Oracle International Corporation

Delphix Corporation

Vinay H Srihari

Yunrui Li

Roger J Bamford

Wilson Wai Shun Chan

Neil MacNaughton

Sashikanth Chandrasekaran

William Bridge

David Brower

Wei Ming Hu

Juan R Loaiza

Mahesh Baburao Girkar

Wei-Ming Hu

Jia Shi

Mark Dilman

Michael James Stewart

Benedicto Elmo Garin, Jr

Jonghyun Lee

Vsevolod Panteleenko

Matthew Allan Ahrens

J William Lee

Adam H Leventhal

Alok Pareek

Wilson Wai Chan

Ramana Yerneni

Subhadeep Sinha

Neil Macnaughton

Tirthankar Lahiri

Angelo Pruscino

Kumar Rajamani

William Havinden Bridge, Jr

Niloy Mukherjee

Vikram Joshi

Jaebock Lee

Margaret Susairaj

Tak Fung Wang

Arvind Nithrakashyap

Tolga Yurek

Bharat Chandra Baddepudi

Neil J S MacNaughton

Kiran Badrinarain Goyal

Roger J Hansen

Stephen John Vivian

Chi Cao Minh

Michael L Stewart

Hochak Hung

Vipin Gokhale

George H Claborn

Steve Lee

Tudor Bosman

Lawrence HinLuen To, Jr

Theocharis Kampouris

Michael Jennings

Ramana Yemeni

A computer is programmed to identify failures and perform recovery of data. Specifically, in several embodiments, the computer is programmed to automatically check integrity of data in a storage structure to identify a set of failures related to the storage structure. The computer is further programmed in some embodiments to identify, based on one failure in the set of failures, a group of repairs to fix that one failure. Each repair in the group of repairs is alternative to another repair in the group. The computer is also programmed in some embodiments to execute at least one repair in the group of repairs, so as to generate corrected data to fix the one failure. In certain embodiments, the corrected data is stored in non-volatile storage media of the computer.

Data recovery advisor

Each of a plurality of Worker processes are allowed to perform any and all of the following tasks involving logged work items: (1) reading a subset of the work items from a log; (2) sequentially ordering work items for corresponding data objects; (3) applying a sequentially ordered set of work items to a corresponding data object; and (4) transmitting a subset of work items to a Worker process running on another database server in a cluster, if necessary. These tasks can be performed concurrently, at will, and as available, by the Worker processes. An improved checkpointing technique eliminates the need for the Worker processes to get to a synchronization point and stop. Instead, a Coordinator process examines the current state of progress of the Worker processes and computes a past point in the sequence of work items at which all work items before that point have been completely processed, and records this point as the checkpoint.

High-performance log-based processing

High performance log-based parallel processing of logs of work items representing operations on data objects

Virtual databases are provisioned using point-in-time copies of a source database. Transaction logs received from the source database are applied to a point-in-time copy of the source database to create recovery milestones comprising database blocks that can be used to provision virtual databases. The recovery milestones are created in advance of receiving a request to provision a virtual database. Each recovery milestone is validated to determine that a consistent database snapshot can be created using the recovery milestone. If a problem is identified that prevents creation of a consistent database snapshot, the problem is fixed before creating the next recovery milestone. In response to receiving a request to provision a virtual database corresponding to a point in time, a recovery milestone is identified and transaction logs applied to the recovery milestone to create a database snapshot corresponding to the requested point in time.

Creating validated database snapshots for provisioning virtual databases

Techniques for processing changes in a cluster database system are provided. A first instance in the cluster transfers a data block to a second instance in the cluster before a redo record that stores one or more changes that the first instance made to the data block is durably stored. The first instance also transfers, to the second instance, a block change timestamp that indicates when a redo record for the one or more changes was generated by the first instance. The first instance also separately sends, to the second instance, a last store timestamp that indicates when the last redo record that was durably stored was generated by the first instance. The block change timestamp and the last store timestamp are used by the second instance when creating redo records for changes (made by the second instance) that depend on the redo record generated by the first instance.

Data transfers between cluster instances with delayed log file flush

Techniques are provided for managing caches in a system with multiple caches that may contain different copies of the same data item. Specifically, techniques are provided for coordinating the write-to-disk operations performed on such data items to ensure that older versions of the data item are not written over newer versions, and to reduce the amount of processing required to recover after a failure. Various approaches are provided in which a master is used to coordinate with the multiple caches to cause a data item to be written to persistent storage. Techniques are also provided for transferring data items and locks associated with the data items from one node to another.

