Dynamic distribution of nodes on a multi-node computer system
I/O nodes are dynamically distributed on a multi-node computing system. An I/O configuration mechanism located in the service node of a multi-node computer system controls the distribution of the I/O nodes. The I/O configuration mechanism uses job information located in a job record to initially configure the I/O node distribution. The I/O configuration mechanism further monitors the I/O performance of the executing job to then dynamically adjusts the I/O node distribution based on the I/O performance of the executing job.
1. A multi-node computer system comprising:
a plurality of compute nodes that each comprise a processor and memory;
a plurality of input/output (I/O) nodes comprising a processor connected to the plurality of compute nodes, wherein the I/O nodes are connected to the compute nodes with a virtual network operating on a physical network and the I/O nodes communicate with a service node to provide I/O communication to network resources;
wherein a block of compute nodes from the plurality of compute nodes is configured to execute a job, and the block of compute nodes has a number of associated I/O nodes that form a ratio of I/O nodes to compute nodes; and
wherein the system is configured to dynamically adjust the ratio of I/O nodes to compute nodes based on I/O characteristics of the job executing on the block of compute nodes by dynamically configuring the virtual network, wherein the I/O characteristics of the job are determined by real-time monitoring of the job's I/O characteristics.
2. The multi-node computer system of claim 1 wherein the multi-node computer system is a massively parallel computer system.
3. The multi-node computer system of claim 1 wherein the I/O configuration mechanism dynamically updates the ratio by suspending the job and re-allocating additional nodes of the plurality of I/O nodes to be associated with the executing job.
4. The multi-node computer system of claim 1 wherein the I/O characteristics of the job are determined from information stored in a job record selected from the following: job description, historical I/O utilization, and application control parameters.
5. The multi-node computer system of claim 4 wherein the I/O characteristics of the job are used for the initial ratio of input nodes to compute nodes upon beginning execution of the job.
6. The multi-node computer system of claim 1 wherein the block of compute nodes are arranged in a virtual tree network and the I/O node connects to the top of the tree network to allow the block of compute nodes to communicate with the service node of a multi-node computer system.
7. The multi-node computer system of claim 6 wherein the virtual tree network is determined by a class routing table on the node.
8. An article of manufacture comprising:
an I/O configuration mechanism that adjusts a ratio of a plurality of input/output (I/O) nodes to a plurality of compute nodes in a multi-node computer system for a job executing on the plurality of compute nodes, wherein the plurality of I/O nodes provide communication to network resources and the plurality of I/O nodes are connected to the plurality of compute nodes with a virtual network operating on a physical network, wherein the I/O configuration mechanism dynamically adjusts the ratio by dynamically configuring the virtual network based on I/O characteristics of the job, wherein the plurality of compute nodes and the plurality of I/O nodes each comprise a processor and memory, wherein the I/O characteristics of the job are determined by real-time monitoring of the job's I/O characteristics, wherein the I/O configuration mechanism dynamically updates the ratio by suspending the job and re-allocating additional nodes of the plurality of I/O nodes to be associated with the executing job; and
non-transitory computer recordable media bearing the I/O configuration mechanism.
9. The article of manufacture of claim 8 wherein the I/O characteristics of the job are determined from information stored in a job record selected from the following: job description, historical I/O utilization, and application I/O control.
10. The article of manufacture of claim 8 wherein the I/O characteristics of the job are used for the initial ratio of input nodes to compute nodes upon beginning execution of the job.
11. The article of manufacture of claim 8 wherein the plurality of compute nodes are arranged in a virtual tree network and further comprising an I/O node that connects to the top of the tree network to allow the compute nodes to communicate with a service node of a massively parallel computer system.
12. The article of manufacture of claim 11 wherein the virtual tree network is determined by a class routing table in the node.