Abstract: User-based recovery point objectives (RPOs) for disaster recovery are described herein. A method as described herein can include obtaining, by a device operatively coupled to a processor, transient information associated with a file stored by a data storage system; determining, by the device, whether the transient information associated with the file indicates that a condition for replicating the file has been met; and inserting, by the device, the file into a replication queue associated with the data storage system in response to a positive result of the determining.

Abstract: Systems and methods for a service based disaster recovery system are disclosed. Embodiments may include the ability to configure and deploy a DR environment, including providing the ability to configure a DR service in the DR environment for one or more deployed primary services in a primary environment. An environment management database holds DR configuration data including the status of the deployed services. An environment manager may interact with the environment management database to determine an associated action for the services. The services may perform activation (e.g., wake up) or enter a standby mode (e.g., sleep) depending on the determined action.

Abstract: A device receives a notification indicating a failure of a first server device responsible for a primary message queue that includes messages at a time of the failure. A second server device is responsible for a standby message queue to which the messages are replicated, where a position in the standby message queue and a message time are assigned to each of the replicated messages. The device obtains a record time that identifies the message time of one of the messages that was last obtained from the primary message queue prior to the failure, compares an adjusted record time and the message time of one or more of the messages of the standby message queue to determine a starting position in the standby message queue, and processes messages obtained from the standby message queue beginning at one of the messages assigned to the position that matches the starting position.

Abstract: For error handling of data communications, in a transmission interval, between first and second tasks for which first and second time intervals are respectively predefined, (1) execution of the first task is omitted in a pending instance of the second time interval responsive to where the transmission interval immediately prior to the pending instance of the second time interval began in, and continued past an end point of, a most recent instance of the first time interval, which was during an immediately preceding instance of the second time interval; or (2) execution of the second task is omitted in the pending instance of the second time interval responsive to where a most recent prior execution of the second task began in, and continued past an end point in time of, a most recent instance of the second time interval immediately prior to the pending instance of the second time interval.

Abstract: Techniques for managing a storage system involve: in response to a storage device in a group of storage devices failing, determining multiple stripes in the storage system which are associated with the failed storage device; selecting a first storage device and a second storage device from the group of storage devices respectively, the first storage device and the second storage device being a first candidate and a second candidate of storage devices for rebuilding a first stripe and a second stripe among the multiple stripes to form a first rebuilt stripe and a second rebuilt stripe respectively; obtaining a coupling degree between the first rebuilt stripe and the second rebuilt stripe, the coupling degree describing a conflict degree of rebuilding the first stripe and the second stripe in parallel; and identifying the first candidate and the second candidate on the basis of the obtained coupling degree.

Abstract: For error handling of data communications between first and second tasks in a data transmission interval, where first time intervals and second time intervals are predefined for the first and second tasks, respectively, the data transmission interval is omitted in one of the second time intervals when (1) execution of the first task immediately prior to the current second time interval, which began in a first time interval, during an immediately preceding second interval, continues past an end point of that first time interval, and an execution of the second task of the current second interval has begun, or (2) execution of the second task, which began in an immediately preceding one of the second intervals, continued past an end point of the preceding second interval and an execution of the first task the current second interval has already begun.

Abstract: A method, computer program product, and computer system to maintain high availability of a service processor. An embodiment provides program code with a location of a second service processor (the second service processor is communicatively coupled to the first service processor). The program code stops a virtual machine during runtime, including instruction execution and IO operations, where during runtime, the virtual machine executes one or more processes to service and manage computing resources in the distributed computing environment. The program code generates a micro-checkpoint of the virtual machine. The program code resumes the instruction execution of the virtual machine and transmits the micro-checkpoint to a second service processor based on the location and then resumes IO operations. The second service processor utilizes the micro-checkpoint to enable a hypervisor on the second service processor to start a virtual machine on the second service processor.

