Abstract: In one embodiment, a computing device accesses an entity database storing information about entities in a virtualization environment. Each of the entities is associated with a single entity type and a shard. The shard further corresponds to the single entity type. The computing device determines an entity trail for the single shard, based on a modification history for one or more of the entities associated with the shard. The entity trail is divided into a plurality of time segments. The computing device receives an update from a client device indicating an update to an entity that is associated with the shard, and a client timestamp. The computing device selects a time segment corresponding to the client timestamp, and updates that portion of the entity trail with the client update.

Abstract: Techniques for rebalancing computing workloads between a set of on-premises resources of an on-premises system and a set of cloud computing resources of a cloud computing system. A method embodiment commences upon detecting a rebalancing trigger event. Responsive to the trigger event, a set of resource data corresponding to observations pertaining to the on-premises resources and resource data corresponding to observations pertaining the cloud resources is accessed. Based at least in part on the resource data, a plurality of candidate workload rebalancing plans are generated. The candidate workload rebalancing plans are evaluated against the likelihood of achieving quantitative objectives resulting from the rebalancing. A portion of the scheduling commands to carry out the workload rebalancing plans are sent to the cloud computing system, and another portion of the scheduling commands are sent to the on-premises computing system. The receiving computing systems carry out the scheduling commands.

Abstract: Techniques for rebalancing computing workloads between a set of on-premises resources of an on-premises system and a set of cloud computing resources of a cloud computing system. A method embodiment commences upon detecting a rebalancing trigger event. Responsive to the trigger event, a set of resource data corresponding to observations pertaining to the on-premises resources and resource data corresponding to observations pertaining the cloud resources is accessed. Based at least in part on the resource data, a plurality of candidate workload rebalancing plans are generated. The candidate workload rebalancing plans are evaluated against the likelihood of achieving quantitative objectives resulting from the rebalancing. A portion of the scheduling commands to carry out the workload rebalancing plans are sent to the cloud computing system, and another portion of the scheduling commands are sent to the on-premises computing system. The receiving computing systems carry out the scheduling commands.

Abstract: Systems for distributed resource management. A method for deploying two or more computing workloads across heterogeneous computing systems commences upon establishing network communications between an on-premises computing system and a cloud computing system. Resource data corresponding to the on-premises resources and the cloud resources are collected continuously and saved as a time series history of observations. Upon a user request or other event, workload placement planning commences. A set of workload placement plans are generated, which workload placement plans are then evaluated in accordance with one or more quantitative objectives. Scheduling commands to carry out the workload placement plans are generated. A first portion of the scheduling commands is executed at the cloud computing system, and another portion of the scheduling commands is executed at the on-premises computing system.

Abstract: In particular embodiments, a system for managing a virtualization environment includes host machines, each of the host machines including a hypervisor, user virtual machines (UVMs) and a virtual machine controller. The virtualization environment also includes virtual disks comprising a plurality of storage devices, and being accessible by the virtual machine controllers. The virtual machine controllers conduct I/O transactions with the virtual disks. The system stores an entity-relationship graph representing elements in the virtualization environment. Each of the elements is represented by an entity-type node in the entity-relationship graph, and relationships between the elements are represented by edges between the nodes.

Abstract: In particular embodiments, a system for managing a virtualization environment includes host machines, each of the host machines including a hypervisor, user virtual machines (UVMs) and a virtual machine controller. The virtualization environment also includes virtual disks comprising a plurality of storage devices, and being accessible by the virtual machine controllers. The virtual machine controllers conduct I/O transactions with the virtual disks. The system stores an entity-relationship graph representing elements in the virtualization environment. Each of the elements is represented by an entity-type node in the entity-relationship graph, and relationships between the elements are represented by edges between the nodes.

Abstract: In particular embodiments, a system for managing a virtualization environment includes host machines, each of the host machines including a hypervisor, user virtual machines (UVMs) and a virtual machine controller. The virtualization environment also includes virtual disks comprising a plurality of storage devices, and being accessible by the virtual machine controllers. The virtual machine controllers conduct I/O transactions with the virtual disks. The system stores an entity-relationship graph representing elements in the virtualization environment. Each of the elements is represented by an entity-type node in the entity-relationship graph, and relationships between the elements are represented by edges between the nodes.

Abstract: In one embodiment, a computing device accesses an entity database storing information about entities in a virtualization environment. Each of the entities is associated with a single entity type and a shard. The shard further corresponds to the single entity type. The computing device determines an entity trail for the single shard, based on a modification history for one or more of the entities associated with the shard. The entity trail is divided into a plurality of time segments. The computing device receives an update from a client device indicating an update to an entity that is associated with the shard, and a client timestamp. The computing device selects a time segment corresponding to the client timestamp, and updates that portion of the entity trail with the client update.

Abstract: In particular embodiments, a system for managing a virtualization environment includes host machines, each of the host machines including a hypervisor, user virtual machines (UVMs) and a virtual machine controller. The virtualization environment also includes virtual disks comprising a plurality of storage devices, and being accessible by the virtual machine controllers. The virtual machine controllers conduct I/O transactions with the virtual disks. The system stores an entity-relationship graph representing elements in the virtualization environment. Each of the elements is represented by an entity-type node in the entity-relationship graph, and relationships between the elements are represented by edges between the nodes.