Abstract: Systems, methods, and computer readable media for snapshotting ephemeral and/or volatile storage systems are provided. These techniques may include synchronizing that state of the ephemeral storage system to a cloud-based storage system and capturing a cloud snapshot of the cloud-based storage system. To reduce the time to capture a cloud snapshot, techniques disclosed herein include obtaining sets of pending data changes to the ephemeral storage system that have not been synchronized to the cloud-based storage system. When the pending data changes satisfy a threshold metric, the pending data changes may then be synchronized to the ephemeral storage system on a rolling basis.

Abstract: Systems, methods, and computer readable media for snapshotting ephemeral and/or volatile storage systems are provided. These techniques may include synchronizing that state of the ephemeral storage system to a cloud-based storage system and capturing a cloud snapshot of the cloud-based storage system. To reduce the time to capture a cloud snapshot, techniques disclosed herein include obtaining sets of pending data changes to the ephemeral storage system that have not been synchronized to the cloud-based storage system. When the pending data changes satisfy a threshold metric, the pending data changes may then be synchronized to the ephemeral storage system on a rolling basis.

Abstract: Systems, methods, and computer readable media for preventing data loss at ephemeral and/or volatile storage of a local storage system are provided. These techniques may include synchronizing that state of the ephemeral storage system to a cloud-based storage system and capturing a cloud snapshot of the cloud-based storage system. In the event of a failure at the volatile storage, the cloud-based snapshot can be used as a restore point for the cloud-based storage system, the state of which can then be synchronized back to the volatile storage. Additionally, the local storage system includes durable storage for storing transaction logs. After synchronizing the state of the cloud-based storage system to the volatile storage, the local storage system can playback transactions in the transaction log to restore the volatile segment to the state at the time of the failure.

Abstract: This disclosure relates to scope-based lock management for shared resources of a distributed computing system. A scope-based lock manager may grant a request for a lock of a shared resource of the distributed computing system, where the lock request includes a scope of the lock and a target having the scope. The target may define entities able to access the shared resource when locked. The scope and target of the lock are stored at a lock database including scopes and targets associated with additional locks of additional shared resources of the distributed computing system. The scope based lock manager may determine when a computing node of the distributed computing system has failed and remove the lock from the lock database responsive to a determination that the target of the lock is hosted on the computing node.

Abstract: Systems and methods for monitoring computing resource usage and computing task progression. The monitoring includes displaying computing system metrics by animating curved bands in a display screen. A method embodiment commences upon identifying computing system metrics that are associated with computing resource usage metrics and/or computing task progression metrics, then assigning one or more curved bands to the computing resource usage and/or computing task progression metrics. The curved bands are configured to be displayed in a user interface. The curved band configuration includes determining graphical characteristics of the curved bands, where the graphical characteristics are determined based on aspects of the system metrics. State parameter values corresponding to the system metrics change over time, which changes are animated by displaying changes via the graphical characteristics of the one or more curved bands to visually indicate usage and/or progress.

Abstract: According to some embodiment, a backup storage system receives, by a first phase of a data deduplication pipeline, a request from a client for reading or writing a data segment associated with a data stream stored in or to a storage system. In response to the request, the system retrieves, by the first phase, load parameters associated with a second phase in the data deduplication pipeline. For each of the load parameters associated with the second phase, the system determines, by the first phase, whether the load parameter has exceeded a load threshold associated with the second phase. The system throttles, by the first phase, performance of a specific job in the data deduplication pipeline by the second phase in response to a determination that at least one of the load parameters associated with the second job phase has exceeded the load threshold associated with the second phase.

Abstract: According to some embodiment, a backup storage system receives a plurality of input/output (IO) requests at the storage system. The IO requests include random IO requests and sequential IO requests. The storage system determines whether there is a pending random IO request from the plurality of IO requests. In response to determining that there is a pending random IO request, the storage system determines whether a total latency of the sequential IO requests exceeds a predicted latency of the pending random IO request. The storage system services the pending random IO request in response to determining that the total latency of the sequential IO requests exceeds the predicted latency of the pending random IO request.

Abstract: A method and system for dynamically allocating memory, the method comprising maintaining a record of allocated memory that should be considered free in a child process, and designating as free any copy of memory identified by the record and created in a forking process.

Abstract: A process replication method for use in a multi-threaded computing environment, for improving application performance, reducing the number of deadlocks and fork safing code. The method comprises acquiring and locking available mutexes from a plurality of mutexes, determining whether any unavailable mutexes from the plurality of mutexes are held by a thread that is waiting for a mutex owned by a forking thread and, if so, ceding any such mutex to the respective waiting thread and boosting priority of the waiting thread, and registering the forking thread as first waiter for any unavailable mutexes. All of the plurality of mutexes are thereby eventually acquired and locked.

Abstract: A method and system for dynamically allocating memory, the method comprising maintaining a record of allocated memory that should be considered free in a child process, and designating as free any copy of memory identified by the record and created in a forking process.

Abstract: A process replication method for use in a multi-threaded computing environment, for improving application performance, reducing the number of deadlocks and fork safing code. The method comprises acquiring and locking available mutexes from a plurality of mutexes, determining whether any unavailable mutexes from the plurality of mutexes are held by a thread that is waiting for a mutex owned by a forking thread and, if so, ceding any such mutex to the respective waiting thread and boosting priority of the waiting thread, and registering the forking thread as first waiter for any unavailable mutexes. All of the plurality of mutexes are thereby eventually acquired and locked.