Distributed Result Retrieval After Server Failure
Find Innovative SolutionsGenerate Solutions
Solution Overview
Problem
In systems where distributed processing is performed by multiple servers, a failure in one server can prevent the completion of the entire processing task due to the inability to acquire the necessary processing results.
Innovation Solution
An information processing apparatus and method that provides an execution instruction for processing, acquires processing results from different servers when a failure occurs, and notifies servers of incomplete processing, allowing for the recovery of results from alternative servers.
Engineering Contradictions & Design Principles
Engineering Contradiction Analysis
1Productivity
If distributed processing is performed by multiple servers, then processing capacity and efficiency are improved, but system reliability deteriorates because a failure in one server can prevent completion of the entire processing task
Solution Approach 1:
The system performs preliminary actions by notifying servers in advance that processing results are needed, and servers prepare results accordingly. The notification mechanism is established before failure occurs, allowing the system to proactively manage result collection and initiate alternative acquisition processes if failures are detected.
Solution Approach 2:
The patent implements beforehand cushioning by creating a notification mechanism that alerts servers about the need for processing results before actual failures occur. This allows the system to prepare alternative acquisition strategies and buffer against potential server failures, ensuring that processing can continue even when individual servers fail.
2Loss of information
If processing results are acquired from all servers in distributed processing, then complete processing results are obtained, but processing time increases when waiting for all servers including those that may fail
Solution Approach 1:
The system applies partial action by acquiring processing results from servers that are confirmed to be operational rather than waiting for all servers. The notification mechanism enables the system to collect results from available servers and proceed with processing, avoiding excessive waiting time while still obtaining sufficient results to complete the task.
Solution Approach 2:
The patent implements feedback through a notification system that provides real-time information about which servers have completed processing and which are still working or have failed. This feedback mechanism allows the system to dynamically adjust result acquisition strategies, collecting results from operational servers while identifying failures and initiating alternative acquisition processes.
3Ease of operation
If a management server is used to manage distributed processing, then processing coordination is improved, but device complexity increases
Solution Approach 1:
The notification server performs multiple functions: it notifies servers of processing requirements, collects processing results, identifies failed servers, and initiates alternative acquisition processes. By consolidating these management functions into a single multi-functional server, the system achieves effective processing coordination without proportionally increasing overall system complexity.
Solution Approach 2:
The notification server acts as an intermediary between processing servers and the system managing processing tasks. It mediates communication by sending notifications to servers, collecting results from servers, and providing information about server status. This intermediary role simplifies the coordination architecture by centralizing communication management while maintaining relatively simple device structure.
Data Source
AI summary
An information processing apparatus includes a processor configured to provide an execution instruction for processing to a server that executes distributed processing, acquire a processing result of the processing, which is distributed to the server in the distributed processing, from each of servers, in a case where a predetermined condition is satisfied without acquiring the processing result for respective processing related to the distributed processing, provide a notification indicating that the processing result has not been acquired to each server that executes the distributed processing, and after the notification, among the servers that execute the distributed processing, acquire the processing result from a different server that is different from the server to which the processing, which is related to the processing result that has not been acquired in a case where the condition is satisfied, is distributed.


