Cloud Event Queue Segmentation for Latency Control

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

In cloud data centers, there is a challenge in balancing operational costs while providing services, as some customers may overload the event server queues with asynchronous tasks, leading to unacceptable delays for other customers with simpler tasks.

Innovation Solution

Implementing a method where each customer has a dedicated event queue, allowing for concurrent dequeuing and prioritization of events based on customer-specific needs, ensuring that tasks are processed efficiently without overloading the system.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of energy

If a single shared event queue is used for all customers, then operational costs are reduced, but service quality deteriorates due to unacceptable delays for customers with simpler tasks

Engineering Contradiction:

Improveoperational costsVSAvoidtask processing delay

Core Design Contradiction:

Loss of energyVSLoss of time

Solution Approach 1:

The patent segments the single shared event queue into multiple customer-specific event queues. Each customer's events are routed to their dedicated queue, preventing one customer's events from blocking others. This segmentation resolves the contradiction by maintaining cost efficiency through a single event server while eliminating delays through isolated queue processing.

Inventive Principle:

Principle #1Segmentation

2Loss of time

If multiple event servers are deployed to prevent delays, then service quality improves, but operational costs increase

Engineering Contradiction:

Improvetask processing delayVSAvoidoperational costs

Core Design Contradiction:

Loss of timeVSLoss of energy

Solution Approach 1:

Instead of deploying multiple event servers, the patent segments the event processing into multiple virtual queues within a single server. This allows simultaneous processing of multiple customers' events through concurrent dequeuing operations, achieving service quality improvement without the cost of additional servers.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The single event server is made multi-functional by implementing support for multiple customer-specific queues. The server can concurrently process events from different queues, making one server perform the work of multiple servers while maintaining cost efficiency.

Inventive Principle:

Principle #6Universality (Multi-functionality)

3Adaptability or versatility

If asynchronous tasks with long processing times are allowed, then system flexibility improves, but other customers' tasks experience unacceptable delays

Engineering Contradiction:

Improvesystem flexibilityVSAvoidtask processing delay

Core Design Contradiction:

Adaptability or versatilityVSLoss of time

Solution Approach 1:

The patent segments the event processing system into customer-specific queues, allowing each customer's tasks to be processed independently. This enables the system to accommodate tasks with varying processing times without one customer's long-running tasks blocking others, maintaining both flexibility and responsiveness.

Inventive Principle:

Principle #1Segmentation

Data Source

PatentUS9990240B2Event handling in a cloud data center

Publication Date: 2018.06.05 SUCCESSFACTORS INC

AI summary

Disclosed is event processing a computing center, which may include receiving events from users of the computing center to be processed. Each received event may be stored in an event queue that is associated with a customer of the user. Events in an event queue may then be processed by an event processor that is associated with that event queue.