Query performance data on parallel computer system having compute nodes
AssigneeInternational Business Machines Corporation
InventorsCharles J. Archer, Michael A. Blocksome, Joseph D. Ratterman, Brian E. Smith
Life of this patent
Estimating the term…
Abstract
Embodiments of the invention provide a method for querying performance counter data on a massively parallel computing system, while minimizing the costs associated with interrupting computer processors and limited memory resources. DMA descriptors may be inserted into an injection FIFO of a remote compute node in the massively parallel computing system. Upon executing the DMA operations described by the DMA descriptors, performance counter data may be transferred from the remote compute node to a destination node.
Claims22
A method for querying performance counter data on a parallel computing system having a plurality of compute nodes, comprising:
- receiving, on a first compute node, from a second compute node, a first direct memory access (DMA) descriptor, wherein the first DMA descriptor identifies performance counter data to be sent to a third compute node;
- inserting the first DMA descriptor into an injection queue on the first compute node;
- executing an operation described by the first DMA descriptor on the first compute node to determine the performance counter data;
- and
- sending the performance counter data from the first compute node to the third compute node.
A computer-readable storage medium containing a program which, when executed, performs an operation for querying performance counter data on a parallel computing system having a plurality of compute nodes, the operation comprising:
- receiving, on a first compute node, from a second compute node, a first direct memory access (DMA) descriptor, wherein the first DMA descriptor identifies performance counter data to be sent to a third compute node;
- inserting the first DMA descriptor into an injection queue on the first compute node;
- executing an operation described by the first DMA descriptor on the first compute node to determine the performance counter data;
- and
- sending the performance counter data from the first compute node to the third compute node.
A parallel computing system, comprising:
- a plurality of compute nodes, wherein each compute node in the plurality of compute node includes at least a processor, a memory, and a DMA controller configured to perform a method for querying performance counter data from other compute nodes in the system by performing the steps of:
- receiving, on a first compute node, from a second compute node, a first direct memory access (DMA) descriptor, wherein the first DMA descriptor identifies performance counter data to be sent to a third compute node,
- inserting the first DMA descriptor into an injection queue stored in the memory on the first compute node,
- executing an operation described by the first DMA descriptor on the first compute node to determine the performance counter data, and
- sending the performance counter data from the first compute node to the third compute node.
A method of performing a fence operation on a parallel computing system having a plurality of compute nodes, comprising:
- sending a first DMA descriptor from a first compute node to a second compute node, wherein the first DMA descriptor identifies transfer data to be transferred to the first compute node;
- sending a second DMA descriptor from the first compute node to the second compute node, wherein the second DMA descriptor contains a fence identifier to be transferred to the first compute node, and wherein the second DMA descriptor is sent after the first DMA descriptor;
- receiving the transfer data on the first compute node from the second compute node, wherein the transfer data was transferred as a result of the second compute node processing the first DMA descriptor;
- and
- receiving the reflect identifier on the first compute node from the second compute node, wherein the reflect identifier was transferred as a result of the second compute node processing the second DMA descriptor.
Dependencies are inferred from the claim text.