Abstract: Techniques for providing metadata (and/or data) protection in a data storage system. The techniques can include storing, for a specific data page, two (2) instances of a main metadata page and one (1) instance of a corresponding alternative metadata page in a protected storage object referred to herein as the “2+1-way mirror”. For the specific data page, two (2) instances of a main metadata page can be stored on a first drive and a second drive, respectively, of the 2+1-way mirror such that the first and second drives each have a copy of the same main metadata page. Further, an instance of a corresponding alternative metadata page can be stored on a third drive of the 2+1-way mirror. The 2+1-way mirror provides protection against metadata loss due to the concurrent failure of two (2) physical drives, and further provides protection against catastrophic metadata errors and/or inconsistencies due to software failures.

Abstract: An apparatus for core repair includes a failure analysis and recovery (“FAR”) probe that accesses a core of a processor and units of the core over a low-level communication bus while the core is operational after a failure notification. The FAR probe compares operational data of the core versus vital product data (“VPD”) while the core is running tests and a thermal, power, functional (“TPF”) workload to determine if the core is in a degraded state and runs tests to identify a failure after determining that the core is in a degraded state. The FAR probe adjusts parameters of the core in response to identifying a failure of the core and re-evaluates the core to determine if the core is functional. The FAR probe returns the core to service after determining that the core is functional. The FAR probe operates independent of other processor cores while the cores are operational.

Abstract: A technique for repairing an indirect addressing structure of a file system damaged by corruption of a mid-level mapping (MID) page includes scanning selected leaf pages to identify leaf pages associated with the corrupted MID page, then recreating the MID page by recreating pointers to the identified leaf pages. The scanning includes (1) based on an association of groups of leaf pages with corresponding sets of families of storage objects, scanning the leaf pages of only those groups of leaf pages associated with the family of storage objects for the corrupted MID page. The scanning further includes (2) performing a two-pass process including first identifying all leaf pages for the logical offset range of the corrupted MID page and then pruning those identified leaf pages that are reachable via non-corrupted MID pages, yielding the leaf pages for the corrupted MID page only, usable to recreate the corrupted MID page.

Abstract: A method of distributed management of recovery of multi-controller NVMe drives includes detecting a path failure of a PCIe path from a first storage node to a first controller on the multi-controller NVMe drive, and initially attempting to correct the path failure using a controller level reset. If the controller level reset is unsuccessful, an alternative path to the controller is sought, and if that is unsuccessful a drive level reset operation is coordinated by all storage nodes with controllers executing on the NVMe drive. To coordinate reset of the NVMe drive, one storage node is elected master. Each node (both slave and master) initiates quiescing of IO operations on its respective controller, and after quiescing has completed, initiates shutdown of its respective controller. Once all controllers are shut down, the master initiates reset of the NVMe drive. Timeouts are used to constrain completion of the quiescing and shutdown operations.

Abstract: Embodiments described herein provide a system for facilitating a training system for a device. During operation, the system determines a system model for the device that can be based on empirical data of the device. The empirical data is obtained based on experiments performed on the device. The system then generates, from the system model, synthetic data that represents behavior of the device under a failure. The system determines uncertainty associated with the synthetic data and, from the uncertainty, determines a set of prediction parameters using an uncertainty quantification model. The system generates training data from the synthetic data based on the set of prediction parameters and learns a set of learned parameters associated with the device by using a machine-learning-based classifier on the training data.

Abstract: Various embodiments of the present disclosure generally relate to a method and a device for managing a hybrid storage disk array. Specifically, the method can include: in response to a first data portion of a first extent in a first set of extents of a first type of storage disk being unavailable, obtaining a metadata portion of a second extent in a second set of extents of the second type of storage disk. The method can also include: reading from the metadata portion reconstruction information for reconstructing the first data portion; and reconstructing, based on the reconstruction information, the first data portion on a third extent in the first set of extents. In addition, there is also provided accordingly a system, an apparatus and a computer program product.

