Red Hat Israel, Ltd.

Cisco Technology, Inc.

Cisco Systems O.i.a. (1988) Ltd.

Roni Luxenberg

Uri Lublin

Doron Oz

Michael Tsirkin

David Delano Ward

Shahar Frank

Nir Ben-Dvora

Danny Karpati

Assaf Ben-Amitai

A request may be received from a second client device to connect to a running virtual machine instance of a virtual machine while a first client device is connected to the running virtual machine instance. The second client device may be connected to the running virtual machine instance in response to receiving the request from the second client device. Furthermore, a connection between the first client device and the running virtual machine instance of the virtual machine may be converted from a primary mode to a secondary mode.

Client live migration for a virtual machine

A second client device may be connected to a running virtual machine instance of a virtual machine while a first client device is connected to the running virtual machine instance. A first input data may be received from the first client device and a second input data may be received from the second client device. A determination may be made as to whether the second input data includes a great number of commands than the first input data in the time period. An output data associated with the running virtual machine instance may be generated by using the second input data in response determining that the second input data includes a greater number of commands than the first input data in the time period.

A logical client includes a primary client device and one or more secondary client devices. Each of the secondary client devices may be coupled to one or more peripherals. The primary client in the logical client may use a virtual machine and/or an application that uses one or more peripheral devices. The primary client device may not be coupled to the one or more peripheral devices used by the application and/or the VM. The primary client device may access the peripheral devices coupled to secondary client devices in order to use the application and/or the VM.

Managing a logical client for an application

A computing device connects a first client device to a session of a virtual machine. The computing device subsequently connects a second client device to the session in response to a request from the second client device to access the virtual machine. The computing device may then disconnect the first device from the virtual machine session or disable one or more functions associated with the first client device.

A mechanism for managed network filter/forward programming in a virtualization system is disclosed. A method of the invention includes receiving a network programming request from a hypervisor managing a virtual machine (VM), wherein the hypervisor intercepted the network programming request from the VM, determining whether the network programming request satisfies an administrative network policy, and updating network configuration settings as requested by the network programming request if the network programming request satisfies the administrative network policy, wherein the updating occurs without restarting the VM.

Managed network filter/forward programming in a virtualization system by an agent more privileged than the hypervisor

A logical client includes a primary client device and one or more secondary client devices. Each of the secondary client devices may be coupled to one or more peripherals. The primary client in the logical client may use a virtual machine (VM) and/or an application that uses one or more peripheral devices. The primary client device may not be coupled to the one or more peripheral devices used by the application and/or the VM. The primary client device may access the peripheral devices coupled to secondary client devices in order to use the application and/or the VM.

Managing a logical client for a virtual machine

A computing device executes a virtual machine that provides a service to a client. The computing device monitors one or more conditions associated with at least one of the virtual machine or the client. The computing device receives a data packet from the client, wherein the data packet is addressed to the virtual machine. The computing device determines, based on the one or more conditions, whether to route the data packet from the client through a virtual appliance before sending the data packet to the virtual machine.

Dynamic routing through virtual appliances

A computing device executing a virtual machine and a hypervisor that manages the virtual machine receives a data packet that is addressed to the virtual machine. The computing device writes the data packet to a buffer in a shared memory. The computing device maps the buffer to a memory of a virtual appliance that is hosted by the hypervisor to enable the virtual appliance to operate on the data packet without making a copy of the data packet. The computing device provides the data packet to the virtual machine after the virtual appliance has operated on the data packet.

Routing data packets between virtual machines using shared memory without copying the data packet

A computing device determines a routing scheme that identifies a plurality of virtual appliances to route data packets through and an order in which to perform the routing. The computing device receives a data packet from a client. The computing device routes the data packet to the plurality of virtual appliances in accordance with the routing scheme, wherein each of the plurality of virtual appliances performs one or more operations on the data packet. The computing device sends the data packet to the virtual machine after the plurality of virtual appliances have completed the one or more operations on the data packet.

Managing a configurable routing scheme for virtual appliances

A processing device of host may receive desktop image data for a first client device that includes a graphical object. The processing device may determine an object identifier of the graphical object and a device identifier of a second client device associated with the graphical object. The processing device may provide the object identifier and the device identifier to the first client device.

Managing a distributed cache for virtual machines

A mechanism for partitioning of a transmission control protocol (TCP) window in a virtual system is disclosed. A method includes estimating effective bandwidth of a communication link between a group of client devices at a first location and at least one host machine at a second location, wherein each client device of the groups of client devices maintains a communication session over the communication link with one of the at least one host machine, partitioning a transport control protocol (TCP) window of the communication link into portions based on quality of service (QoS) policy data of the communication sessions, where each portion is associated with one of the communication sessions, and sending, to each client device in the group of client devices, a parameter identifying the portion of the TCP window partitioned for the client device, the parameter sent to communication endpoints of each communication session.

Partitioning of a transmission control protocol window in a virtual system

A mechanism for distributed multiple monitor display split using multiple client devices in a virtualization system is disclosed. A method of the invention includes determining, by a remote access protocol server executed by a computing device, that a monitor configuration of a primary client device is not compatible with a multiple monitor configuration of a virtual machine (VM) hosted by a hypervisor of the computing device, establishing a communication session with remote access protocol clients of multiple client devices of a logical client unit comprising the primary client device, wherein each of the multiple client devices comprises at least one monitor, and for each VM monitor of the multiple monitors configured for the VM, mapping, by the remote access protocol server, the VM monitor to a monitor of a client device in the logical client unit.

Distributed multiple monitor display split using multiple client devices in a virtualization system

A mechanism for a virtual machine (VM) monitor display split using multiple client devices in a virtualization system is disclosed. A method of the invention includes receiving, by a remote access protocol server, resolution data of a monitor of a virtual machine (VM), receiving resolution data for each monitor of multiple client devices forming a logical client unit, determining, based on the resolution data of the VM monitor and the resolution data of the monitors of the multiple client devices, client devices of the multiple client devices in the logical client unit to display the monitor of the VM, dividing, based on the resolutions of the determined client devices, the monitor of the VM into a number of portions that is equal to a number of the determined client devices, and mapping each of the portions of the monitor of the VM to a monitor of the determined client devices.

Virtual machine monitor display split using multiple client devices in a virtualization system

Clients may display desktop environments to provide users with access to virtual machines (VMs). Graphical objects that displayed in the desktop environments are stored in caches in multiple clients. A host that hosts a VM may track or manage the graphical objects that are in the caches of the multiple clients. The host may instruct a first client to obtain a graphical object from a second client that is near the first client, instead of providing the graphical object to the first client directly.

The load of a server and its client is balanced during connection setup time and dynamically during operation. In one embodiment, a server that hosts virtual machines for a plurality of clients balances its load with its clients. The server receives a request from a requesting client to establish a connection. The request includes capacity information about the requesting client. The server then negotiates with the requesting client about load allocation based on the capacity information. The negotiated load can be dynamically rebalanced after the connection is established.

Growing community of inventors

Raanana, Israel

Roni Luxenberg