Abstract: Configuring compute nodes in a parallel computer using remote direct memory access (‘RDMA’), the parallel computer comprising a plurality of compute nodes coupled for data communications via one or more data communications networks, including: initiating, by a source compute node of the parallel computer, an RDMA broadcast operation to broadcast binary configuration information to one or more target compute nodes in the parallel computer; preparing, by each target compute node, the target compute node for receipt of the binary configuration information from the source compute node; transmitting, by each target compute node, a ready message to the target compute node, the ready message indicating that the target compute node is ready to receive the binary configuration information from the source compute node; and performing, by the source compute node, an RDMA broadcast operation to write the binary configuration information into memory of each target compute node.

Abstract: Configuring compute nodes in a parallel computer using remote direct memory access (‘RDMA’), the parallel computer comprising a plurality of compute nodes coupled for data communications via one or more data communications networks, including: initiating, by a source compute node of the parallel computer, an RDMA broadcast operation to broadcast binary configuration information to one or more target compute nodes in the parallel computer; preparing, by each target compute node, the target compute node for receipt of the binary configuration information from the source compute node; transmitting, by each target compute node, a ready message to the target compute node, the ready message indicating that the target compute node is ready to receive the binary configuration information from the source compute node; and performing, by the source compute node, an RDMA broadcast operation to write the binary configuration information into memory of each target compute node.

Abstract: Configuring compute nodes in a parallel computer using remote direct memory access (‘RDMA’), the parallel computer comprising a plurality of compute nodes coupled for data communications via one or more data communications networks, including: initiating, by a source compute node of the parallel computer, an RDMA broadcast operation to broadcast binary configuration information to one or more target compute nodes in the parallel computer; preparing, by each target compute node, the target compute node for receipt of the binary configuration information from the source compute node; transmitting, by each target compute node, a ready message to the target compute node, the ready message indicating that the target compute node is ready to receive the binary configuration information from the source compute node; and performing, by the source compute node, an RDMA broadcast operation to write the binary configuration information into memory of each target compute node.

Abstract: Configuring compute nodes in a parallel computer using remote direct memory access (‘RDMA’), the parallel computer comprising a plurality of compute nodes coupled for data communications via one or more data communications networks, including: initiating, by a source compute node of the parallel computer, an RDMA broadcast operation to broadcast binary configuration information to one or more target compute nodes in the parallel computer; preparing, by each target compute node, the target compute node for receipt of the binary configuration information from the source compute node; transmitting, by each target compute node, a ready message to the target compute node, the ready message indicating that the target compute node is ready to receive the binary configuration information from the source compute node; and performing, by the source compute node, an RDMA broadcast operation to write the binary configuration information into memory of each target compute node.

Abstract: Configuring compute nodes in a parallel computer using remote direct memory access (‘RDMA’), the parallel computer comprising a plurality of compute nodes coupled for data communications via one or more data communications networks, including: initiating, by a source compute node of the parallel computer, an RDMA broadcast operation to broadcast binary configuration information to one or more target compute nodes in the parallel computer; preparing, by each target compute node, the target compute node for receipt of the binary configuration information from the source compute node; transmitting, by each target compute node, a ready message to the target compute node, the ready message indicating that the target compute node is ready to receive the binary configuration information from the source compute node; and performing, by the source compute node, an RDMA broadcast operation to write the binary configuration information into memory of each target compute node.

Abstract: Configuring compute nodes in a parallel computer using remote direct memory access (‘RDMA’), the parallel computer comprising a plurality of compute nodes coupled for data communications via one or more data communications networks, including: initiating, by a source compute node of the parallel computer, an RDMA broadcast operation to broadcast binary configuration information to one or more target compute nodes in the parallel computer; preparing, by each target compute node, the target compute node for receipt of the binary configuration information from the source compute node; transmitting, by each target compute node, a ready message to the target compute node, the ready message indicating that the target compute node is ready to receive the binary configuration information from the source compute node; and performing, by the source compute node, an RDMA broadcast operation to write the binary configuration information into memory of each target compute node.

Abstract: Techniques for transferring files between machines include creating a zero-length target file on non-volatile storage, truncating the file to a desired size, and allocating storage on the non-volatile storage for each block of the target file. The technique also includes determining a logical block address (LBA) for each location in the target file. The technique further includes sending a request to an input/output (I/O) node to transfer a source file to the non-volatile storage, where the request includes a mapping between the LBAs and file offsets. The technique includes opening the source file and a block device at the I/O node. The technique further includes reading each block from the source file and writing each block to the target file on the non-volatile storage utilizing the block device, and then closing the source file and the block device.

Abstract: Techniques are disclosed for loading programs efficiently in a parallel computing system. In one embodiment, nodes of the parallel computing system receive a load description file which indicates, for each program of a multiple program multiple data (MPMD) job, nodes which are to load the program. The nodes determine, using collective operations, a total number of programs to load and a number of programs to load in parallel. The nodes further generate a class route for each program to be loaded in parallel, where the class route generated for a particular program includes only those nodes on which the program needs to be loaded. For each class route, a node is selected using a collective operation to be a load leader which accesses a file system to load the program associated with a class route and broadcasts the program via the class route to other nodes which require the program.

Abstract: Method and apparatus for detecting a hung up and/or slow-running syscall without affecting the performance of the syscall. Before a syscall is started, a time stamp can be created at a memory address that is distinct from memory addresses to be used by the syscall. While a syscall thread handles the syscall operation, a separate monitor thread monitors the time stamp to track the length of time the syscall operation has been running. If the syscall thread operation exceeds a threshold time limit, then a flag can be sent to a network administrator to indicate that the syscall may be hung up and/or slow running.

