Abstract: Systems, methods, and computer-readable media are disclosed for an improved database. The systems, methods, and computer-readable media described herein may enhance the response time of databases and improve user experiences. In an example method described herein, a database may receive a search request for data stored in the database. The database may be stored in association with a b-tree to facilitate searching and retrieving data from the database. The method may include allocating a portion of a first memory for read ahead operations and performing the read ahead operations, wherein the read ahead operations comprise loading, into the portion of the first memory, data associated with one or more siblings of a current node, wherein the one or more siblings of the current node are one or more nodes that have a same number of parent nodes as the current node.

Abstract: A replica database may receive a replication instruction from a master. The replica may maintain a version of a collection of data in database pages stored on a storage device. A replication thread may process the instruction. The replication thread may determine that an item affected by the instruction is located in a page that is both loaded into a memory buffer and subject to contention. The page may be modified with information indicating that the instruction has not been processed. A subsequent reader thread may, while processing a request to access the page, apply the instruction and complete processing of the request.

Abstract: A data store may implement on-demand data schema modifications. Request to change a schema for a data set in a data store may be received and a description of the change recorded as part of a schema history for the data set. The request to change the schema may then be acknowledged. When access requests directed to the data set are received at the data store, the schema history for the data set may be evaluated. If the schema history indicates that data that is to be accessed in order to service the data store needs to include one or more changes in the schema history, then a version of the schema that includes the one or more changes may be applied to the data.

Abstract: A database system may implement asynchronous processing for synchronous requests received at the database. A pool of request processing threads may be maintained. As access requests for the database are received from clients, an available request processing thread in the pool may process the access request. The access request may be dependent on a persistent storage I/O operation, such that processing of the access request waits at least until the persistent storage I/O operation is complete. The request processing thread may perform processing operations for the access request so that persistent storage I/O operation is performed and become available to process other access requests. A response processing thread may determine that the persistent storage I/O operation for the access request is complete, and send a response to the client.

Abstract: A lock-free data structure may be used to provide concurrent access to a data retention index for a data store. Different processes may obtain keys to the data retention index from the lock-free data structure. The different processes may update entries in the data retention index corresponding to the keys to identify data objects to retain in the data store. The data retention index may be evaluated to determine data objects not identified in the data retention index and reclaim portions of the data store that store the determined data objects for storing other data.

Abstract: Read-only nodes of a distributed database system may perform replicated database startup for common database storage. A read-only node may request a view of a database from a master node to initialize the read-only node for servicing queries. An indication of the view may be received and portions of data maintained for the distributed database may be read from a common data store maintaining the data for the database according to the indicated view of the database. In-memory data structures at the read-only node for servicing queries may be initialized based on the portions of data read from the common data store. The read-only node may then be made available for servicing queries.

Abstract: A system for updating a database instance with a patch or new binary image. A database instance includes a database engine and a client-side storage layer driver to manage database state information and volume geometry state information associated with client process connections to a volume of a storage layer of a database system during the patching process. The database engine and the client-side storage service driver save database state information and volume geometry state information, execute the new binary image, and restore the updated database engine with the database state information and volume geometry state information. The new binary image may be executed using a fork/exec model, wherein file descriptors used to identify sockets used in existing client connections are preserved, enabling restoration of the client connections without interruption or downtime.

Abstract: Read-only nodes of a distributed database system may perform replicated database startup for common database storage. A read-only node may request a view of a database from a master node to initialize the read-only node for servicing queries. An indication of the view may be received and portions of data maintained for the distributed database may be read from a common data store maintaining the data for the database according to the indicated view of the database. In-memory data structures at the read-only node for servicing queries may be initialized based on the portions of data read from the common data store. The read-only node may then be made available for servicing queries.

Abstract: Read-only nodes of a distributed database system may perform replicated database startup for common database storage. A read-only node may request a view of a database from a master node to initialize the read-only node for servicing queries. An indication of the view may be received and portions of data maintained for the distributed database may be read from a common data store maintaining the data for the database according to the indicated view of the database. In-memory data structures at the read-only node for servicing queries may be initialized based on the portions of data read from the common data store. The read-only node may then be made available for servicing queries.

Abstract: Systems and methods that enable resolving deadlock cycles by victim selection based on cost and/or priority of strongly connected components in a static wait-for graph. As such, victims can be selected by iteratively estimating total deadlock resolution cost (based on cost of strongly connected components) for remaining deadlock participants, selecting the participant with the least estimated total cost as the next victim, and reducing the wait-for graph after removing the victim vertex. The victim selection algorithm can be a cost-based greedy algorithm, wherein during each victim selection iteration, a single optimal victim is selected based on estimated remaining cost.

Abstract: Systems and methods that enable resolving deadlock cycles by victim selection based on cost and/or priority of strongly connected components in a static wait-for graph. As such, victims can be selected by iteratively estimating total deadlock resolution cost (based on cost of strongly connected components) for remaining deadlock participants, selecting the participant with the least estimated total cost as the next victim, and reducing the wait-for graph after removing the victim vertex. The victim selection algorithm can be a cost-based greedy algorithm, wherein during each victim selection iteration, a single optimal victim is selected based on estimated remaining cost.