Storage Array Predicts Event Preferences to Reduce Polling
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
Current data storage systems face inefficiencies due to frequent polling by hosts seeking event updates, leading to increased computational resource usage and network bandwidth consumption.
Innovation Solution
A storage array system that predicts future event-response preferences using heat maps, proactively sends unsolicited data to hosts, and updates approved preferences to reduce duplicate requests and resource utilization.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Reliability
If hosts frequently poll for event updates, then hosts can obtain timely event information, but computational resource usage and network bandwidth consumption increase
Solution Approach 1:
The storage array proactively sends unsolicited event data to hosts before hosts need to request it. The system predicts which events hosts will likely need and pushes this data in advance, eliminating the need for frequent polling and reducing computational overhead while maintaining timely information delivery
Solution Approach 2:
The system monitors host behavior patterns and uses this feedback to refine predictions about which events hosts will need. By analyzing historical polling patterns and event data requests, the system learns to send more accurate predictions, further reducing unnecessary data transmission and computational resources
2Reliability
If hosts frequently poll for event updates, then hosts can obtain timely event information, but network bandwidth consumption increases
Solution Approach 1:
The storage array proactively sends unsolicited event data to hosts before hosts need to request it. The system predicts which events hosts will likely need and pushes this data in advance, eliminating the need for frequent polling and reducing network bandwidth consumption while maintaining timely information delivery
Solution Approach 2:
The system creates and sends copies of relevant event data to hosts based on predicted needs. Instead of hosts repeatedly requesting the same data through polling, the storage array generates and transmits appropriate data copies proactively, reducing network traffic
3Productivity
If the storage array sends unsolicited data proactively, then duplicate requests are reduced, but the complexity of predicting host preferences increases
Solution Approach 1:
The storage array system performs self-learning by automatically analyzing host polling patterns and event data requests. The system improves its prediction capabilities over time without external intervention, using the hosts' own behavior patterns to refine predictions about which unsolicited data they will need
Solution Approach 2:
The system dynamically adjusts prediction parameters based on observed host behavior. By monitoring changes in polling frequency, event types requested, and temporal patterns, the system adapts its prediction model parameters to match actual host needs, improving accuracy without requiring complex manual configuration
Data Source
AI summary
A storage array for providing data storage services includes persistent storage that stores approved preferences and a storage array manager that obtains event-response preferences for hosts receiving the data storage services; predicts, based on the event-response preferences, a set of additional event-response preferences for a host of the hosts; predicts an additional event-response preference for a host of the hosts based on a request associated with a second host of the hosts; generates a heat map for the host based on: the event-response preferences, the set of additional event-response preferences, the additional event-response preference, and global event-response preferences that associate events with responses for all of the hosts; identifies a final set of event-response preferences for the host using the heat map; updates the approved preferences based on the final set of event-response preferences; and provides unsolicited data to the host based on the updated approved preferences.