International Business Machines Corporation

Michael E Lyons

Bruce Gerard Mealey

Bret Ronald Olszewski

Omar Cardona

James Anthony Pafumi

Vaijayanthimala K Anand

Larry Bert Brenner

Basu Vaidyanathan

Ronen Grosman

Michael Anthony Perez

Andre Laurent Albot

Robert G Kovacs

Amit H Lamba

James Bernard Moody

Jimmy Ray Hill

Jonathan Allen Wildstrom

Shibhani Rai

Nicholas Georg Lange

Sean McNeal

Wojciech T Stryjewski

Methods and systems for data governance in a dispersed storage network using bucket templates are disclosed. A method includes: creating, by a computing device, a bucket template in a dispersed storage network (DSN) based on a first request, the bucket template including at least a first bucket parameter and at least one allowed value for the first bucket parameter; receiving, by the computing device, a second request to create a bucket in the DSN, the second request including at least a first value for the first bucket parameter; and creating, by the computing device, the bucket based on the first value for the first bucket parameter in the second request and the at least one allowed value for the first bucket parameter in the bucket template.

Data governance in a dispersed storage network using bucket templates

Provided are techniques generating a data structure, wherein the data structure specifies both a specified size of a memory space to allocate within an application and a virtual address within the application to locate a data path transmission queue; including within a verb for allocating the data path transmission queue the defined data structure; in response to a call of the verb, allocate, within the application, the data path transmission queue of the specified size and at the virtual location; in response to a request to transmit control data, employ a remote direct memory access (RDMA) transmission path; and, in response to a request to transmit data, employ the data path transmission queue rather than an RDMA transmission path.

Remote direct memory access acceleration via hardware context in non-native applciations

The illustrative embodiments provide a computer implemented method, apparatus, and computer usable program code for managing a heap. The heap is partitioned into at least one sub heap based on a relationship to at least one memory class of a plurality of memory classes. A memory allocation request comprising a memory class is received from a requester. A unique heap handle based on the memory class and associated with a specific sub heap is generated. The unique heap handle is then returned to the requester.

Memory class based heap partitioning

Software components 'self-scrub' to improve software reliability, serviceability and availability (RAS). Each component designates a routine to perform a component level consistency check on major data structures and to verify the state of component. This is performed as an on-going task during the life of the component. The component registers an entry point with the system to receive notification of scrubbing parameter changes. The entry point is also called with the request to perform component-scrubbing operations. The entry point functions are responsible for executing within limitations on central processing unit (CPU) usage and memory footprint when performing scrubbing operations.

Software component self-scrubbing

A software thread is dispatched for causing the system to poll a device for determining whether a condition has occurred. Subsequently, the software thread is undispatched and, in response thereto, an interrupt is enabled on the device, so that the device is enabled to generate the interrupt in response to an occurrence of the condition, and so that the system ceases polling the device for determining whether the condition has occurred. Eventually, the software thread is redispatched and, in response thereto, the interrupt is disabled on the device, so that the system resumes polling the device for determining whether the condition has occurred.

Polling in a virtualized information handling system

Methods, systems, and products are disclosed for implementing a generic kernel extension framework that operate generally to allow one kernel extension to register a function callable by other kernel extensions. A kernel extension can be loaded even if a function it potentially could need is not available. If in fact the running kernel extension does eventually request a function that no other kernel extension has registered to provide, it is failed gracefully with a return error code. The client kernel extension can still function normally, assuming the desired service is non-critical. Example services in the non-critical category may include, for example, advanced error handling functions, statistical functions, reporting functions, and so on, including many others as will occur to those of skill in the art.

Administration of kernel extensions

A method, computer program product, and data processing system for preventing the occurrence of undetectable dangling pointers through memory reallocation are disclosed. Allocated memory regions that are no longer needed are deallocated but are not immediately freed for reallocation, being designated as 'retained.' A memory retention metric is computed as a measure of an extent of the retained memory regions in the computer's memory space. Once the memory retention metric exceeds a pre-determined threshold, some or all of the retained memory regions are freed for reuse. In this manner, improper accesses to deallocated memory regions can be detected more easily, since reuse of those regions is delayed, while at the same time excessive resource usage and heap fragmentation is avoided by only retaining deallocated regions for a limited time (determined by the threshold and retention metric chosen).

Detecting illegal reuse of memory with low resource impact

A method and system for choosing an optimal PCI adapter burst length is disclosed. The optimal burst length is automatically determined by the adapter configuration feature of AIX software using a cache-line size of a PCI bridge and the latency timer value of the target PCI adapter as inputs. The method also provides for a user to be able to override the software-calculated setting.

System for optimally tuning a burst length by setting a maximum burst length based on a latency timer value and adjusting the maximum burst length based on a cache line size

Growing community of inventors

Austin, TX, United States of America

Michael E Lyons