PREDICTIVE LOAD BALANCING FOR A DIGITAL ENVIRONMENT
A system and method for predictive load balancing for digital multiplayer gaming or simulated environments. The system includes an environment monitor subsystem that monitors a simulation or game condition and system load condition on a computational node of a networked computer environment over time, the game condition comprising a condition within a computer game operating on the computational node and across system nodes. The system also includes an automated planning service subsystem that performs a predictive analysis resulting in a system load forecast for the computational node and changes the system load on the computational node based on the system load forecast by changing a game control and adds or removes nodes, the game control comprising a limitation on the operation of the computer game operating on the computational node or the relationship between nodes or the number of nodes.
1 . A computing system for predictive load balancing for a networked computer gaming environment employing an automated planning service, the computing system comprising:
one or more hardware processors configured for:
monitoring a game condition and a system load condition on a computational node of a networked computer gaming environment over time, the game condition comprising a condition within a computer game operating on the computational node;
storing time series data on a non-volatile data storage device, the time series data comprising changes of the game condition and the system load condition at a series of time points;
performing a predictive analysis based on the time series data to generate a system load forecast for the computational node; and
changing the system load on the computational node based on the system load forecast by changing a game control, the game control comprising a limitation on the operation of the computer game operating on the computational node.
2 . The computing system of claim 1 , wherein the load on the computational node is changed by adding or removing computational nodes.
3 . The computing system of claim 1 , wherein the load on the computational node is changed by reallocating computing resources from other nodes to the computational node.
4 . The computing system of claim 1 further comprising a connector subsystem, wherein a node may utilize the connector subsystem to communicate and exchange data with other nodes to facilitate load balancing.
5 . The computing system of claim 1 , wherein at least a portion of the load-related data used for predictive analysis is user contribution to system load.
6 . The computing system of claim 1 , wherein the time-series data further comprise scheduled events.
7 . The computing system of claim 1 , wherein the predictive analysis is performed using model-based forecasts or simulations based on the time series data.
8 . The computing system of claim 1 , wherein the game condition further comprises state information about a user agent, a game agent, or another entity.
9 . The computing system of claim 1 , wherein the game condition further comprises player activity data, the player activity data comprising login information, actions, and location data.
10 . The computing system of claim 1 , wherein the load on the computational node is changed by adjusting the relationship between the computational node and at least one other node.
11 . A computer-implemented method executed on an automated planning service for predictive load balancing for a networked computer gaming environment, the computer-implemented method comprising:
monitoring a game condition and a system load condition on a computational node of a networked computer gaming environment over time, the game condition comprising a condition within a computer game operating on the computational node;
storing time series data on a non-volatile data storage device, the time series data comprising changes of the game condition and the system load condition at a series of time points;
performing predictive analysis based on the time series data to generate a system load forecast for the computational node; and
changing the system load on the computational node based on the system load forecast by changing a game control, the game control comprising a limitation on the operation of the computer game operating on the computational node.
12 . The computer-implemented method of claim 11 , further comprising the step of changing the system load on the computational node by adding computational nodes.
13 . The computer-implemented method of claim 11 , further comprising the step of changing the system load on the computational node by reallocating computing resources from other nodes to the computational node.
14 . The computer-implemented method of claim 11 , further comprising a connector subsystem, wherein a node may utilize the connector subsystem to communicate and exchange data with other nodes to facilitate load balancing.
15 . The computer-implemented method of claim 11 , wherein the time-series data further comprise user contribution to the load.
16 . The computer-implemented method of claim 11 , wherein the time-series data further comprise scheduled events.
17 . The computer-implemented method of claim 11 , wherein the predictive analysis is performed using simulations based on the time series data.
18 . The computer-implemented method of claim 11 , wherein the game condition further comprises state information about a user agent, a game agent, or another entity.
19 . The computer-implemented method of claim 11 , wherein the game condition further comprises player activity data, the player activity data comprising login information, actions, and location data.
20 . The computer-implemented method of claim 11 , wherein the load on the computational node is changed by adjusting the relationship between the computational node and at least one other node.
21 . A system for predictive load balancing for a networked computer gaming environment employing an automated planning service, comprising one or more computers with executable instructions that, when execute, cause the system to:
monitor a game condition and a system load condition on a computational node of a networked computer gaming environment over time, the game condition comprising a condition within a computer game operating on the computational node;
store time series data on a non-volatile data storage device, the time series data comprising changes of the game condition and the system load condition at a series of time points;
perform a predictive analysis based on the time series data to generate a system load forecast for the computational node; and
change the system load on the computational node based on the system load forecast by changing a game control, the game control comprising a limitation on the operation of the computer game operating on the computational node.
22 . The system of claim 21 , wherein the load on the computational node is changed by adding computational nodes.
23 . The system of claim 21 , wherein the load on the computational node is changed by reallocating computing resources from other nodes to the computational node.
24 . The system of claim 21 further comprising a connector subsystem, wherein a node may utilize the connector subsystem to communicate and exchange data with other nodes to facilitate load balancing.
25 . The system of claim 21 , wherein at least a portion of the load-related data used for predictive analysis is user contribution to system load.
26 . The system of claim 21 , wherein the time-series data further comprise scheduled events.
27 . The system of claim 21 , wherein the predictive analysis is performed using model-based forecasts or simulations based on the time series data.
28 . The system of claim 21 , wherein the game condition comprises state information about a user agent, a game agent, or another entity.
29 . The system of claim 21 , wherein the game condition further comprises player activity data, the player activity data comprising login information, actions, and location data.
30 . The system of claim 21 , wherein the load on the computational node is changed by adjusting the relationship between the computational node and at least one other node.
31 . Non-transitory, computer-readable storage media having computer-executable instructions embodied thereon that, when execute by one or more processors of a computing system employing an automated planning service for predictive load balancing for a networked computer gaming environment, cause the computing system to:
monitor a game condition and a system load condition on a computational node of a networked computer gaming environment over time, the game condition comprising a condition within a computer game operating on the computational node;
store time series data on a non-volatile data storage device, the time series data comprising changes of the game condition and the system load condition at a series of time points;
perform a predictive analysis based on the time series data to generate a system load forecast for the computational node; and
change the system load on the computational node based on the system load forecast by changing a game control, the game control comprising a limitation on the operation of the computer game operating on the computational node.
32 . The system of claim 31 , wherein the load on the computational node is changed by adding computational nodes.
33 . The system of claim 31 , wherein the load on the computational node is changed by reallocating computing resources from other nodes to the computational node.
34 . The system of claim 31 further comprising a connector subsystem, wherein a node may utilize the connector subsystem to communicate and exchange data with other nodes to facilitate load balancing.
35 . The system of claim 31 , wherein at least a portion of the load-related data used for predictive analysis is user contribution to system load.
36 . The system of claim 31 , wherein the time-series data further comprise scheduled events.
37 . The system of claim 31 , wherein the predictive analysis is performed using model-based forecasts or simulations based on the time series data.
38 . The system of claim 31 , wherein the game condition comprises state information about a user agent, a game agent, or another entity.
39 . The system of claim 31 , wherein the game condition further comprises player activity data, the player activity data comprising login information, actions, and location data.
40 . The system of claim 31 , wherein the load on the computational node is changed by adjusting the relationship between the computational node and at least one other node.