host servers in a data center. The system is comprised of a central server and multiple agents, each agent installed on a virtual machine. When a VM receives an update from the central server, the files are stored in a single instance store on the host server. On a first request to open the directory containing the files on the VM, corresponding stub files are created on-demand for each file in the single instance store. The stub files are marked to indicate that they are mapped to the single instance store. Subsequently, when the guest operating system requests to read the stub file, the virtual disk layer of the host server detects that the requested block has been marked and fetches the content of the file from the single instance store.

Growing community of inventors

Zikhron Yaakov, Israel

Leonid Vasetsky