Abstract: Direct injection of a data to be transferred in a hybrid computing environment that includes a host computer and a plurality of accelerators, the host computer and the accelerators adapted to one another for data communications by a system level message passing module. Each accelerator includes a Power Processing Element (‘PPE’) and a plurality of Synergistic Processing Elements (‘SPEs’). Direct injection includes reserving, by each SPE, a slot in a shared memory region accessible by the host computer; loading, by each SPE into local memory of the SPE, a portion of data to be transferred to the host computer; executing, by each SPE in parallel, a data processing operation on the portion of the data loaded in local memory of each SPE; and writing, by each SPE, the processed data to the SPE's reserved slot in the shared memory region accessible by the host computer.

Abstract: Systems, methods and articles of manufacture are disclosed for effecting a desired collective operation on a parallel computing system that includes multiple compute nodes. The compute nodes may pipeline multiple collective operations to effect the desired collective operation. To select protocols suitable for the multiple collective operations, the compute nodes may also perform additional collective operations. The compute nodes may pipeline the multiple collective operations and/or the additional collective operations to effect the desired collective operation more efficiently.

Abstract: Data communications in a parallel active messaging interface (‘PAMI’) of a parallel computer, the parallel computer including a plurality of compute nodes that execute a parallel application, the PAMI composed of data communications endpoints, each endpoint including a specification of data communications parameters for a thread of execution on a compute node, including specifications of a client, a context, and a task, the compute nodes and the endpoints coupled for data communications through the PAMI and through data communications resources, including receiving in an origin endpoint of the PAMI a data communications instruction, the instruction characterized by an instruction type, the instruction specifying a transmission of transfer data from the origin endpoint to a target endpoint and transmitting, in accordance with the instruction type, the transfer data from the origin endpoint to the target endpoint.

Abstract: Processing data communications events in a parallel active messaging interface (‘PAMI’) of a parallel computer that includes compute nodes that execute a parallel application, with the PAMI including data communications endpoints, and the endpoints are coupled for data communications through the PAMI and through other data communications resources, including determining by an advance function that there are no actionable data communications events pending for its context, placing by the advance function its thread of execution into a wait state, waiting for a subsequent data communications event for the context; responsive to occurrence of a subsequent data communications event for the context, awakening by the thread from the wait state; and processing by the advance function the subsequent data communications event now pending for the context.

Abstract: Performing a scatterv operation on a hierarchical tree network optimized for collective operations including receiving, by the scatterv module installed on the node, from a nearest neighbor parent above the node a chunk of data having at least a portion of data for the node; maintaining, by the scatterv module installed on the node, the portion of the data for the node; determining, by the scatterv module installed on the node, whether any portions of the data are for a particular nearest neighbor child below the node or one or more other nodes below the particular nearest neighbor child; and sending, by the scatterv module installed on the node, those portions of data to the nearest neighbor child if any portions of the data are for a particular nearest neighbor child below the node or one or more other nodes below the particular nearest neighbor child.

Abstract: Systems, methods and articles of manufacture are disclosed for performing a vector collective operation on a parallel computing system that includes multiple compute nodes and a network connecting the compute nodes that includes an ALU. A collective operation may be performed to determine displacements for the vector collective operation. Descriptors for the vector collective operation may be generated based on the displacements. The vector collective operation may then be performed using the descriptors.

Abstract: In an embodiment, a send thread receives an identifier that identifies a destination node and a pointer to data. The send thread creates a first send request in response to the receipt of the identifier and the data pointer. The send thread selects a selected channel from among a plurality of channels. The selected channel comprises a selected hand-off queue and an identification of a selected message unit. Each of the channels identifies a different message unit. The selected hand-off queue is randomly accessible. If the selected hand-off queue contains an available entry, the send thread adds the first send request to the selected hand-off queue. If the selected hand-off queue does not contain an available entry, the send thread removes a second send request from the selected hand-off queue and sends the second send request to the selected message unit.

Abstract: Methods, apparatus, and products are disclosed for profiling an application for power consumption during execution on a compute node that include: receiving an application for execution on a compute node; identifying a hardware power consumption profile for the compute node, the hardware power consumption profile specifying power consumption for compute node hardware during performance of various processing operations; determining a power consumption profile for the application in dependence upon the application and the hardware power consumption profile for the compute node; and reporting the power consumption profile for the application.

Abstract: Methods, apparatus, and products are disclosed for reducing power consumption during execution of an application on a plurality of compute nodes that include: powering up, during compute node initialization, only a portion of computer memory of the compute node, including configuring an operating system for the compute node in the powered up portion of computer memory; receiving, by the operating system, an instruction to load an application for execution; allocating, by the operating system, additional portions of computer memory to the application for use during execution; powering up the additional portions of computer memory allocated for use by the application during execution; and loading, by the operating system, the application into the powered up additional portions of computer memory.

Abstract: A system and method for enhancing barrier collective synchronization on a computer system comprises a computer system including a data storage device. The computer system includes a program stored in the data storage device and steps of the program being executed by a processor. The system includes providing a plurality of communicators for storing state information for a bather algorithm. Each communicator designates a master core in a multi-processor environment of the computer system. The system allocates or designates one counter for each of a plurality of threads. The system configures a table with a number of entries equal to the maximum number of threads. The system sets a table entry with an ID associated with a communicator when a process thread initiates a collective. The system determines an allocated or designated counter by searching entries in the table.