Methods to perform disk writes in a distributed shared disk system needing consistency across failures

A method, system, and computer program product for adaptive high-performance database redo log synchronization. The method commences upon performing a write operation of a redo log entry, the write operation concluding with an indication of completion of the write operation of the redo log entry. Any number of committing processes may be waiting for the indication of completion, and upon indication of completion, then (using a first synchronization mode) the processes or proxy measures the waiting time as experienced by the committing processes (e.g., while waiting for the indication of completion of the write operation of the redo log entry). In some cases a second synchronization mode would introduce less latency than the first synchronization mode, so the system changes to a second synchronization mode. The system can also change mode when a predicted second mode waiting time is smaller than the measured waiting time.

Adaptive high-performance database redo log synchronization

Techniques for moving data files without interrupting access are described. A first process moves a database file from a first location to a second location while the database file is accessible to one or more other processes for read or write operations. According to one technique, the first process communicates a move status and a copy range into the database file to one or more database server instances executing the one or more other processes. The one or more other processes then perform input/output (IO) operations on the database file based at least in part on the move status and the copy range communicated by the first process.

Techniques for moving data files without interrupting access

A method and system for replicating database data is provided. One or more standby database replicas can be used for servicing read-only queries, and the amount of storage required is scalable in the size of the primary database storage. One technique is described for combining physical database replication to multiple physical databases residing within a common storage system that performs de-duplication. Having multiple physical databases allows for many read-only queries to be processed, and the de-duplicating storage system provides scalability in the size of the primary database storage. Another technique uses one or more diskless standby database systems that share a read-only copy of physical standby database files. Notification messages provide consistency between each diskless system's in-memory cache and the state of the shared database files. Use of a transaction sequence number ensures that each database system only accesses versions of data blocks that are consistent with a transaction checkpoint.

Reduced disk space standby

An improved approach is described for implementing transformations of data records in high concurrency environments. Each transformation is performed in parallel at the source when the data record is first generated. According to one approach for data integrity validation, record generators compute an integrity checksum for a newly generated record before copying into a data unit in shared memory. Subsequent generators may aggregate integrity checksums for data records into checksums for data units incrementally. This approach achieves end-to-end protection of data records against corruption using an efficient method of maintaining verifiable data integrity. In another approach, compression and encryption data transformations may be performed by themselves, or in combination with an integrity checksum transformation.

Method and system for implementing parallel transformations of records

A method and system is provided for reducing delay to applications connected to a database server that guarantees no data loss during failure or disaster. After storing a log record persistently in a local primary log, the log writer returns control to the application which continues running concurrently with the database server sending the session's log records to a standby database. A separate back channel is used by the standby to communicate, out-of-band to the primary, the location of the last log record stored persistently to the standby log. An application waiting for a transaction to commit may wait until the transaction's commit record has been persisted. Also described is a technique for reducing application delay when there is contention between nodes of a multi-node cluster for updating the same block. The technique provides for an asynchronous ping protocol that guarantees zero data loss during failure or disaster.

Method for maximizing throughput and minimizing transactions response times on the primary system in the presence of a zero data loss standby replica

A method and system is provided for measuring, guaranteeing, and reducing replication data lag time between a primary system and one or more standby systems. Each standby system determines the lag time between the generation of a consistent version of data on the primary system and the time that the consistent version is applied on the standby system. Applications can request and be guaranteed to receive data from a standby system that is identical to the state on the primary system at the time of the query, or lag the primary state only by a maximum tolerable amount. A standby system may also publish a service that guarantees a maximum lag time and withdraw the service offer when the actual lag time exceeds the guaranteed lag time. Implications for implementing synchronous and asynchronous replication as well as performance optimizations are also discussed.

Controlling data lag in a replicated computer system

A method and apparatus for detecting split brain in a distributed system is provided. After determining that a rogue instance is no longer an active member of the cluster, a recovery instance detects activity associated with a redo log that is updated by the rogue instance to store log records that describe changes made by the rogue instance to data associated with the cluster.

Methods for detecting split brain in a distributed system

Growing community of inventors

San Mateo, CA, United States of America

Vinay H Srihari