International Business Machines Corporation

Matthew Albert Huras

Aamer U Sachedina

Keriley Kay Romanufa

Michael Jeffrey Winer

Xun Xue

Catherine Suzanne McArthur

Mark Francis Wilding

Sam Sampson Lightstone

Ronen Grosman

Leslie Anne Cranston

Timothy Ray Malkemus

Kevin R Rose

Adam J Storm

Bishwaranjan Bhattacharjee

Sriram Kolinjivadi Padmanabhan

Christian Marcelo Garcia-Arellano

Roger L Q Zheng

Dale Murray McInnis

Lance Warren Russell

Richard Sefton Sidle

Bruce Gilbert Lindsay

Joseph L Hellerstein

Thomas Stanley Mathews

Sujay Sunil Parekh

Miroslaw Adam Flasza

Jeffrey James Goss

James Liam Finnie

Nelson Hop Hing

Vijayshankar Raman

Bret Ronald Olszewski

Ronald Jason Barber

Mir Hamid Pirahesh

Gary Valentin

Timothy Jon Vincent

Pinar Tozun

Effi Ofer

Sean W McKeough

Andrew David Buckler

Haider Abbas Rizvi

Tony Wen Hsun Lai

Joseph Serge Limoges

Lan Pham

Roger L Zheng

Robert T Berks

Roger Luo Q Zheng

Lorysa M Meadowcroft

Michael Edward Factor

Chandrasekaran Mohan

Garret Frederick Swart

Robert William Lyle

Paul Miller Bird

Thomas James Heller, Jr

David Alan Hepkin

Robert J Blainey

James P Allen

David C Kalmuk

Paula Kim Ta-Shma

Steven Robert Pearson

David C Sharpe

Ramesh Chandra Agarwal

Alexander Abrashkevich

David Tremaine

Dmitri Abrashkevich

Sunil Jeevananda Kamath

Avishay Traeger

Frank-Martin Haas

Walter Duke Alvey

Leo Tat Man Lau

Stewart L Palmer

Kaarel Truuvert

James J Seeger

Leo T M Lau

Herbert Walter Pereyra

Roger Luo Quan Zheng

Matthew James Carroll

Dominique Jeremy Evans

John Paul Kennedy

Quanhua Hong

Wojciech Kuczynski

Sarah Posner

Yogendra K Srivastava

Roger Lq Zheng

Tim J Vincent

David C Smith

Taavi Andrew Burns

Iqbal A Goralwalla

Sam B Lightstone

Sridhar Munireddy

Liam James Finnie

Torsten W E Ziegler

Kelly W Schlamb

Nickolay V Tchervenski

Shaun K Thornborough

Dominque J Evans

Russell M Stocker

Effi Ofar

Dale J Hagen

Stephen A Schormann

One embodiment provides for a method for grooming transactional operations in a multi-master distributed data management system including maintaining transactional operations in at least one transaction-local side-log. A table log scan is performed for a table including recording identification for each record in a log. Tuple sequence numbers are generated for the records and the records are written on groomed transaction blocks to the table.

Transactional operations in multi-master distributed data management systems

One embodiment provides for a method for grooming transactional operations in a multi-master distributed data management system including maintaining transactional operations in at least one transaction-local side-log. A table log scan is performed for a table including recording identification for each record in a log. Each column in the table is loaded with column values based on the identifications. Records based on the column values are written to the groomed transaction blocks that are table based, and the tuple sequence numbers for each record on groomed blocks are generated. The table is shared/sharded across nodes handling transactional operations in the multi-master distributed data management system to increase processing handling of the transactional operations.

Embodiments relate to database systems. An aspect includes deferring row insert operations until occurrence of a triggering event. One method includes receiving a row insert for a tuple into a column group store table, where the tuple includes one or more tuplets and each of the tuplets corresponds to a column group in the column group store table. The method also includes copying at least one of the tuplets into an insert buffer that is specific to one of the column groups in the column group store table. The method also includes deferring the row insert into the column group store table until an occurrence of one or more triggering events. The method also includes flushing the row insert into storage associated with the column group store table, in response to the occurrence of the one or more triggering events.

Buffering inserts into a column store database

Technology for applying locks to memory pages. More specifically, a sticky lock is chosen by matching observed usage patterns with predetermined patterns associated with various sticky locks and/or types of stick locks increase database performance by reducing required communications and information transfers.

Increase database performance by reducing required communications and information transfers

According to embodiments of the present invention, one or more computer processors determine that a predetermined workload threshold associated with an invalidated object is not exceeded and transmits an invalidation command associated with the invalidated object to the owner of the invalidated object. The one or more computer processors instruct the modifier of the invalidated object to retain possession of a first lock on the object beyond the transaction. The one or more computer processors determine that a usage pattern associated with the invalidated object matches a predetermined usage pattern for selecting a lock that can be retained beyond an associated transaction and transmits a second lock a requestor. The one or more computer processors instruct the modifier to release possession of the first lock to the owner and transmit a first image of the current version of the invalidated object to the owner for subsequent transmission to the requestor.

