International Business Machines Corporation

Alfred F Foster

Robert Miller, Jr

Charles E Mari

Elpida Tzortzatos

Harris M Morgenstern

Thomas F Rankin

Scott B Tuttle

Peter Grimm Sutton

Matthew J Mauriello

Mariama Ndoye

Steven M Partlow

Walter W Otto

Michael Gary Spiegel

Duane C Hughes

David Horn

David Hom

Christopher G Brooker

Christopher Lee Wood

Chun Kwan K Yee

Chun-Kwan K Yee

A transient input/output in progress state is established during processing of a testcase by a test infrastructure in a computing environment. The method includes obtaining the testcase for an object having one or more pages, and processing the testcase by the test infrastructure. Processing the testcase by the test infrastructure includes, for a page of the object, generating a delay in the processing of the testcase for the page of the object. The delay opens a transient input/output in progress state during which one or more test operations reference the page of the object.

Controlled input/output in progress state during testcase processing

A transient input/output in progress state is established during processing of an input/output testcase by a test infrastructure in a computing environment. The method includes obtaining the input/output testcase for an object having one or more pages, and processing the input/output testcase by the test infrastructure. Processing the input/output testcase by the test infrastructure includes, for a page of the object, generating a delay in the processing of the input/output testcase for the page of the object. The delay opens a transient input/output in progress state during which one or more concurrent test operations are to reference the page of the object.

A method, system, and computer program product to verify serialization of storage frames within an address space via multi-threaded programs is described. The method includes dynamically scaling a number of units of work based on a number of available processors, each of the units of work configured to execute actions, and dynamically scaling an amount and page size of virtual storage accessed by each of the units of work based on a total available memory. The method also includes obtaining, at each of the units of work, different types of storage pages and accessing storage pages corresponding with the respective different types of virtual storage pages associated with the different frame sizes and attributes and performing a respective action, and verifying, for each of the units of work performing the respective action, a state and data content of the storage pages.

Verification of serialization of storage frames within an address space via multi-threaded programs

A method, system, and computer program product to verify management of real storage via multi-threaded thrashers in multiple address spaces are described. The method includes dynamically scaling a number of units of work and a number of address spaces based on a number of available processors and dynamically scaling an amount and page size of storage pages representing virtual storage accessed by each of the number of units of work based on a total available memory. The method also includes obtaining, at each of the units of work, different types of storage frame sizes and attributes, accessing the storage pages corresponding with the respective different types of storage frame sizes and attributes and performing a respective function, and verifying, for each of the units of work performing the respective function, a location of the storage pages and content of the storage pages based on the respective function.

Verification of management of real storage via multi-threaded thrashers in multiple address spaces

Embodiments of the disclosure include a method for reserving large pages in a large frame area (LFAREA) of a main memory. The method includes pre-scanning a plurality of storage increments and counting a number of available large pages that are online and issuing a message that indicates the number of available large pages. The method also includes receiving and parsing an LFAREA request including a target number of large pages to be reserved. The method further includes calculating an optimal number of large pages to be reserved, based upon the target number of available pages and a system limit. The method includes determining if the LFAREA request is valid and if the LFAREA request can be satisfied and scanning the plurality of the storage increments and reserving the optimal number of pages in the LFAREA.

Reserving fixed page areas in real storage increments

Embodiments of the disclosure include a method for optimizing large page processing. The method includes receiving an indication that a real memory includes a first page. The first page includes a plurality of smaller pages. The method also includes determining a page frame table entry associated with a first smaller page of the first page and storing data associated with the first page in the page frame table entry associated with the first smaller page. The page frame table entry associated with the first smaller page of the first page is a data repository for the plurality of smaller pages of the first page.

Optimizing large page processing

Embodiments relate to methods, systems and computer program products for defragmenting storage class memory by comparing a utilization rate of the storage class memory to a threshold value. If the utilization rate of the storage class memory is greater than the threshold value, the potentially wasted storage space is then compared to the combined storage capacity of the unclaimed extents of the storage class memory. If the potentially wasted storage space is greater than the combined storage capacity of the unclaimed extents of the storage class memory, a determination is made whether a defragmentation was recently performed. Based on determining that the defragmentation was not recently performed, or that it was recently performed and was productive, performing a defragmentation of the storage class memory.

Defragmentation of storage class memory managed in extents

Dedicated large page memory pools are provided to, at least in part, facilitate access to large pages. The large page memory is managed by: establishing multiple large page memory pools, each large page memory pool of the multiple large page memory pools including a number of large pages; and dedicating each large page memory pool of the multiple large page memory pools to a respective processor of multiple processors of the computing environment, wherein processors of the multiple processors can concurrently access pages from the respective large page memory pools of the multiple large page memory pools.

Dedicated large page memory pools

Allocation of pages of memory is managed in computing environments that include multiple sized memory pools. Responsive to a request for a page of memory, one or more memory pools are searched for an available frame of memory to service the request. The search uses a predefined order of search, which includes multiple types of memory pools in a specific order based on the requested size of the page of memory.

Managing allocation of memory pages

Large page memory pools are managed. Thresholds are used to determine if the number of pages in a large page memory pool is to be adjusted. If the number of pages is to be increased, a particular technique is provided for adding additional pages to the pool. Further, if there are too many pages in the pool, one or more pages may be removed.

Managing large page memory pools

An auxiliary storage medium is selected for writing data of a set of one or more pages being paged-out from real memory. The auxiliary storage medium is selected from among a plurality of auxiliary storage media, including differing types of storage media, based on characteristics of the plurality of storage media and/or the attributes of the data being written to the auxiliary storage media.

Selecting an auxiliary storage medium for writing data of real storage pages

Auxiliary storage is segmented into different types of extents (i.e., ranges of storage), including extents with 4K blocks of storage, extents with 1M blocks of storage, empty extents, and mixed extents that include blocks of storage of various sizes (e.g., 4K blocks and 1M blocks). The auxiliary storage, and in particular, the extents and blocks of storage therein, are managed to reduce storage fragmentation and optimize system performance.

Growing community of inventors

Wappingers Falls, NY, United States of America

Alfred F Foster