Abstract: In one aspect, multiple data path error collection is provided in a storage management system. In one embodiment, an error condition in a main data path between the storage controller and at least one of a host and a storage unit is detected, and in response, a sequence of error data collection operations to collect error data through a main path is initiated. In response to a failure to collect error data at a level of the sequential error data collection operations, error data is collected through an alternate data path as a function of the error data collection level at which the failure occurred. Other aspects are described.

Abstract: Provided are a computer program product, system, and method for using send buffers and receive buffers for sending messages among nodes in a network. A send buffer is provided for each of at least one receiving node comprising one of the nodes to which messages are sent. Each of the receiving nodes includes at least one receive buffer to receive messages from the send buffer at the sending node. A determination is made of a buffer entry for a send buffer and receive buffer pair that is available for use. Indication is made of the message in the determined buffer entry of the send buffer. The message is sent from the send buffer to the receiving node indicating the determined buffer entry in which the message is indicated to cause the receiving node to include the message in the indicated buffer entry in the receive buffer.

Abstract: A plurality of processing entities in which a plurality of tasks are executed are maintained. Memory access patterns are determined for each of the plurality of tasks by dividing a memory associated with the plurality of processing entities into a plurality of memory regions, and for each of the plurality of tasks, determining how many memory accesses take place in each of the memory regions, by incrementing a counter associated with each memory region in response to a memory access. Each of the plurality of tasks are allocated among the plurality of processing entities, based on the determined memory access patterns for each of the plurality of tasks.

Abstract: Provided are techniques for utilization based multi-buffer self-calibrated dynamic adjustment management. A sub-buffer is assigned to each entity of multiple entities. Under control of each of the multiple entities, a current utilization rate of the entity and each other entity is summed up; a number of data segments for the entity are determined based on the current utilization rate of the entity relative to the current utilization rate of each of the other entities; and a size of the assigned sub-buffer is adjusted based on the determination.

Abstract: A determination is made of a plurality of components whose states are to be determined to generate a statesave. At least one central processing unit that determines a state of a first component of the plurality of components faster than other central processing units is assigned to determine the state of the first component to include in the statesave, where more processing operations have to be performed to determine the state of the first component in comparison to any other component of the plurality of component. One or more of the other central processing units are assigned to other components of the plurality of components to determine states of the other components to include in the statesave.

Abstract: Provided are a computer program product, system, and method for managing processor threads of a plurality of processors. In one embodiment, a parameter of performance of the computing system is measured, and the configurations of one or more processor nodes are dynamically adjusted as a function of the measured parameter of performance. In this manner, the number of processor threads being concurrently executed by the plurality of processor nodes of the computing system may be dynamically adjusted in real time as the system operates to improve the performance of the system as it operates under various operating conditions. It is appreciated that systems employing processor thread management in accordance with the present description may provide other features in addition to or instead of those described herein, depending upon the particular application.

Abstract: Provided are techniques for utilization based multi-buffer dynamic adjustment management. A sub-buffer is assigned to each entity of multiple entities. A percentage utilization is determined for each entity. Based on the percentage utilization, for each sub-buffer assigned to each entity, one of one of allocating at least one random data segment from a free list of data segments and removing at least one data segment to change a size of the sub-buffer is performed.

Abstract: A set of like tasks to be performed is organized into a first group. A last used processing group assigned to the set of like tasks is stored. The set of like tasks is reassigned to an additional group having a minimal queue length upon a determination that the difference between the queue lengths of the additional processing group and the stored processing group is greater than a predetermined threshold.

Abstract: Provided are a computer program product, system, and method for using a plurality of sub-buffers and a free segment list to allocate segments to a plurality of threads to use for writing data. Groups of threads are assigned to sub-buffers. At least one segment of data is indicated in each of the sub-buffers. The threads assigned to one of the sub-buffers write to the at least one segment indicated in the sub-buffer to which the threads are assigned. A free segment list indicates segments which are not indicated in one of the sub-buffers. In response to one of the segments in a containing sub-buffer comprising one of the sub-buffers becoming a full segment having less than a threshold amount of free space, the indication of the full segment in the containing sub-buffer is replaced with one of the segments indicated in the free segment list.

