A multi-tenant computing system provides services to a number of different tenant organizations. To address the problem of failure of portions of the system, the hardware infrastructure of the system is located at a number of different geographical locations. The various tenants are assigned to one of a set of “cells,” each cell corresponding to one of the geographical locations. Additionally, each cell has another one of the cells assigned to it as a backup cell, and the data of each cell is replicated within its assigned backup cell. At system run time, if a failure is detected within one of the cells, the network redirection is used within the multi-tenant system to reflect that the backup cell for the failing cell is now handling requests for the failing cell. Upon determination that the failing cell has been repaired and is now again correctly functioning, the network redirection is no longer employed, such that the (formerly) failing cell again handles its own requests.

A copy data management system for a modern application includes a modern application module and a copy data management module. The modern application module includes a service unit, an object storage gateway, a production object storage, and a database. The service unit is configured to generate a large object, a small object, and service metadata. The object storage gateway is configured to write and read the large object and the small object in the production object storage, merge small objects into a large object, and generate object metadata. The database is configured to store the service metadata and the object metadata. The copy data management module includes a metadata processing unit and a disaster recovery object storage. The metadata processing unit is configured to back up the database. The object storage gateway is also configured to back up the production object storage to the disaster recovery object storage.

Synchronization metadata is read from non-volatile storage. The synchronization metadata comprises indications of one or more synchronization targets. A synchronization target is a node of a clustered storage system. A synchronization cache is populated with the synchronization metadata. After populating the synchronization cache with at least a portion of the synchronization metadata, a connection to a network is established. After the connection to the network is established, a connection to a first of the one or more synchronization targets is established.

The present disclosure includes apparatuses and methods related to hybrid memory management. An example apparatus can include a first memory array, a number of second memory arrays, and a controller coupled to the first memory array and the number of second memory arrays configured to execute a write operation, wherein execution of the write operation writes data to the first memory array starting at a location indicated by a write cursor, and place the write cursor at an updated location in the first memory array upon completing execution of the write operation, wherein the updated location is a next available location in the first memory array.

A distributed data storage system using erasure coding (EC) provides advantages of EC data storage while retaining high resiliency for EC data storage architectures having fewer data storage nodes than the number of EC data-plus-parity fragments. An illustrative embodiment is a three-node data storage system with EC 4+2. Incoming data is temporarily replicated to ameliorate the effects of certain storage node outages or fatal disk failures, so that read and write operations can continue from/to the storage system. The system is equipped to automatically heal failed EC write attempts in a manner transparent to users and/or applications: when all storage nodes are operational, the distributed data storage system automatically converts the temporarily replicated data to EC storage and reclaims storage space previously used by the temporarily replicated data. Individual hardware failures are healed through migration techniques that reconstruct and re-fragment data blocks according to the governing EC scheme.

A method for protecting data in a storage system is disclosed. In one embodiment, such a method includes detecting, by a first hardware management console, first battery-on status associated with a first uninterruptible power supply. The method further detects, by a second hardware management console, second battery-on status associated with a second uninterruptible power supply. The method communicates, from the first hardware management console to the second hardware management console, the first battery-on status. The method then triggers, by the second hardware management console, a dump of modified data from memory to more persistent storage upon detecting both the first battery-on status and the second battery-on status. A corresponding system and computer program product are also disclosed.

A computing system providing high-availability access to computing resources includes: a plurality of interfaces; a plurality of sets of computing resources, each of the sets of computing resources including a plurality of computing resources; and at least three switches, each of the switches being connected to a corresponding one of the interfaces via a host link and being connected to a corresponding one of the sets of computing resources via a plurality of resource connections, each of the switches being configured such that data traffic is distributed to remaining ones of the switches through a plurality of cross-connections between the switches if one of the switches fails.

Methods and systems are described for provisioning cloud-based database systems and performing decoupled maintenance. For example, conventional systems may rely on database management systems to provision and modify databases hosted by a service provider. However, for entities operating complex database systems with the need for highly customized cloud infrastructure, database management systems fail to provide the granular customization and the control necessary to create and service these systems. In contrast, the described solutions provide an improvement over conventional database management system architecture by providing direct communication between an entity and its cloud-based database systems via command line prompts or API calls, decoupling database system maintenance from database system provisioning process to increase the speed and granular customization of the database system. Moreover, the disclosed solution leverages machine learning to predict optimal database system provisioning and maintenance processes and resources.

In some examples, a system detects recovery, from an unavailable state, of a communication link between a first storage system that includes a first storage volume and a second storage system that includes a second storage volume that is to be a synchronized version of the first storage volume, where while the communication link is in the unavailable state the second storage volume is in an offline state and the first storage volume is in an online state. In response to detecting the recovery of the communication link, the system sends a first tracking metadata for the first storage volume from the first storage system to the second storage system, and in response to receipt of the first tracking metadata at the second storage system that maintains a second tracking metadata for the second storage volume, the system transitions the second storage volume from the offline state to a controlled online state, and initiates a synchronization process to synchronize the second storage volume with the first storage volume.

A plurality of computing devices are communicatively coupled to each other via a network, and each of the plurality of computing devices is operably coupled to one or more of a plurality of storage devices. A plurality of failure resilient address spaces are distributed across the plurality of storage devices such that each of the plurality of failure resilient address spaces spans a plurality of the storage devices. The plurality of computing devices maintains metadata that maps each failure resilient address space to one of the plurality of computing devices. The metadata is grouped into buckets. Each bucket is stored in a group of computing devices. However, only the leader of the group is able to directly access a particular bucket at any given time.