International Business Machines Corporation

Michael Robert Scott

David C Reed

Max D Smith

Philip R Chauvet

Franklin Emmert McCune

Gregg Leonard Liguori

Gregory E McBride

Sosuke Matsui

Richard Albert Welp

Ryan J Wisniewski

Takeshi Nohta

Aderson J Pacini

Derek Logan Erdmann

Ronald David Boenig, Ii

Franklin E Mccune

Ryan Arthur Bouchard

Samuel Ryan Smith

Parker Mathewson

A computer-implemented method (CIM), according to one embodiment, includes performing a reorganization analysis for mitigating latency during reorganization of a data set. The reorganization analysis includes using regression analysis to predict a time duration that reorganization of the data set will take based on a total size of the data set, relating the predicted time duration to a previous data set reorganization time duration, and performing linear extrapolation to estimate a time at which the total size of the data set will exceed a predetermined percentage of a maximum size of the data set. The method further includes reorganizing the data set before the estimated time, where the data set is reorganized at a time determined based on the reorganization analysis or derivative thereof.

Reorganization of a data set

Provided are a computer program product, system, and method to determine whether a holding of a lock to the resource by a first task satisfies a condition and whether the lock is swappable. The lock is released from the first task and granted to a second task waiting in a queue for the lock in response to determining that the holding of the lock satisfies the condition and that the lock is swappable. The first task is indicated in the queue waiting for the lock in response to granting the lock to the second task.

Determining when to release a lock from a first task holding the lock to grant to a second task waiting for the lock

Provided are a computer program product, system, and method for selecting a plurality of processing systems to sort a data set. A determination is made of processing systems to perform a sort of records in a data set. The data set is divided into data subsets, each data subset comprising a portion of the records of the data set. Each of the determined processing systems of is notified of a data subset to sort. In response to receiving notification from the determined processing systems that the records in the data subsets have been sorted, the sorted data subsets are merged into a sorted data set comprising the records of the data set sorted.

Selecting a plurality of processing systems to sort a data set

Provided are a computer program product, system, and method for determining when to release a lock from a first task holding the lock to grant to a second task waiting for the lock. A determination is made as to whether a holding of a lock to the resource by a first task satisfies a condition and whether the lock is swappable. The lock is released from the first task and granted to a second task waiting in a queue for the lock in response to determining that the holding of the lock satisfies the condition and that the lock is swappable. The first task is indicated in the queue waiting for the lock in response to granting the lock to the second task.

A method for allocating a resource to multiple requesters is disclosed. In one embodiment, such a method includes maintaining, for a resource, a regular queue and an express queue. The method receives requests to control the resource and determines, for each request, an anticipated amount of time that the request needs to control the resource. In the event the anticipated amount of time for a request is greater than a selected threshold, the method allocates the request to the regular queue. In the event the anticipated amount of time for a request is less than the selected threshold, the method allocates the request to the express queue. The method provides priority to requests allocated to the express queue over requests allocated to the regular queue. A corresponding system and computer program product are also disclosed.

Short duration serialization express queue

A method includes receiving a request to store a record in a base data set and determining whether the base data set can accommodate the record. In the event the base data set cannot accommodate the record, the method automatically creates an overflow data set to accommodate the record, stores the record in the overflow data set, and updates an index of the base data set to include a key associated with the record. The key points to the overflow data set as containing the record. In certain embodiments, the base data set and overflow data set are both keyed VSAM data sets. A corresponding system and computer program product are also disclosed.

Full data set avoidance

A method for managing volumes in a scratch pool of a virtual tape system is disclosed. In one embodiment, such a method provides a scratch pool containing volumes for use in a virtual tape system. The method further enables a user to predefine an external pool of volumes residing outside of the scratch pool. This external pool may be hidden to a host system accessing the virtual tape system. The method monitors current and/or past usage of the volumes in the scratch pool and, based on the usage, predicts a future need for volumes in the scratch pool. The method automatically moves volumes between the external pool and the scratch pool in accordance with the future need. A corresponding system and computer program product are also disclosed.

Dynamic scratch pool management on a virtual tape system

Provided are a computer program product, system, and method for data unit classification in accordance with one embodiment of the present description, in which in response to a data processing command, a storage controller classifies data units of a storage unit as either allocated to a data set or as unallocated to any data set. If allocated to a data set, the storage controller can further classify data set-allocated data units as either containing client data or metadata or as empty. In accordance with one aspect of the present description, the storage controller may bypass data processing of the data units which have not been allocated to any data set or otherwise do not contain client data or metadata. Other aspects of data unit classification in accordance with the present description are described.

Metadata-based bypassing in a controller

Various embodiments for retaining diagnostic information for data in a computing storage environment. In one such embodiment, a diagnostic component, apart from a volume table of contents (VTOC), associated with an integrated catalog facility (ICF) catalog and with a base data set from data sets via a catalog association record, is initialized. The diagnostic component is configured to retain base data set-specific diagnostic information retrievable by the computing storage environment to assist in error diagnosis. The base data set-specific diagnostic information is stored pursuant to at least one detected event associated with the base data set.

Diagnostic data set component

Embodiments for managing data in a virtual tape storage environment, by a processor device, are provided. A management system on a host is used to define volume expiration attributes for virtual tape volumes and the volume expiration attributes are forwarded to a virtual tape server. Return-to-scratch processing is performed on the virtual tape volumes directly on the virtual tape server by using the volume expiration attributes to compile candidate volumes without querying the host, where the candidate volumes are expired and converted to scratch by the virtual tape server.

Self-expiring data in a virtual tape server

Provided are a computer program product, system, and method for modifying memory space allocation for inactive tasks. Information is maintained on computational resources consumed by tasks running in the computer system allocated memory space in the memory. The information on the computational resources consumed by the tasks is used to determine inactive tasks of the tasks. The allocation of the memory space allocated to at least one of the determined inactive tasks is modified.

Growing community of inventors

Ocean View, HI, United States of America

Michael Robert Scott