International Business Machines Corporation

Roity Prieto Perez

Neal E Bohling

Terri A Menendez

Helen Maria Witter

Austin Willoughby

Debra J Benjamin-Deckert

Tony Xu

Lawrence L Law

Brian Lee

Elaine Lai

Gary Pizl

In one embodiment, a method includes receiving an unstructured data record and parsing the unstructured data record to determine a characterization of the unstructured data record. The characterization includes determination of: a number of fields, a length for the fields, and an order of the fields as stored in the unstructured data record. The method also includes obtaining a first key pair that includes a primary key-name associated with a first key-value and hashing at least the first key-value to obtain a hash value. In addition, the method includes updating the unstructured data record to include the hash value as an indexing key thereby creating a modified data record. Also, the method includes storing the modified data record in a structured database. Moreover, the method includes updating an index of the structured database to include an entry for the modified data record, the entry including the hash value.

Storing unstructured data in a structured framework

Selectively committing or rolling-back in-flight units of recovery is provided. An indicator is read in a transaction identifier information record corresponding to a unit of recovery that is in-flight. It is determined whether the indicator indicates a commit for the unit of recovery that is in-flight. In response determining that the indicator does indicate the commit for the unit of recovery that is in-flight, the unit of recovery that is in-flight corresponding to the transaction identifier information record is committed to form a committed unit of recovery. The committed unit of recovery corresponding to the transaction identifier information record is sent to a target system for further processing.

Replicating changes written by a transactional virtual storage access method

In one embodiment, a method includes receiving, at a job control manager, a command specifying execution of a batch application. The method also includes receiving, at the job control manager, a commit count associated with the batch application, and initiating the batch application, with the batch application processing a group of records. The method also includes locking a first record of the group of records in response to the first record being processed by the batch application with the proviso that the batch application does not cause the first record to be unlocked unless the batch application has finished processing a last record in the group of records. Moreover, the method includes committing, in response to the batch application having completed processing of an nth record of the group of records, all records of the group of records that are locked resulting from execution of the batch application.

Preventing long running transactions from holding record locks

A method, apparatus and computer program product for performing database maintenance is described. Index information for data stored in a tree based file system running in a computer is received. The file system includes a set of data blocks arranged in logical layers and a set of pointers within the set of data blocks which reference respective data blocks in different logical layers. The received index information is parsed. A model of the file system is built based on the parsed index information. A graphical user interface having an index panel is presented. In the index panel, ones of a plurality of icons respectively represent ones of the set of data blocks and are arranged in layers representing the logical layers.

Data set visualizer for tree based file systems

In one general embodiment, a computer-implemented method includes receiving at a first system a request for data, searching one or more local buffers within the first system for the requested data, determining whether the requested data is located within an additional buffer of an additional system in communication with the first system, in response to determining that the one or more local buffers within the first system do not contain the requested data, receiving the requested data by the first system from the additional buffer of the additional system, in response to determining that the requested data is located within the additional buffer of the additional system, and retrieving the requested data from a data disk within the first system, in response to determining that the requested data is not located within the additional buffer of the additional system.

Performing caching utilizing dispersed system buffers

In one general embodiment, a computer-implemented method includes identifying a record having a lock being held being held by a first process, where the process is running within a first system of a plurality of systems, sending an identification of the lock for the record to a second system of the plurality of systems, receiving, at the first system, from the second system, information associated with a first request for the lock of the record, determining at the first system a second request for the lock of the record held by the first process, and conditionally releasing the lock being held by the first process, based on an analysis of the information associated with the first request and information associated with the second request.

Growing community of inventors

San Jose, CA, United States of America

Roity Prieto Perez