Abstract: A method for detecting a failure in an electronic signal processing system having a signal processing path comprises a configurable functional unit for performing a given function and at least one redundant version of the signal processing path including a corresponding configurable functional unit for performing the given function and configuring a first operating point for the functional unit in the signal processing path for performing the given function and configuring a second operating point for the corresponding functional unit in the redundant version of the signal processing path. The second operating point is different from the first operating point. The method further comprises applying a same input signal to the functional unit and the corresponding functional unit, comparing a first output signal produced by the functional unit with a second output signal produced by the corresponding functional unit, and deriving a failure indication from the comparing.

Abstract: In one aspect, a computerized method for managing consistency and availability tradeoffs in a real-time operational database management system (DBMS) includes the step of implementing consistency in an AP mode of the real-time operational DBMS by implementing the following steps. The method adds a set of schemes that enable a real-time operational DBMS to linearize read/write operations in all situations except a first specified situation and a second specified situation. The real-time operational DBMS is in AP mode, at least one master node for every data item is available in the database cluster of the real-time operational DBMS at all times. The method implements a CP mode of operation.

Abstract: Techniques are directed to failure recovery of a storage system. In accordance with certain techniques, in response to detecting that a disk group of a memory system failed, failure duration of the disk group is recorded. If the failure duration does not reach a predetermined ready time limit and the disk group is in a degraded state, the disk group is maintained in a degraded but not ready state. The predetermined ready time limit is shorter than a logic unit number debounce time limit to avoid a data unavailable event. With such techniques, the possibility of occurrence of a data loss event may be reduced significantly while avoiding a data unavailable event.

Abstract: There is described an apparatus for diagnosing a problem in a simulator comprising a plurality of components, comprising: a communication unit; a memory having stored thereon a database containing a plurality of lists of events each associated with a respective anomaly of the simulator and a respective source of anomaly; a processing unit configured for: receiving a detected anomaly of the simulator via the communication unit; retrieving from the database at least a given one of the plurality of lists of events that correspond to the detected anomaly; receiving an actual state of operation for at least some of the plurality of components; identifying a source of the detected anomaly by comparing the received actual state of operation and at least a given one of the plurality of lists of events; and outputting the source of the detected anomaly via the communication unit.

Abstract: There is described an apparatus for preventing a problem in a simulator comprising a plurality of components, comprising: a communication unit for at least one of transmitting data and receiving data; a memory having stored thereon a database containing a plurality of lists of events each associated with a respective source of anomaly; a processing unit configured for: receiving an actual state of operation for at least some of the plurality of components; determining that a given one of the plurality of lists of event has partially occurred using the actual state of operation; and outputting an alert indicative of the respective source of anomaly associated with the given one of the plurality of lists of event via the communication unit.

Abstract: An accelerator manager monitors hardware accelerators that are called by one or more computer programs. A virtual function table includes multiple entries, where each entry correlates a call from a computer program to a corresponding call to either a software library or a hardware accelerator. A call by the computer program to a function in the virtual function table results in the call being routed to either the software library or to a hardware accelerator depending on the contents of the corresponding entry in the virtual function table. The accelerator manager, in response to a detected failure in an accelerator, replaces one or more calls in the virtual function table to the failed accelerator with calls to the software library. The accelerator manager can then retry the call that caused the accelerator to fail, which will then be executed by the software library.

Abstract: Methods and systems for recovering a host image of a client machine to a recovery machine comprise comparing a profile of a client machine of a first type to be recovered to a profile of a recovery machine of a second type different from the first type, to which the client machine is to be recovered, by a first processing device. The first and second profiles each comprise at least one property of the first type of client machine and the second type of recovery machine, respectively. At least one property of a host image of the client machine is conformed to at least one corresponding property of the recovery machine. The conformed host image is provided to the recovery machine, via a network. The recovery machine is configured with at least one conformed property of the host image by a second processing device of the recovery machine.