Abstract: Internode data communications in a parallel computer that includes compute nodes that each include main memory and a messaging unit, the messaging unit including computer memory and coupling compute nodes for data communications, in which, for each compute node at compute node boot time: a messaging unit allocates, in the messaging unit's computer memory, a predefined number of message buffers, each message buffer associated with a process to be initialized on the compute node; receives, prior to initialization of a particular process on the compute node, a data communications message intended for the particular process; and stores the data communications message in the message buffer associated with the particular process. Upon initialization of the particular process, the process establishes a messaging buffer in main memory of the compute node and copies the data communications message from the message buffer of the messaging unit into the message buffer of main memory.

Abstract: Embodiments of the invention provide a method of calculating performance counter data for a computer simulator, while minimizing the performance costs associated with cycle-accurate simulation. A callback may be associated with the instructions of a user program and, when the instructions are executed, the associated callbacks may be executed as well. Upon execution, the callbacks may calculate performance counter data related to the associated instruction.

Abstract: Intranode data communications in a parallel computer that includes compute nodes configured to execute processes, where the data communications include: allocating, upon initialization of a first process of a compute node, a region of shared memory; establishing, by the first process, a predefined number of message buffers, each message buffer associated with a process to be initialized on the compute node; sending, to a second process on the same compute node, a data communications message without determining whether the second process has been initialized, including storing the data communications message in the message buffer of the second process; and upon initialization of the second process: retrieving, by the second process, a pointer to the second process's message buffer; and retrieving, by the second process from the second process's message buffer in dependence upon the pointer, the data communications message sent by the first process.

Abstract: An apparatus, program product and method for detecting nodal faults may simultaneously cause designated nodes of a cell to communicate with all nodes adjacent to each of the designated nodes. Furthermore, all nodes along the axes of the designated nodes are made to communicate with their adjacent nodes, and the communications are analyzed to determine if a node or connection is faulty.

Abstract: Methods, apparatus, and products are disclosed for generating an executable version of an application using a distributed compiler operating on a plurality of compute nodes that include: receiving, by each compute node, a portion of source code for an application; compiling, in parallel by each compute node, the portion of the source code received by that compute node into a portion of object code for the application; performing, in parallel by each compute node, inter-procedural analysis on the portion of the object code of the application for that compute node, including sharing results of the inter-procedural analysis among the compute nodes; optimizing, in parallel by each compute node, the portion of the object code of the application for that compute node using the shared results of the inter-procedural analysis; and generating the executable version of the application in dependence upon the optimized portions of the object code of the application.

Abstract: Performing a deterministic reduction operation in a parallel computer that includes compute nodes, each of which includes computer processors and a CAU (Collectives Acceleration Unit) that couples computer processors to one another for data communications, including organizing processors and a CAU into a branched tree topology in which the CAU is a root and the processors are children; receiving, from each of the processors in any order, dummy contribution data, where each processor is restricted from sending any other data to the root CAU prior to receiving an acknowledgement of receipt from the root CAU; sending, by the root CAU to the processors in the branched tree topology, in a predefined order, acknowledgements of receipt of the dummy contribution data; receiving, by the root CAU from the processors in the predefined order, the processors' contribution data to the reduction operation; and reducing, by the root CAU, the processors' contribution data.

Abstract: Methods, apparatus, and products are disclosed for thread selection during context switching on a plurality of compute nodes that includes: executing, by a compute node, an application using a plurality of threads of execution, including executing one or more of the threads of execution; selecting, by the compute node from a plurality of available threads of execution for the application, a next thread of execution in dependence upon power characteristics for each of the available threads; determining, by the compute node, whether criteria for a thread context switch are satisfied; and performing, by the compute node, the thread context switch if the criteria for a thread context switch are satisfied, including executing the next thread of execution.

Abstract: Compute nodes of a parallel computer organized for collective operations via a network, each compute node having a receive buffer and establishing a topology for the network; selecting a schedule for a broadcast operation; depositing, by a root node of the topology, broadcast data in a target node's receive buffer, including performing a DMA operation with a well-known memory location for the target node's receive buffer; depositing, by the root node in a memory region designated for storing broadcast data length, a length of the broadcast data, including performing a DMA operation with a well-known memory location of the broadcast data length memory region; and triggering, by the root node, the target node to perform a next DMA operation, including depositing, in a memory region designated for receiving injection instructions for the target node, an instruction to inject the broadcast data into the receive buffer of a subsequent target node.

Abstract: Paging memory from random access memory (‘RAM’) to backing storage in a parallel computer that includes a plurality of compute nodes, including: executing a data processing application on a virtual machine operating system in a virtual machine on a first compute node; providing, by a second compute node, backing storage for the contents of RAM on the first compute node; and swapping, by the virtual machine operating system in the virtual machine on the first compute node, a page of memory from RAM on the first compute node to the backing storage on the second compute node.

Abstract: Internode data communications in a parallel computer that includes compute nodes that each include main memory and a messaging unit, the messaging unit including computer memory and coupling compute nodes for data communications, in which, for each compute node at compute node boot time: a messaging unit allocates, in the messaging unit's computer memory, a predefined number of message buffers, each message buffer associated with a process to be initialized on the compute node; receives, prior to initialization of a particular process on the compute node, a data communications message intended for the particular process; and stores the data communications message in the message buffer associated with the particular process. Upon initialization of the particular process, the process establishes a messaging buffer in main memory of the compute node and copies the data communications message from the message buffer of the messaging unit into the message buffer of main memory.