The State University of New York

The Research Foundation for the State University

Kartik Gopalan

Jian Wang

Michael Hines

Ping Yang

Yaohui Hu

Jonathan Terner

Dinuni K Fernando

Umesh Deshpande

Hui Lu

A multi-hypervisor system, comprising: a plurality of hypervisors comprising a first hypervisor and a second hypervisor, at least one of the plurality of hypervisors being a transient hypervisor; and at least one Span VM, concurrently executing on each of the plurality of hypervisors, the at least one transient hypervisor being adapted to be dynamically at least one of injected and removed under the at least one Span VM concurrently with execution of the at least one Span VM on another hypervisor, wherein the at least one Span VM has a single and consistent at least one of memory space, virtual CPU state, and set of input/output resources, shared by the plurality of hypervisors.

Multi-hypervisor virtual machines that run on multiple co-located hypervisors

Systems and methods for live updating virtualization systems are disclosed. One method may include identifying a container on a first virtual machine of a virtualization system. The container may include at least one process associated with performing a task on a first virtual machine. The method may also include determining a first communication path between the process included in the container on the first virtual machine and a hypervisor of the virtualization system, and migrating the container and the process included in the container to a second virtual machine. The second virtual machine may be distinct from the first virtual machine. Additionally, the method may include determining a second communication path between the process included in the migrated container on the second virtual machine and the hypervisor of the virtualization system.

Systems and methods for live update of operating systems and hypervisors within virtualization systems

Post-copy is one of the two key techniques (besides pre-copy) for live migration of virtual machines in data centers. Post-copy provides deterministic total migration time and low downtime for write-intensive VMs. However, if post-copy migration fails for any reason, the migrating VM is lost because the VM's latest consistent state is split between the source and destination nodes during migration. PostCopyFT provides a new approach to recover a VM after a destination or network failure during post-copy live migration using an efficient reverse incremental checkpointing mechanism. PostCopyFT was implemented and evaluated in the KVM/QEMU platform. Experimental results show that the total migration time of post-copy remains unchanged while maintaining low failover time, downtime, and application performance overhead.

Recovering a virtual machine after failure of post-copy live migration

Standard nested virtualization allows a hypervisor to run other hypervisors as guests, i.e. a level-0 (L0) hypervisor can run multiple level-1 (L1) hypervisors, each of which can run multiple level-2 (L2) virtual machines (VMs), with each L2 VM is restricted to run on only one L1 hypervisor. Span provides a Multi-hypervisor VM in which a single VM can simultaneously run on multiple hypervisors, which permits a VM to benefit from different services provided by multiple hypervisors that co-exist on a single physical machine. Span allows (a) the memory footprint of the VM to be shared across two hypervisors, and (b) the responsibility for CPU and I/O scheduling to be distributed among the two hypervisors. Span VMs can achieve performance comparable to traditional (single-hypervisor) nested VMs for common benchmarks.

Multi-hypervisor virtual machines

Standard nested virtualization allows a hypervisor to run other hypervisors as guests, i.e. a level-(L) hypervisor can run multiple level-(L) hypervisors, each of which can run multiple level-(L) virtual machines (VMs), with each LVM is restricted to run on only one Lhypervisor. Span provides a Multi-hypervisor VM in which a single VM can simultaneously run on multiple hypervisors, which permits a VM to benefit from different services provided by multiple hypervisors that co-exist on a single physical machine. Span allows (a) the memory footprint of the VM to be shared across two hypervisors, and (b) the responsibility for CPU and I/O scheduling to be distributed among the two hypervisors. Span VMs can achieve performance comparable to traditional (single-hypervisor) nested VMs for common benchmarks.

A checkpointing method for creating a file representing a restorable state of a virtual machine in a computing system, comprising identifying processes executing within the virtual machine that may store confidential data, and marking memory pages and files that potentially contain data stored by the identified processes; or providing an application programming interface for marking memory regions and files within the virtual machine that contain confidential data stored by processes; and creating a checkpoint file, by capturing memory pages and files representing a current state of the computing system, which excludes information from all of the marked memory pages and files.

System and method for security and privacy aware virtual machine checkpointing

Gang migration refers to the simultaneous live migration of multiple Virtual Machines (VMs) from one set of physical machines to another in response to events such as load spikes and imminent failures. Gang migration generates a large volume of network traffic and can overload the core network links and switches in a datacenter. In this paper, we present an approach to reduce the network overhead of gang migration using global deduplication (GMGD). GMGD identifies and eliminates the retransmission of duplicate memory pages among VMs running on multiple physical machines in the cluster. The design, implementation and evaluation of a GMGD prototype is described using QEMU/KVM VMs. Evaluations on a 30-node Gigabit Ethernet cluster having 10 GigE core links shows that GMGD can reduce the network traffic on core links by up to 65% and the total migration time of VMs by up to 42% when compared to the default migration technique in QEMU/KVM. Furthermore, GMGD has a smaller adverse performance impact on network-bound applications.

Gang migration of virtual machines using cluster-wide deduplication

Datacenter clusters often employ live virtual machine (VM) migration to efficiently utilize cluster-wide resources. Gang migration refers to the simultaneous live migration of multiple VMs from one set of physical machines to another in response to events such as load spikes and imminent failures. Gang migration generates a large volume of network traffic and can overload the core network links and switches in a data center. The present technology reduces the network overhead of gang migration using global deduplication (GMGD). GMGD identifies and eliminates the retransmission of duplicate memory pages among VMs running on multiple physical machines in the cluster. A prototype GMGD reduces the network traffic on core links by up to 51% and the total migration time of VMs by up to 39% when compared to the default migration technique in QEMU/KVM, with reduced adverse performance impact on network-bound applications.

Growing community of inventors

Vestal, NY, United States of America

Kartik Gopalan