Abstract: Method and apparatus for detecting a hung up and/or slow-running syscall without affecting the performance of the syscall. Before a syscall is started, a time stamp can be created at a memory address that is distinct from memory addresses to be used by the syscall. While a syscall thread handles the syscall operation, a separate monitor thread monitors the time stamp to track the length of time the syscall operation has been running. If the syscall thread operation exceeds a threshold time limit, then a flag can be sent to a network administrator to indicate that the syscall may be hung up and/or slow running.

Abstract: Techniques for transferring files between machines include creating a zero-length target file on non-volatile storage, truncating the file to a desired size, and allocating storage on the non-volatile storage for each block of the target file. The technique also includes determining a logical block address (LBA) for each location in the target file. The technique further includes sending a request to an input/output (I/O) node to transfer a source file to the non-volatile storage, where the request includes a mapping between the LBAs and file offsets. The technique includes opening the source file and a block device at the I/O node. The technique further includes reading each block from the source file and writing each block to the target file on the non-volatile storage utilizing the block device, and then closing the source file and the block device.

Abstract: Techniques are disclosed for loading programs efficiently in a parallel computing system. In one embodiment, nodes of the parallel computing system receive a load description file which indicates, for each program of a multiple program multiple data (MPMD) job, nodes which are to load the program. The nodes determine, using collective operations, a total number of programs to load and a number of programs to load in parallel. The nodes further generate a class route for each program to be loaded in parallel, where the class route generated for a particular program includes only those nodes on which the program needs to be loaded. For each class route, a node is selected using a collective operation to be a load leader which accesses a file system to load the program associated with a class route and broadcasts the program via the class route to other nodes which require the program.

Abstract: Techniques are disclosed for loading programs efficiently in a parallel computing system. In one embodiment, nodes of the parallel computing system receive a load description file which indicates, for each program of a multiple program multiple data (MPMD) job, nodes which are to load the program. The nodes determine, using collective operations, a total number of programs to load and a number of programs to load in parallel. The nodes further generate a class route for each program to be loaded in parallel, where the class route generated for a particular program includes only those nodes on which the program needs to be loaded. For each class route, a node is selected using a collective operation to be a load leader which accesses a file system to load the program associated with a class route and broadcasts the program via the class route to other nodes which require the program.

Abstract: Collectively loading an application in a parallel computer, the parallel computer comprising a plurality of compute nodes, including: identifying, by a parallel computer control system, a subset of compute nodes in the parallel computer to execute a job; selecting, by the parallel computer control system, one of the subset of compute nodes in the parallel computer as a job leader compute node; retrieving, by the job leader compute node from computer memory, an application for executing the job; and broadcasting, by the job leader to the subset of compute nodes in the parallel computer, the application for executing the job.

Abstract: Aggregating job exit statuses of a plurality of compute nodes executing a parallel application, including: identifying a subset of compute nodes in the parallel computer to execute the parallel application; selecting one compute node in the subset of compute nodes in the parallel computer as a job leader compute node; initiating execution of the parallel application on the subset of compute nodes; receiving an exit status from each compute node in the subset of compute nodes, where the exit status for each compute node includes information describing execution of some portion of the parallel application by the compute node; aggregating each exit status from each compute node in the subset of compute nodes; and sending an aggregated exit status for the subset of compute nodes in the parallel computer.

Abstract: Method and apparatus for detecting a hung up and/or slow-running syscall without affecting the performance of the syscall. Before a syscall is started, a time stamp can be created at a memory address that is distinct from memory addresses to be used by the syscall. While a syscall thread handles the syscall operation, a separate monitor thread monitors the time stamp to track the length of time the syscall operation has been running. If the syscall thread operation exceeds a threshold time limit, then a flag can be sent to a network administrator to indicate that the syscall may be hung up and/or slow running.

Abstract: Method and apparatus for detecting a hung up and/or slow-running syscall without affecting the performance of the syscall. Before a syscall is started, a time stamp can be created at a memory address that is distinct from memory addresses to be used by the syscall. While a syscall thread handles the syscall operation, a separate monitor thread monitors the time stamp to track the length of time the syscall operation has been running. If the syscall thread operation exceeds a threshold time limit, then a flag can be sent to a network administrator to indicate that the syscall may be hung up and/or slow running.

Abstract: A hybrid computing environment in which the host computer allocates, in the shadow memory area of the host computer, a memory region for a packet to be written to the shared memory of an accelerator; writes packet data to the accelerator's shared memory in a memory region corresponding to the allocated memory region; inserts, in a next available element of the accelerator's descriptor array, a descriptor identifying the written packet data; increments the copy of the head pointer of the accelerator's descriptor array maintained on the host computer; and updates a copy of the head pointer of the accelerator's descriptor array maintained on the accelerator with the incremented copy.

Abstract: Computer program product and system to limit core file generation in a massively parallel computing system comprising a plurality of compute nodes each executing at least one task, of a plurality of tasks, by: upon determining that a first task executing on a first compute node has failed, performing an atomic load and increment operation on a core file count; generating a first core file upon determining that the core file count is below a predefined threshold; and not generating the first core file upon determining that the core file count is not below the predefined threshold.

Abstract: Computer program product and system to limit core file generation in a massively parallel computing system comprising a plurality of compute nodes each executing at least one task, of a plurality of tasks, by: upon determining that a first task executing on a first compute node has failed, performing an atomic load and increment operation on a core file count; generating a first core file upon determining that the core file count is below a predefined threshold; and not generating the first core file upon determining that the core file count is not below the predefined threshold.