Abstract: An example system may provide for plurality of object block stores and a store management system. The store management system is configured to access a first bucket including a first storage class bucket tag identified by a first key-value pair corresponding to a first one of the plurality of object block stores with a first storage class. The store management system is further configured to write a first data object to the first one of the plurality of object block stores with a first storage class. The first one of the plurality of object block stores is represented by the first bucket. The first one of the plurality of object block stores is configured to store the first data object according to the first storage class specified by the first storage class bucket tag.

Abstract: An example system may provide for plurality of object block stores and a store management system. The store management system is configured to access a first bucket including a first storage class bucket tag identified by a first key-value pair corresponding to a first one of the plurality of object block stores with a first storage class. The store management system is further configured to write a first data object to the first one of the plurality of object block stores with a first storage class. The first one of the plurality of object block stores is represented by the first bucket. The first one of the plurality of object block stores is configured to store the first data object according to the first storage class specified by the first storage class bucket tag.

Abstract: Techniques for resource allocation are described. Some embodiments provide a computing system and method for resource allocation in a virtualized computing environment comprising at least one physical computing system hosting multiple virtual machines, that performs at least some of the described techniques. In one embodiment, a user connection server is configured to receive a request for allocation of a virtual machine, for a user. The user connection server determines an attribute value of the user. Based on the attribute value of the user, allocation of physical computing resources for the virtual machine is determined. A management server is configured to boot the virtual machine for access by the user, the virtual machine booted with the determined allocation of physical computing resources for the virtual machine.

Abstract: Aspects of peripheral device sharing for virtual machines are described. In some aspects, a virtual machine requires access to a peripheral device. The virtual machine is executed in a first host computing system. A table is accessed, and includes a list of peripheral devices, an indication that the peripheral device is connected to a second host computing system, a bandwidth requirement of the peripheral device, and a network bandwidth allocated to the virtual machine. The virtual machine is provided with access to the peripheral device based on a comparison between the bandwidth requirement of the peripheral device, and a threshold percentage of the network bandwidth allocated to the virtual machine.

Abstract: Aspects of peripheral device sharing for virtual machines are described. In some aspects, a virtual machine requires access to a peripheral device. The virtual machine is executed in a first host computing system. A table is accessed, and includes a list of peripheral devices, an indication that the peripheral device is connected to a second host computing system, a bandwidth requirement of the peripheral device, and a network bandwidth allocated to the virtual machine. The virtual machine is provided with access to the peripheral device based on a comparison between the bandwidth requirement of the peripheral device, and a threshold percentage of the network bandwidth allocated to the virtual machine.

Abstract: Techniques for sharing a peripheral device connected to a first host computing system in a cluster are disclosed. In one embodiment, a request to access the peripheral device connected to the first host computing system is received from a virtual machine running on a second host computing system. Further, a bandwidth requirement associated with the peripheral device is determined. Furthermore, one of enabling the virtual machine to remotely access the peripheral device over a network and recommending migration of the virtual machine to the first host computing system to locally access the peripheral device is performed based on the bandwidth requirement of the peripheral device.

Abstract: Particular embodiments change a current storage I/O path used by a host computer to access networked storage to an alternative storage I/O path by considering traffic load at a networked switch in the current storage I/O path. The host computer transmits a request to the networked switch in the current storage I/O path to provide network load information currently experiences by the networked switch. After receiving network load information from the networked switch, the host computer then evaluates whether the networked switch is overloaded based on the received network load information. Based on the evaluation, the host computer selects a new alternative storage I/O path to the networked storage that does not include the networked switch, and then forwards future storage I/O communications to the networked storage using the new alternative storage I/O path.

Abstract: A virtual machine (VM) is designated as a guarded VM so that restricted operations may not be performed on the VM without permission from the VM's owner. A request to perform at least one of the restricted operations on the VM is received. When the VM is a guarded VM, the request to perform at least one of the restricted operations on the VM is sent to the VM's owner. When the VM's owner at least partially approves the request, at least some of the restricted operations on the VM are enabled.

