Data Center Load Estimation for Switching Control
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Existing information processing systems face performance reductions when switching data center destinations, as the new destination may experience increased loads due to high user activity in specific time zones, leading to system overload.
Innovation Solution
An information processing apparatus that calculates estimated load values for each data center based on received load information and time zone considerations to determine the most suitable data center for processing requests, thereby reducing the likelihood of switching to a heavily loaded destination.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Ease of operation
If the access destination is switched to the closest data center when the current data center is under heavy load, then the user can access a nearby data center, but the destination data center may experience heavy load due to high user activity in its time zone, leading to performance reduction
Solution Approach 1:
The system performs preliminary actions by calculating estimated load values for candidate data centers before switching the access destination. The load calculation unit computes the estimated load for each candidate data center based on current load information and time zone characteristics, allowing the system to predict future load conditions and avoid switching to data centers that would become overloaded, thus preventing performance degradation before it occurs
Solution Approach 2:
The system dynamically adjusts the data center selection criteria by incorporating time zone-based load estimation. Instead of using a static selection rule (closest data center), the system dynamically evaluates candidate data centers based on their estimated load values, which vary depending on the time zone and current load conditions. This dynamic approach allows the system to adapt to changing load patterns and maintain reliable performance
2Productivity
If load balancing is achieved by switching to the closest data center, then the current overloaded data center can offload requests, but the destination data center's load increases, potentially causing system overload
Solution Approach 1:
The system performs preliminary load estimation for candidate data centers before executing the switching operation. By calculating the estimated load values in advance based on time zone characteristics and current load information, the system can predict the impact of load redistribution and avoid switching to data centers that would become overloaded, thus maintaining balanced load distribution across the system
Solution Approach 2:
The system implements a feedback mechanism where the load calculation unit continuously monitors load information from multiple data centers and uses this feedback to adjust switching decisions. The estimated load value serves as feedback that informs the switching control unit whether to proceed with switching to a candidate data center, ensuring that load balancing actions do not inadvertently create new overload conditions
Data Source
AI summary
An information processing apparatus includes: a memory; and a processor coupled to the memory and configured to: receive, from each of a plurality of data centers located in a plurality of positions, load information pertaining to a load of each data center; calculate an estimated load value for each of the plurality of data centers based on the received load information; determine a data center to execute a requested process based on the estimated load value and a time zone of the position where each of the plurality of data centers is located; and control the determined data center to execute the requested process.