Vmware, Inc.

Bromium, Inc.

Jacob Gorm Hansen

Vikram Kapoor

Ian Pratt

Gaurav Banga

Kiran Bondalapati

Krzysztof Uchronski

Martin O'Brien

A cluster of computer systems, each of which is configured with a virtualization software layer to support execution of virtual computing instances, includes a first computer system in which a first virtual computing instance is executing, the first computer system including a first local storage unit in which a first log file is stored to capture write operations to a virtual disk of the first virtual computing instance. The cluster also includes a second computer system, networked to the first computer system, in which a second virtual computing instance is executing, the second computer system including a second local storage unit in which a second log file is stored to capture write operations to a virtual disk of the second virtual computing instance and in which a replica of the first log file is stored.

System and method for using local storage to emulate centralized storage

Approaches for creating a template virtual machine. An in-memory state of a virtual machine and/or a set of applications executing within the virtual machine are adjusted and/or configured based on the intended use of the template virtual machine. Thereafter, the virtual machine is established as a template virtual machine. The template virtual machine may be used to create one or more virtual machines using a copy-on-write memory process.

Preparing a virtual machine for template creation

Methods and systems for utilizing local storage of host computers to emulate a centralized storage system comprise receiving a write operation from a virtual machine running on a first host computer, wherein the write operation corresponds to a disk image that is locally stored in a first host computer. The disk image and a second disk image is updated in accordance with the write operation, wherein (i) the second disk image is locally stored on a second host computer that is networked to the first host computer, (ii) the second disk image serves as a back-up to the disk image, and (ii) the second host computer is configured to support the running of other virtual machines that access other disks images that (a) are locally stored on the second host computer, and (b) are further backed-up on at least a third host computer.

Managing the guest operating system's eviction of memory pages from a virtual machine. A guest operating system or a hypervisor may cause one or more memory pages within a guest physical frame to become unlikely or ineligible for selection as a candidate for eviction by the guest operating system. Each of the one or more memory pages may also reside, or be intended to reside, in the memory of one or more other virtual machines. In this way, memory pages that are shared across multiple virtual machines may become less likely to be evicted, thereby using memory more efficiently.

Managing the eviction process

Reducing an amount of memory used by a virtual machine. A system includes multiple virtual machines that share common pages of memory. The number of private pages associated with each virtual machine is minimized by ensuring that pages that a guest operating system regards as now free or zeroed are efficiently mapped by the hypervisor to a shared zero page. Upon a hypervisor determining that one or more guest physical frame numbers are assigned to free memory pages, the hypervisor updates mapping data to map the one or more guest physical frame numbers to a shared zero page within the machine frame.

Mapping free memory pages maintained by a guest operating system to a shared zero page within a machine frame

Methods and system for adding a data record to a file comprise maintaining a data structure to track locations of data records within the file, wherein each entry in the data structure is a Bloom filter that corresponds to a different portion of the file that includes a plurality of data records. When an instruction to write data to the file is received, a data record is generated to append to the file, wherein the data record comprises a randomized unique id and the data from the received instruction. A Bloom filter bit pattern is extracted from the randomized unique id and bits in a current Bloom filter entry of the data structure are set, wherein the set bits correspond to the same bit positions as the Bloom filter bit pattern. The data record is thus able to be added to the file.

Method and system for inserting data records into files

Mitigating eviction of the memory pages of virtualized machines. Upon detecting that a request to perform an I/O operation has been issued against a block stored a disk, a determination is made as to whether a pristine copy of the contents of the block is stored in memory. If a pristine copy of the contents of the block is stored in memory, then the request may be performed by updating mapping data that maps a page of memory to a location in memory at which the pristine copy is stored. In this way, the request is performed without performing the I/O operation against the block stored on disk. Various approaches for resharing memory, including memory of a template virtual machine, are discussed.

Mitigating eviction by maintaining mapping tables

A two dimensional bit array is maintained to keep track of the location of data records in a log file that is organized as a plurality of log segments, each comprising a plurality of data records. The data records are indexed in the log file according to randomized unique ids. Each column of the two dimensional bit array represents a Bloom filter corresponding to a log segment of the log file such that, given a particular randomized unique id, log segments that may contain the data record corresponding to the particular randomized unique id can be efficiently identified by utilizing Bloom filter techniques to analyze the columns of the two dimensional bit array.

Method and system for locating update operations in a virtual machine disk image

A replicated decentralized storage system comprises a plurality of servers that locally store disk images for locally running virtual machines as well as disk images, for failover purposes, for remotely running virtual machines. To ensure that disk images stored for failover purposes are properly replicated upon an update of the disk image on the server running the virtual machine, a hash of a unique value known only to the server running the virtual machine is used to verify the origin of update operations that have been transmitted by the server to the other servers storing replications of the disk image for failover purposes. If verified, the update operations are added to such failover disk images.

System and method for replicating disk images in a cloud computing based virtual machine file system

A replicated decentralized storage system comprises a plurality of servers that locally store disk images for locally running virtual machines as well as disk images, for failover purposes, for remotely running virtual machines. To ensure that disk images stored for failover purposes are properly replicated upon an update of the disk image on the server running the virtual machine, a hash of a unique value known only to the server running the virtual machine is used to verify the origin of update operations that have been transmitted by the server to the other servers storing replications of the disk image for failover purposes. If verified, the update operations are added to such failover disk images. To enable the replicated decentralized system to recover from a failure of the primary server, the master secret is subdivided into parts and distributed to other servers in the cluster. Upon a failure of the primary server, a secondary server receives a threshold number of the parts and is able to recreate the master secret and failover virtual machines that were running in the failed primary server.

Growing community of inventors

Ryomgaard, Denmark

Jacob Gorm Hansen