Vmware, Inc.

Digital Equipment Corporation

Hewlett-Packard Development Company, L.p.

Compaq Computer Corporation, Inc.

Compaq Information Technologies Group, L.p.

Daniel J Scales

Michael Nelson

Satyam B Vaghani

Pratap Subrahmanyam

Ganesh Venkitachalam

Kourosh Gharachorloo

Min Xu

Carl Alan Waldspurger

Hao Xu

Boris Weissman

Luiz Andre Barroso

Jeffrey W Sheldon

Vyacheslav Vladimirovich Malyugin

Matthew Benjamin Amdur

Rohit Jain

Peter Chen

Andrew G Tucker

Robert J Stets, Jr

Matthew David Ginzton

Mosur Kumaraswamy Ravishankar

Olivier F Lecomte

Mallik Mahalingam

Mayank Rawat

Murali Vilayannur

Jinyuan Li

Norman Paul Jouppi

Chandramohan Amyangot Thekkath

Deng Liu

Leslie B Lamport

Tariq Magdon-Ismail

Richard McDougall

Lan Huang

Yuen-Lin Tan

Razvan Cheveresan

Eric E Lowe

Abhishek Rai

Wayne R Mack

Dragan Stancevic

Thorbioern Donbaek Jensen

Anshu Aggarwal

System and method for providing fault tolerance in virtualized computer systems use a first guest and a second guest running on virtualization software to produce outputs, which are produced when a workload is executed on the first and second guests. An output of the second guest is compared with an output of the first guest to determine if there is an output match. If there is no output match, the first guest is paused and a resynchronization of the second guest is executed to restore a checkpointed state of the first guest on the second guest. After the resynchronization of the second guest, the paused first guest is caused to resume operation.

Virtual machine fault tolerance

To generate a checkpoint for a virtual machine (VM), first, while the VM is still running, a copy-on-write (COW) disk file is created pointing to a parent disk file that the VM is using. Next, the VM is stopped, the VM's memory is marked COW, the device state of the VM is saved to memory, the VM is switched to use the COW disk file, and the VM begins running again for substantially the remainder of the checkpoint generation. Next, the device state that was stored in memory and the unmodified VM memory pages are saved to a checkpoint file. Also, a copy may be made of the parent disk file for retention as part of the checkpoint, or the original parent disk file may be retained as part of the checkpoint. If a copy of the parent disk file was made, then the COW disk file may be committed to the original parent disk file.

Generating and using checkpoints in a virtual computer system

In a computer system running at least a first virtual machine (VM) and a second VM on virtualization software, a computer implemented method for the second VM to provide quasi-lockstep fault tolerance for the first VM includes executing a workload on the first VM and the second VM that involves producing at least one externally visible output and comparing an externally visible output of the second VM with an externally visible output of the first VM to determine if there is an output match. In response to a determination that the externally visible output of the second VM does not match the externally visible output of the first VM, a resynchronization of the second VM is executed. The externally visible output of the first VM is kept from being output externally until completion of the resynchronization.

An elastic filesystem for temporary data provides storage space for virtual machines (VMs) in a distributed computing system. The filesystem redirects accesses to virtual disks in VMs to a common pool file. The system provides performance and storage efficiency at least on par with local, direct attached virtual disks, while providing a single pool of shared storage that is provisioned and managed independently of the VMs. The system provides storage isolation between VMs storing temporary data in that shared pool. Also, storage space for temporary data may be allocated on demand and reclaimed when no longer needed, thereby supporting a wide variety of temporary space requirements for different Hadoop jobs.

Elastic temporary filesystem

A virtualization platform provides fault tolerance for a primary virtual machine by continuously transmitting checkpoint information of the primary virtual machine to a collector process, such as a backup virtual machine. When implemented on a hardware platform comprising a multi-processor that supports nested page tables, the virtualization platform leverages the nested page table support to quickly identify memory pages that have been modified between checkpoints. The backup virtual machine provides feedback information to assist the virtualization platform in identifying candidate memory pages for transmitting actual modifications to the memory pages rather than the entire memory page as part of the checkpoint information. The virtualization platform further maintains a modification history data structure to identify memory pages that can be transmitted simultaneous with the execution of the primary virtual machine rather than while the primary virtual machine has been stunned.

