摘要:
A system and related method of operation for migrating the memory of a virtual machine from one NUMA node to another. Once the VM is migrated to a new node, migration of memory pages is performed while giving priority to the most utilized pages, so that access to these pages becomes local as soon as possible. Various heuristics are described to enable different implementations for different situations or scenarios.
摘要:
A system and related method of operation for migrating the memory of a virtual machine from one NUMA node to another. Once the VM is migrated to a new node, migration of memory pages is performed while giving priority to the most utilized pages, so that access to these pages becomes local as soon as possible. Various heuristics are described to enable different implementations for different situations or scenarios.
摘要:
A system and related method of operation for migrating the memory of a virtual machine from one NUMA node to another. Once the VM is migrated to a new node, migration of memory pages is performed while giving priority to the most utilized pages, so that access to these pages becomes local as soon as possible. Various heuristics are described to enable different implementations for different situations or scenarios.
摘要:
A system for increasing the efficiency of migrating, at least in part, a virtual machine from a source host to a destination host is described wherein the content of one or more portions of the address space of the virtual machine are each uniquely associated at the source host with a signature that may collide, absent disambiguation, with different content at the destination host. Code in both the source and destination hosts disambiguates the signature(s) so that each disambiguated signature may be uniquely associated with content at the destination host, and so that collisions with different content are avoided at the destination host. Logic is configured to determine whether the content uniquely associated with a disambiguated signature at the destination host is already present in the destination host memory, and, if so, to back one or more portions of the address space of the virtual machine having this content with one or more portions of the destination host memory already holding this content.
摘要:
An “old” hypervisor is upgraded to or otherwise replaced by a “new” hypervisor without migrating virtual machines to a standby computer. The old hypervisor partitions the computer that it controls between a source partition and a target partition. The hypervisor and its virtual machines initially run on the source partition, while a new hypervisor is installed on the target partition. The virtual machines are migrated to the new hypervisor without physically moving the in-memory virtual-machine data. Instead, the old hypervisor sends memory pointers, and the new hypervisor claims the respective memory locations storing the virtual-machine data. After all virtual machines are migrated, the old hypervisor bequeaths the hypervisor memory and a last processor that the old hypervisor requires to run. The new hypervisor claims the bequeathed processor and hypervisor memory after the old hypervisor terminates to complete the upgrade/exchange.
摘要:
Miss rate curves are constructed in a resource-efficient manner so that they can be constructed and memory management decisions can be made while the workloads are running. The resource-efficient technique includes the steps of selecting a subset of memory pages for the workload, maintaining a least recently used (LRU) data structure for the selected memory pages, detecting accesses to the selected memory pages and updating the LRU data structure in response to the detected accesses, and generating data for constructing a miss-rate curve for the workload using the LRU data structure. After a memory page is accessed, the memory page may be left untraced for a period of time, after which the memory page is retraced.
摘要:
A hypervisor exchange, e.g., an upgrade, can include consolidating resident virtual machines into a single host virtual machine, exchanging an old hypervisor with a new (upgraded) hypervisor, and disassociating the virtual resident virtual machines by migrating them to the new hypervisor. The consolidating can involve migrating the resident virtual machines from the old hypervisor to a guest hypervisor on the host virtual machine. The exchange can involve: 1) suspending the host virtual machine before the exchange; and 2) resuming the host virtual machine after the exchange; or migrating the host virtual machine from a partition including the old hypervisor to a partition hosting the new hypervisor. Either way, an exchange (upgrade) is achieve without requiring a bandwidth consuming migration over a network to a standby machine.
摘要:
A hypervisor-exchange process includes: suspending, by an “old” hypervisor, resident virtual machines; exchanging the old hypervisor for a new hypervisor, and resuming, by the new hypervisor, the resident virtual machines. The suspending can include “in-memory” suspension of the virtual machines until the virtual machines are resumed by the new hypervisor. Thus, there is no need to load the virtual machines from storage prior to the resuming. As a result, any interruption of the virtual machines is minimized. In some embodiments, the resident virtual machines are migrated onto one or more host virtual machines to reduce the number of virtual machines being suspended.
摘要:
An “old” hypervisor is upgraded to or otherwise replaced by a “new” hypervisor without migrating virtual machines to a standby computer. The old hypervisor partitions the computer that it controls between a source partition and a target partition. The hypervisor and its virtual machines initially run on the source partition, while a new hypervisor is installed on the target partition. The virtual machines are migrated to the new hypervisor without physically moving the in-memory virtual-machine data. Instead, the old hypervisor sends memory pointers, and the new hypervisor claims the respective memory locations storing the virtual-machine data. After all virtual machines are migrated, the old hypervisor bequeaths the hypervisor memory and a last processor that the old hypervisor requires to run. The new hypervisor claims the bequeathed processor and hypervisor memory after the old hypervisor terminates to complete the upgrade/exchange.
摘要:
Page data of a virtual machine is represented for efficient save and restore operations. One form of representation applies to each page with an easily identifiable pattern. The page is described, saved, and restored in terms of metadata reflective of the pattern rather than a complete page of data reflecting the pattern. During a save or restore operation, however, the metadata of the page is represented, but not the page data. Another form of representation applies to each page sharing a canonical instance of a complex pattern that is instantiated in memory during execution, and explicitly saved and restored. Each page sharing the canonical page is saved and restored as a metadata reference, without the need to actually save redundant copies of the page data.