A storage appliance system is disclosed which may include at least one application server for locally executing an application, and one or more storage servers in communication with the application server for I/O transmission therebetween. Also disclosed are an application server, a method, and a computer program product.

Storage appliance, application server and method thereof

Provided are techniques for reclamation of empty pages in database tables. In response to receiving a plurality of records for insertion into a database table, the plurality of records are inserted into one or more contiguous pages. In response to at least some of the plurality of records being deleted from the database table, the one or more contiguous pages are reclaimed by: relocating any records from the plurality of records remaining in the one or more contiguous pages and releasing the one or more contiguous pages.

Reclamation of empty pages in database tables

According to one embodiment of the present invention, a system partitions data for parallel processing and comprises one or more computer systems with at least one processor. The system partitions data of a data object into a plurality of data partitions within a data structure based on a plurality of keys. The data structure includes a plurality of dimensions and each key is associated with a corresponding different dimension of the data structure. Portions of the data structure representing different data partitions are assigned to the computer systems for parallel processing, and the assigned data structure portions are processed in parallel to perform an operation. Embodiments of the present invention further include a method and computer program product for partitioning data for parallel processing in substantially the same manner described above.

Partitioning data for parallel processing

An ordering of operations in log records includes: performing update operations on a database object by a node; writing log records for the update operations into a local buffer by the node, the log records each including a local virtual timestamp; determining that a log flush to write the log records in the local buffer to a persistent storage is to be performed; in response, sending a request from the node to a log flush sequence server for a log flush sequence number; receiving the log flush sequence number by the node; inserting the log flush sequence number into the log records in the local buffer; and performing the log flush to write the log records in the local buffer to the persistent storage, where the log records written to the persistent storage comprises the local virtual timestamps and the log flush sequence number.

Methods to minimize communication in a cluster database system

A method for logical buffer pool extension identifies a page in a memory for eviction, and analyzes characteristics of the page to form a differentiated page. The characteristics of the page include descriptors that include a workload type, a page weight, a page type, frequency of access and timing of most recent access. The method also identifies a target location for the differentiated page from a set of locations including a fastcache storage and a hard disk storage to form an identified target location. The method further selects an eviction operation from a set of eviction operations using the characteristics of the differentiated page and the identified target location. The differentiated page is written to the identified target location using the selected eviction operation, where the differentiated page is written only to the fastcache storage.

Logical buffer pool extension

Lock communications in a shared disk database system is reduced by several approaches: sending a single message to a lock manager for locks required for an operation and locks that will be required in the future, the operations may include row update, row delete, large scan, and row insert; performing an operation in two executions, where in a first execution locks required for the operation is determined without performing the operation, and in a second execution a single message is sent to a lock manager for the required locks and the operation is performed; in free space searching, sending a single message to a lock manager for row and page locks before a row insert is actually performed; and granting locks in an unfair manner.

Approaches to reducing lock communications in a shared disk database

A method, computer program product, and system for detecting and fencing off a failed entity instance so that failover time in the computing system is reduced. Upon detection of a failed entity, for example a failed process, a signal handler invokes an operating system interface to fence off the failed instance from modifying its persistent state, so that the entity may be re-instantiated prior to termination of the failed instance. This approach reduces failover time and eliminates split-brain problems without compromising access to the failed instance to obtain diagnostic information, core dumps, and the like.

Failure detection and fencing in a computing system

According to one embodiment of the present invention, a method of detecting and correcting hot pages in a database system includes monitoring transactions involving pages in a database and determining if one of the transactions had to wait for access to one of the pages; incrementing a page contention counter each time one of the transactions had to wait for access to one of the pages. The method then determines if the page contention counter exceeds a predetermined threshold and monitors transactions on one of the pages for which the page contention count was exceeded. The method then determines if a row was accessed in the page for which the page contention count was exceeded and increments a reference count for the accessed row. The accessed rows are flagged when the reference count exceeds a second predetermined threshold. The flagged rows are moved to another page in the database.

Automatic detection and correction of hot pages in a database system

An adaptive throttling system for minimizing the impact of non-production work on production work in a computer system is provided. The adaptive throttling system throttles production work and non-production work to optimize production. The adaptive throttling system allows system administrators to specify a quantified limit on the performance impact of non-production or utility work on production work. The throttling rate of the utility is then automatically determined by a supervisory agent, so that the utilities' impact is kept within the specified limit. The adaptive throttling system adapts dynamically to changes in workloads so as to ensure that valuable system resources are well utilized and utility work is not delayed unnecessarily.

Adaptive throttling for data processing systems

Growing community of inventors

Ajax, Canada

Matthew Albert Huras