International Business Machines Corporation

Steven Robert Pearson

Richard Sefton Sidle

Adam J Storm

Chandrasekaran Mohan

Roger L Q Zheng

Walter Duke Alvey

Gary Nanhao Jin

Prasadarao Akulavenkatavara

Kevin James Cherkauer

Madhusudan Kalarikal Janardhana

Andanagouda F Patil

Xun Xue

James John Seeger, Jr

Ronald Jason Barber

Yi Ding

Scott David Lashley

Dale Murray McInnis

Effi Ofer

Matthew Albert Huras

Vincent Kulandaisamy

Jon A Lind

Yuke Zhuge

James J Seeger

Steve Raspudic

David C Smith

Sam B Lightstone

Aaron I Henner

Techniques for data replication from a primary system to a standby system. A first buffer portion of a transaction log buffer is allocated to a first transaction at a primary system. Upon determining that the first buffer portion is fully formed, before the log data is written to the storage on the primary system, and before the transaction log buffer is entirely full, the log data in the first buffer portion is transmitted to the standby system, from the primary system. Upon writing the log data to the storage on the primary system and after the log data in the first buffer portion is transmitted to the standby system, a notification is transmitted to the standby system, from the primary system, where the standby system is configured to process the received log data responsive to receiving the notification from the primary system.

Log-shipping data replication with early log record fetching

Provided are techniques for using parallel inserts to insert into a column store. A list of sub-ranges for a columnar table is created. A change statement for one or more rows to be inserted into the columnar table is received. For each of the received change statements, a sub-range from the list of sub-ranges is identified and data is serially inserted into the identified sub-range of the columnar table, while allowing parallel processing of each change statement for different sub-ranges of the columnar table.

Using parallel insert sub-ranges to insert into a column store

According to one embodiment of the present invention, a method for processing a query in a column database is provided. The method for analyzing test results may include a computer, associating a first set of metadata values with a set of records in a table, wherein each record has an associated state indicated by a metadata value. The method may further include the computer receiving a query to return one or more records of the set of records. The method may further include the computer determining that one or more records in the set of records satisfies the query. The method may further include the computer determining whether at least a first record in the one or more records that satisfies the query is in a returnable state based, at least in part, on the metadata value.

Query handling in a columnar database

An approach for synopsis guided scan processing of MIN/MAX data queries where a minimal or maximal data query operation and a synopsis store are received. The synopsis store includes at least one of tuple range identifiers, column minimal and maximal metadata values. The synopsis store is sorted, where the column minimal or maximal metadata values search order is created for the tuple range identifiers. Dataset data associated with the tuple range identifiers is searched in the search order for a minimal or maximal value result until the minimal value result is less than or equal to a next column minimal metadata values or the maximal value result is greater than or equal to a next column maximal metadata values respectively and the minimal or maximal value result associated with the respective minimal or maximal data query operation is output.

Min/max query with synopsis guided scan order

Methods, systems and program products for log-shipping data replication from a primary system to a communicatively-coupled standby system. Embodiments of the invention may receive transactional log data at a standby system, from the primary system, and before the transactional log data is written to storage on the primary system. Embodiments may then receive a notification from the primary system indicating that the corresponding log data was written to storage on the primary system, and responsive to receiving the notification, may process the received transactional log data.

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

MIN/MAX query with synopsis guided scan order

A method of optimizing the number and type of database backups to achieve a given RTO is provided and may include receiving a RTO and receiving a heuristic for determining an amount of unencumbered processing time. A type of next backup, (i.e., a next backup), is determined wherein the type of next backup is an incremental backup when the sum of the heuristic, and the times to: restore the latest full backup, restore zero or more incremental backups, complete a current incremental backup, and perform a full backup is less than the received RTO, else the type of the next backup is a full backup. A time to schedule the next backup is scheduled based on the received RTO being a total of an amount of time to: complete the type of next backup; rollforward zero or more transaction log records; and to restore at least one backup.

Optimizing the number and type of database backups to achieve a given recovery time objective (RTO)

A method of classifying and monitoring database operations based on a recovery cost may include receiving an indication of a recoverable operation. A count in a persistent storage, such as a catalog, corresponding to an occurrence of the recoverable operation is incremented.

Classifying and monitoring database operations based on a cost of recovery

Growing community of inventors

Portland, OR, United States of America

Steven Robert Pearson