Network aware process scheduling
View Patent ↗A schedule graph may be used to identify executable elements that consume data from a network interface or other input/output interface. The schedule graph may be traversed to identify a sequence or pipeline of executable elements that may be triggered from data received on the interface, then a process scheduler may cause those executable elements to be executed on available processors. A queue manager and a load manager may optimize the resources allocated to the executable elements to maximize the throughput for the input/output interface. Such as system may optimize processing for input or output of network connections, storage devices, or other input/output devices.
1. A system comprising:
a plurality of processors, receiving input stream of data from a source;
a queue manager that:
receives a scheduling graph from an analyzer, said scheduling graph comprising executable elements and relationships between said executable elements, said relationships comprising data dependency relationships;
from said scheduling graph,
identifies a first executable element that receives said input stream from the source;
identifies a first set of executable elements comprising a plurality of executable elements being executed downstream from said first executable element; and
identifies a second set of executable elements comprising a plurality of executable elements being executed upstream from said first executable element;
schedules said second set of executable elements, said second set of executable elements being executed at an elevated priority by preempting other executable elements; and
schedules said first set of executable elements such that a process scheduler causes said first set of executable elements to process said input stream of data;
a load manager that:
monitors performance of said executable elements, by
when processing of said input stream of data is operating at maximum capacity, continues said monitoring;
when processing of said input stream is not operating at maximum capacity, identifies a second executable element downstream from said first executable element, said second executable element being a bottleneck for processing output from said first executable element; and allocates additional processing resources to said second executable element until said executable elements is operating at full capacity.
2. The system of claim 1 , said source being a network input device.
3. The system of claim 1 , said source being a data storage device.
4. A computerized method comprising:
receiving a scheduling graph from an analyzer, said scheduling graph comprising executable elements and relationships between said executable elements, said relationships comprising data dependency relationships;
receiving an input stream, comprising a plurality of data items from a data source, said input stream being consumed by a first executable element defined within said scheduling graph;
identifying a second executable element from said scheduling graph, said second executable element having a data dependency relationship with said first executable element;
identifying a third executable element from said scheduling graph, said first executable element being dependent on said third executable element prior to consuming said input stream;
scheduling said third executable element such that said process scheduler causes said third executable element to be executed prior to said first executable element;
scheduling said second executable element such that a process scheduler causes said second executable element to be executed after said first executable element;
identifying a fourth executable element being dependent on said first executable element, said fourth executable element causing said first executable element to wait to process said input stream; and allocating additional processing resources to said fourth executable element;
determining a current rate at which said first executable element processes said input stream; and measuring a change in said current rate when said fourth executable element has been allocated said additional processing resources;
when said change is positive, adding additional processing resources; and when said change is not positive, removing at least some of said additional processing resources.
5. The method of claim 4 , said second executable element not having a message passing relationship with said first executable element.
6. The method of claim 4 further comprising: storing configuration of resource allocation in the scheduling graph.
7. The system of claim 1 , said scheduling graph being defined within a development environment.
8. The system of claim 1 , said scheduling graph being defined by an analyzer executed after compilation.
9. The system of claim 1 , said dependency relationships being identified after monitoring said executable elements.
10. The system of claim 9 , said dependency relationships comprising a shared data object.
11. The system of claim 1 , said queue manager being part of an execution environment.
12. The system of claim 1 , said additional processing resources comprising executing said second executable element at an elevated priority.
13. The system of claim 1 , said additional processing resources comprising assigning additional processors.
14. The system of claim 1 , said additional processing resources comprising assigning additional memory resources.
15. The method of claim 4 , said additional processing resources comprising executing said fourth executable element an elevated priority.
16. The method of claim 4 , said additional processing resources comprising assigning additional processors.
17. The method of claim 4 , said additional processing resources comprising assigning additional memory resources.