Abstract: Provided are a computer program product, system, and method for managing a lock to a resource shared among a plurality of processors. Slots in a memory implement the lock on the shared resource. The slots correspond to counter values that are consecutively numbered and indicate one of busy and free. A requesting processor fetches a counter value comprising a fetched counter value. A determination is made as to whether the slot corresponding to the fetched counter value indicates free. A processor identifier of the requesting processor is inserted into the slot corresponding to the fetched counter value in response to determining that the slot corresponding to the fetched counter value indicates not free. The requesting processor accesses the shared resource in response to determining that the slot corresponding to the fetched counter value indicates free.

Abstract: A rack-power control module (RPC) module is used for allowing a local storage partition, located on a local server, for controlling a destination storage partition, located on a destination server, by piggybacking commands on power alerts issued by the RPC module in a clustered storage system. The commands are sent from the local storage partition to the RPC module, where the commands are RPC commands and include a destination server identification (ID) and payload data that includes a hypervisor command containing a command type, a subtype, and a destination ID. The RPC module then parses the commands received from the local storage partition.

Abstract: In a distributed server storage environment, a set of like tasks to be performed is organized into a first group, and a last used processing group associated with the like tasks is stored. Upon a subsequent dispatch, the last used processing group is compared to other processing groups and the tasks are assigned to a processing group based upon a predetermined threshold.

Abstract: A set of like tasks to be performed is organized into a first group, and a last used processing group associated with the like tasks is stored. Upon a subsequent dispatch, the last used processing group is compared to other processing groups and the tasks are assigned to a processing group based upon a predetermined threshold.

Abstract: Provided are a computer program product, system, and method for managing a lock to a resource shared among a plurality of processors. Slots in a memory implement the lock on the shared resource. The slots correspond to counter values that are consecutively numbered and indicate one of busy and free. A requesting processor fetches a counter value comprising a fetched counter value. A determination is made as to whether the slot corresponding to the fetched counter value indicates free. A processor identifier of the requesting processor is inserted into the slot corresponding to the fetched counter value in response to determining that the slot corresponding to the fetched counter value indicates not free. The requesting processor accesses the shared resource in response to determining that the slot corresponding to the fetched counter value indicates free.

Abstract: In a computing storage environment having multiple processor devices, lists of Task Control Blocks (TCBs) are maintained in a processor-specific manner, such that each of the multiple processor devices is assigned a local TCB list.

Abstract: A rack-power control module (RPC) module is used for allowing a local storage partition, located on a local server, for controlling a destination storage partition, located on a destination server, by piggybacking commands on power alerts issued by the RPC module in a clustered storage system.

Abstract: Mechanisms for improving computing system performance by a processor device. System resources are organized into a plurality of groups. Each of the plurality of groups is assigned one of a plurality of predetermined task pools. Each of the predetermined task pools has a plurality of tasks. Each of the plurality of groups corresponds to at least one physical boundary of the system resources such that a speed of an execution of those of the plurality of tasks for a particular one of the plurality of predetermined task pools is optimized by a placement of an association with the at least one physical boundary and the plurality of groups.

Abstract: In a distributed server storage environment, a set of like tasks to be performed is organized into a first group, and a last used processing group associated with the like tasks is stored. Upon a subsequent dispatch, the last used processing group is compared to other processing groups and the tasks are assigned to a processing group based upon a predetermined threshold.

Abstract: A set of like tasks to be performed is organized into a first group, and a last used processing group associated with the like tasks is stored. Upon a subsequent dispatch, the last used processing group is compared to other processing groups and the tasks are assigned to a processing group based upon a predetermined threshold.

Abstract: A set of like tasks to be performed is organized into a first group. Upon a determined imbalance between dispatch queue depths greater than a predetermined threshold, the set of like tasks is reassigned to an additional group.