Method and system for determining a plurality of scheduling endpoints in a grid network
View Patent ↗A method for determining a plurality of scheduling endpoints in a grid network, each scheduling endpoint comprising a device for processing job requests communicated across the network, the method comprising the steps of: examining a communication across the network in order to determine a current activity status of a requesting device; identifying an activity history for the requesting device; comparing the current activity status with the activity history to derive a statistical data set; and analysing the statistical data set to derive a time period in which the requesting device is available for processing job requests.
1. A method for assigning a job request to a device in a grid network, the method comprising the steps of:
receiving a job request;
monitoring traffic across the grid network;
copying data packets sent over the grid network by a requesting device;
extracting a set of network parameters from the data packets, the set of network parameters comprising a current activity status of the requesting device, an IP address of the requesting device, an IP address of a destination device, and a type of resource the requesting device is requesting from the destination device;
analyzing the destination IP address and the type of resource to determine a category of device the requesting device falls into, wherein the category of the requesting device is one of a server and a client;
identifying an activity history for the requesting device; deriving a statistical data set by comparing the current activity status with the activity history;
deriving a time period in which the requesting device is likely to be available for processing job requests by analyzing the statistical data set;
adjusting the time period based on the determined category of the requesting device, and;
assigning the job request to the requesting device based on the adjusted time period.
2. A method as claimed in claim 1 , wherein the analyzing step further comprises deriving a probability score indicative of a likelihood of a pattern of activity re-occurring.
3. A method as claimed in claim 1 , wherein examining a copy of a data packet from a requesting device indicates that the requesting device is active and available on the network.
4. A method as claimed in claim 1 , further comprising identifying a type of data packet and performing a lookup, in a data store, to identify a format of the data within the identified data packet type.
5. A method as claimed in claim 4 , wherein network parameters are extracted from the data packet and stored in accordance with the format of the data packet type.
6. A method as claimed in claim 5 , wherein the extracted network parameters further comprise a vendor extension identifying hardware and software installed on the requesting device.