Request Isolation Mechanism for Server Load Reduction

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In network systems, redundant or expensive requests for resources can flood servers, leading to performance issues as multiple requests for the same resource are not effectively isolated, causing unnecessary load on servers.

Innovation Solution

Implementing a method where a first server receives requests for resources stored on a second server, stores entries in data structures to track client devices and request references, and cancels requests when a cancellation is detected, ensuring only one request is made to the second server, even if multiple requests are received from client devices.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If multiple requests for the same resource are allowed to pass through to the second server, then client devices can access resources they need, but the second server experiences overload and performance degradation

Engineering Contradiction:

Improveresource access efficiencyVSAvoidserver overload

Core Design Contradiction:

ProductivityVSObject-affected harmful factors

Solution Approach 1:

The patent segments requests by tracking them individually in a data structure that associates each request with a unique identifier. This allows the system to distinguish between redundant and non-redundant requests, enabling selective blocking of duplicate requests while allowing unique requests to proceed to the second server.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent implements preliminary action by checking whether a request for a resource already exists in the data structure before forwarding the request to the second server. This proactive approach prevents redundant requests from reaching the server, thereby avoiding overload before it occurs.

Inventive Principle:

Principle #10Preliminary action

2Reliability

If redundant requests are blocked at the first server, then the second server is protected from overload, but the system requires additional data structures and tracking mechanisms

Engineering Contradiction:

Improveserver performance stabilityVSAvoidrequest tracking system

Core Design Contradiction:

ReliabilityVSDevice complexity

Solution Approach 1:

The patent introduces an intermediary data structure that sits between the first server and the second server to track requests. This intermediary component absorbs the complexity of request management, protecting the second server from direct exposure to redundant requests while maintaining a relatively simple implementation through the use of hash tables or similar data structures.

Inventive Principle:

Principle #24Intermediary (Mediator)

3Productivity

If the system tracks and cancels requests using reference counts, then redundant requests are effectively isolated, but additional memory operations and processing are required

Engineering Contradiction:

Improverequest isolation efficiencyVSAvoidrequest processing overhead

Core Design Contradiction:

ProductivityVSLoss of time

Solution Approach 1:

The patent replaces complex mechanical request tracking with a data structure-based system using hash tables or similar efficient data structures. This substitution enables O(1) average-time complexity for request lookups and insertions, significantly reducing the time overhead compared to traditional mechanical or sequential tracking methods while maintaining effective request isolation.

Inventive Principle:

Principle #28Mechanics substitution (Replace mechanical system)

Data Source

PatentUS9674277B1Systems and methods for request isolation protection

Publication Date: 2017.06.06 GOOGLE LLC

AI summary

Systems and methods for request isolation protection at a first server. The first server receives a resource request from a client device. The first server may receive multiple requests from the same client device or other client devices. The first server caches the multiple requests so that only one resource request is made to a second server storing the resource. The first server detects when the client device cancels the request to the resource. In response, the first server cancels the request made to the second server if there are no other requests made to the resource. If there are other requests remaining, then the first server does not cancel the request to the second server.