Ncr Corporation

Microsoft Technology Licensing, LLC

Teradata Us, Inc.

Other

Nec Corporation

Teradota Us, Inc.

Jeffrey F Naughton

Curt J Ellmann

Gang Luo

Michael Warren Watzke

Karthikeyan Ramasamy

Prasad Manikarao Deshpande

Amit Shukla

Alan D Halverson

Srinath Shankar

Eric R Robinson

Vahit Hakan Hacigumus

Yun Chi

Nikhil Teletia

Jignesh M Patel

Wentao Wu

Robert Hedgcock

Timothy J Keane

The present invention extends to methods, systems, and computer program products for formulating global statistics for parallel databases. In general, embodiments of the invention merge (combine) information in multiple compute node level histograms to create a global histogram for a table that is distributed across a number of compute nodes. Merging can include aligning histogram step boundaries across the compute node histograms. Merging can include aggregating histogram step-level information, such as, for example, equality rows and average range rows (or alternately equality rows, range rows, and distinct range rows), across the compute node histograms into a single global step. Merging can account for distinct values that do not appear at one or more compute nodes as well as distinct values that are counted at multiple compute nodes. A resulting global histogram can be coalesced to reduce the step count.

Optimizing queries using global statistics for distributed databases

Systems and methods for predicting query execution time for concurrent and dynamic database workloads include decomposing each query into a sequence of query pipelines based on the query plan from a query optimizer, and predicting an execution time of each pipeline with a progress predictor for a progress chart of query pipelines.

System and methods for predicting query execution time for concurrent and dynamic database workloads

Distributing rows of data in a distributed table distributed across a plurality of nodes. A method includes identifying skewed rows of a first table to be distributed in a distributed database system. The skewed rows include a common data value in a column such that the skewed rows are skewed, according to a predetermined skew factor, with respect to other rows in the first table not having the common data value. Non-skewed rows of the first table that are not skewed according to the skew factor are identified. The skewed rows of the first table are distributed across nodes in a non-deterministic fashion. The non-skewed rows of the first table are distributed across nodes in a deterministic fashion. The rows of the first table distributed across the nodes, whether distributed in a deterministic fashion or non-deterministic fashion, are stored in a single table at each of the nodes.

Skew-aware storage and query execution on distributed database systems

Formulating global statistics for distributed databases

A database system is capable of processing a query and providing a progress indicator of the processing of the query based on at least a first cost factor. Initially, an estimate is set for the first cost factor. During processing of the query, the estimate of the first cost factor is updated to enable refinement of the progress indicator.

Providing a progress indicator in a database system

A method or apparatus for use with a database system that stores a join view associated with plural base relations includes receiving modification operations to modify at least two of the base relations of the join view, and re-ordering the received modification operations to avoid concurrent execution of modification operations of more than one of at least two base relations.

Rescheduling of modification operations for loading data into a database system

Auxiliary relations are used to maintain a materialized view. The materialized view comprises results of some operation performed between two or more base relations of a parallel database management system comprising a plurality of nodes. The two or more base relations are distributed to the nodes according to some partitioning strategy. Each auxiliary relation is partitioned according to the join attribute in the materialized view definition. During join or other operations involving the relations, the auxiliary relations, not the base relations, are used to construct join results and to maintain the materialized view.

Auxiliary relation for materialized view

A database system includes a rescheduler that performs rescheduling of transactions in a scheduling queue based on one or more criteria. One criterion is whether a first transaction in the queue can be performed as a synchronized scan of a table with a second transaction that is currently executing. This enables the first and second transactions to share a segment of a buffer storage pool.

Rescheduling table scan transactions

A database system receives a transaction that causes a tuple to be integrated into a join view. In response to receiving the transaction, a predefined first lock is placed on at least a portion of the join view, the predefined first lock conflicting with either a shared lock or an exclusive lock placed on the join view, but not conflicting with another predefined first lock placed on the join view. Also, a latch or semaphore associated with a value of a predetermined attribute of the tuple is obtained before integrating the tuple into the join view.

Locking mechanism for materialized views in a database system

A database system includes a locking mechanism for a view defined on one or more base relations. A B-tree index is defined on the view. In response to updates of a base relation, predefined locks are placed on the view. With locking mechanisms according to some embodiments of the invention, concurrency of transactions that access the view and B-tree index is enhanced by enabling concurrent updates of the materialized view by plural transactions in certain cases, while avoiding read-write and write-read conflicts.

Locking mechanism for views associated with B-tree indexes

A database system includes a locking mechanism for a materialized view defined on base relations. In response to updates of a base relation, a name lock is placed on the materialized view. The name lock is a type of exclusive lock and is associated with a data structure containing a first parameter to identify a base relation of the materialized view being updated, and a second parameter to indicate a number of transactions updating the base relation. With locking mechanisms according to some embodiments of the invention, transaction concurrency is enhanced by enabling concurrent updates of a materialized view by plural transactions in certain cases.

Locking mechanism employing a name lock for materialized views

A database system includes a locking mechanism for a materialized view defined on base relations. In response to updates of a base relation, a predefined lock (e.g., Y lock or IY lock) is placed on the materialized view. With the locking mechanism, transaction concurrency is enhanced by enabling concurrent updates of a materialized view by plural transactions in certain cases.

Locking mechanism using a predefined lock for materialized views in a database system

A parallel hash ripple join algorithm partitions tuples of two relations for localized processing. The algorithm is non-blocking and may be performed in a parallel, multi-processor environment. At each processing node, the tuples are further partitioned such that join operations may be performed as tuples are redistributed to each node during the partitioning.

Method and apparatus for performing hash join

A database system includes a transaction rescheduling mechanism that reorders pending transactions based on whether one or more of the pending transactions conflict with any transactions that have been scheduled for execution. In one arrangement, the received pending transactions are stored in a first queue. The transactions that are scheduled for execution are stored in a second queue. The transactions are moved from the first queue to the second queue based on the reordering by the transaction rescheduling mechanism.

Growing community of inventors

Madison, WI, United States of America

Jeffrey F Naughton