Abstract: Techniques for placing a first fiber channel (FC) switch into maintenance mode in a virtualized computing environment in which each data store is connected to at least one host computing system via at least two FC switches are described. In one embodiment, a first active input/output (I/O) path from a data store to a host computing system via the first FC switch is identified. Further, a path change is initiated from the data store to the host computing system via any other FC switch coupled to the data store and the host computing system.

Abstract: Techniques for sharing a peripheral device connected to a first host computing system in a cluster are disclosed. In one embodiment, a request to access the peripheral device connected to the first host computing system is received from a virtual machine running on a second host computing system. Further, a bandwidth requirement associated with the peripheral device is determined. Furthermore, one of enabling the virtual machine to remotely access the peripheral device over a network and recommending migration of the virtual machine to the first host computing system to locally access the peripheral device is performed based on the bandwidth requirement of the peripheral device.

Abstract: High availability of a virtual machine is ensured even when all of the virtual machine's IO paths fail. In such a case, the virtual machine is migrated to a host that is sharing the same storage system as the current host in which the virtual machine is being executed and has at least one functioning IO path to the shared storage system. After execution control of the virtual machine is transferred to the new host, IO operations from the virtual machine are issued over the new IO path.

Abstract: Techniques for placing a first storage network device into maintenance mode in a virtualized computing environment in which each data store is connected to at least one host computing system via a storage network device are described. In one embodiment, a first data store having an active input/output (I/O) path to a first host computing system via the first storage network device is identified. Further, migration of virtual machine disks (VMDKs) on the first data store and/or migration of workloads running on the first host computing system are recommended such that the VMDKs can be accessed by the workloads via any other storage network device.

Abstract: Techniques for placing a first storage network device into maintenance mode in a virtualized computing environment in which each data store is connected to at least one host computing system via a storage network device are described. In one embodiment, a first data store having an active input/output (I/O) path to a first host computing system via the first storage network device is identified. Further, migration of virtual machine disks (VMDKs) on the first data store and/or migration of workloads running on the first host computing system are recommended such that the VMDKs can be accessed by the workloads via any other storage network device.

Abstract: Techniques for placing a first fibre channel (FC) switch into maintenance mode in a virtualized computing environment in which each data store is connected to at least one host computing system via at least two FC switches are described. In one embodiment, a first active input/output (I/O) path from a data store to a host computing system via the first FC switch is identified. Further, a path change is initiated from the data store to the host computing system via any other FC switch coupled to the data store and the host computing system.

Abstract: A virtual machine (VM) is designated as a guarded VM so that restricted operations may not be performed on the VM without permission from the VM's owner. A request to perform at least one of the restricted operations on the VM is received. When the VM is a guarded VM, the request to perform at least one of the restricted operations on the VM is sent to the VM's owner. When the VM's owner at least partially approves the request, at least some of the restricted operations on the VM are enabled.

Abstract: Particular embodiments change a current storage I/O path used by a host computer to access networked storage to an alternative storage I/O path by considering traffic load at a networked switch in the current storage I/O path. The host computer transmits a request to the networked switch in the current storage I/O path to provide network load information currently experiences by the networked switch. After receiving network load information from the networked switch, the host computer then evaluates whether the networked switch is overloaded based on the received network load information. Based on the evaluation, the host computer selects a new alternative storage I/O path to the networked storage that does not include the networked switch, and then forwards future storage I/O communications to the networked storage using the new alternative storage I/O path.

Abstract: Techniques for resource allocation are described. Some embodiments provide a computing system and method for resource allocation in a virtualized computing environment comprising at least one physical computing system hosting multiple virtual machines, that performs at least some of the described techniques. In one embodiment, a user connection server is configured to receive a request for allocation of a virtual machine, for a user. The user connection server determines an attribute value of the user. Based on the attribute value of the user, allocation of physical computing resources for the virtual machine is determined. A management server is configured to boot the virtual machine for access by the user, the virtual machine booted with the determined allocation of physical computing resources for the virtual machine.

Abstract: High availability of a virtual machine is ensured even when all of the virtual machine's IO paths fail. In such a case, the virtual machine is migrated to a host that is sharing the same storage system as the current host in which the virtual machine is being executed and has at least one functioning IO path to the shared storage system. After execution control of the virtual machine is transferred to the new host, IO operations from the virtual machine are issued over the new IO path.