Method and system for frequent checkpointing

In a computer system running at least a first virtual machine (VM) and a second VM on virtualization software, a computer implemented method for the second VM to provide quasi-lockstep fault tolerance for the first VM. The method includes enabling operations of the second VM to diverge from being in lockstep with operations with the first VM without resynchronizing the second VM with the first VM, if output of the second VM is not diverging from output of the first VM. The output of the second VM is compared with the output of the first VM to determine whether the output of the second VM has diverged from the output of the first VM. In response to a determination that the output of the second VM has diverged from the output of the first VM, a resynchronization of the second VM is executed.

A virtualized computer system provides fault tolerant operation of a primary virtual machine. In one embodiment, this system includes a backup computer system that stores a snapshot of the primary virtual machine and a log file containing non-deterministic events occurring in the instruction stream of the primary virtual machine. The primary virtual machine periodically updates the snapshot and the log file. Upon a failure of the primary virtual machine, the backup computer can instantiate a failover backup virtual machine by consuming the stored snapshot and log file.

Low overhead fault tolerance through hybrid checkpointing and replay

A technique for efficient cache management demotes a unit of data from a higher cache level to a lower cache level in a cache hierarchy when the higher level cache evicts the unit of data. In a virtualization computing environment, eviction of the unit of data may be inferred by observing privileged memory and disk operations performed by a guest operating system and trapped by virtualization software for execution. When the unit of data is inferred to be evicted, the unit of data is demoted by transferring the unit of data into the lower cache level. This technique enables exclusive caching without direct involvement or modification of the guest operating system. In alternative embodiments, a pseudo-driver installed within the guest operating system explicitly tracks memory operations and transmits page eviction information to the lower level cache, which is able to cache evicted pages while maintaining cache exclusivity.

System and method for exclusive read caching in a virtualized computing environment

A method of acquiring a lock by a node, on a shared resource in a system of a plurality of interconnected nodes, is disclosed. Each node that competes for a lock on the shared resource maintains a list of locks currently owned by the node. A lock metadata is maintained on a shared storage that is accessible to all nodes that may compete for locks on shared resources. A heartbeat region is maintained on a shared resource corresponding to each node so nodes can register their liveness. A lock state is maintained in the lock metadata in the shared storage. A lock state may indicate lock held exclusively, lock free or lock in managed mode. If the lock is held in the managed mode, the ownership of the lock can be transferred to another node without a use of a mutual exclusion primitive such as the SCSI reservation.

Hybrid locking using network and on-disk based schemes

Multiple computers are connected to a data storage unit that includes a file system, which further includes multiple data entities, including files, directories and the file system itself. The file system also includes, for each data entity, an owner field for indicating which computer, if any, has exclusive or shared access to the data entity, along with a time field for indicating when a lease of the data entity began. When a computer wants to lease a data entity, the computer uses a disk reservation capability to temporarily lock the data storage unit, and, if the data entity is not currently leased, the computer writes its own identification value into the owner field and a current time into the time field for the data entity, to claim the data entity for a renewable lease period. If a prior lease of a data entity has expired, another computer may break the lease and claim ownership for itself.

Providing multiple concurrent access to a file system

A virtualization technique, in accordance with one embodiment of the present invention, includes emulating the small computing system interface (SCSI) protocol to access a virtual SCSI storage device backed by a file stored on network attached storage (NAS).

SCSI protocol emulation for virtual storage device stored on NAS device

Techniques for enabling a virtual machine (VM) executing on a physical node to access a partition of a storage system are described. One embodiment associates an identifier with the VM for use when communicating with the storage system, wherein the identifier differs from a default identifier of a host bust adapter (HBA) usable by the physical node to communicate with the storage system.

Storage area network access for virtual machines

A method and system is provided for initializing files such as, for example and without limitation, pre-allocated files or raw device mapping (RDM) files, by delaying initializing file blocks. In accordance with one or more embodiments of the present invention, file blocks are associated with corresponding indicators to track un-initialized blocks.

Initializing file data blocks

Growing community of inventors

Mountain View, CA, United States of America

Daniel J Scales