Distributed Data Caching for Low-Latency Ad Bidding

Resolve Bottlenecks,
Find Innovative SolutionsGenerate Solutions

Solution Overview

Problem

The existing advertising exchange systems face significant latency issues due to centralized data storage, which hampers the speed and efficiency of ad bidding processes, leading to suboptimal user experiences and reduced ad impression value.

Innovation Solution

Implementing a low-latency high-throughput scalable data caching system where user profiles are stored on load balancer servers with local data caches, reducing network latency and search times by caching data in fast memory, and correlating cache locations with user access regions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Speed

If centralized data storage is used, then data consistency is maintained, but latency increases and throughput decreases

Engineering Contradiction:

Improvedata retrieval speedVSAvoidsystem architecture complexity

Core Design Contradiction:

SpeedVSDevice complexity

Solution Approach 1:

The patent segments the centralized data storage system into multiple distributed data caches across different load balancer servers. Each cache stores a portion of the user profile data, allowing parallel access and reducing retrieval latency. The segmentation enables the system to handle multiple requests simultaneously without contending for a single centralized resource.

Inventive Principle:

Principle #1Segmentation

Solution Approach 2:

The patent introduces a spatial dimension to data storage by distributing caches across multiple servers and geographic locations. User profiles are replicated across different data caches in different regions, allowing the system to serve users from the nearest cache location, thereby reducing network latency and improving retrieval speed.

Inventive Principle:

Principle #17Another dimension (Dimensionality change)

2Loss of time

If data is cached locally on load balancer servers, then latency is reduced, but data consistency and synchronization become more difficult

Engineering Contradiction:

Improvedata access latencyVSAvoiddata consistency

Core Design Contradiction:

Loss of timeVSReliability

Solution Approach 1:

The patent implements a feedback mechanism where data caches periodically synchronize with the central data source or with each other. When data is updated in one cache, the system detects the change and propagates the update to other caches, ensuring eventual consistency. This feedback loop maintains data reliability while allowing local caches to operate independently for low-latency access.

Inventive Principle:

Principle #23Feedback

Solution Approach 2:

The patent performs preliminary actions by pre-caching user profile data in multiple data caches before it is needed. User profiles are proactively replicated across different caches based on predicted access patterns, so that when a request arrives, the data is already available locally, eliminating latency. This preliminary caching maintains consistency by ensuring all caches have the latest data before access occurs.

Inventive Principle:

Principle #10Preliminary action

3Ease of operation

If user profiles are stored in centralized databases, then data management is simplified, but search times and retrieval latency increase

Engineering Contradiction:

Improvedata management simplicityVSAvoidsearch and retrieval time

Core Design Contradiction:

Ease of operationVSLoss of time

Solution Approach 1:

The patent creates copies of user profile data and stores them in multiple data caches distributed across load balancer servers. Instead of managing a single centralized database, the system maintains replicated copies in different locations. This copying approach reduces retrieval time by serving data from the nearest cache while simplifying management through automated replication and synchronization mechanisms.

Inventive Principle:

Principle #26Copying

4Productivity

If data caches are distributed across multiple servers, then throughput and scalability improve, but system complexity and coordination overhead increase

Engineering Contradiction:

Improvesystem throughputVSAvoidsystem coordination complexity

Core Design Contradiction:

ProductivityVSDevice complexity

Solution Approach 1:

The patent makes each data cache universal by designing them to perform the same functions: storing user profiles, handling read requests, and participating in synchronization. Each load balancer server with its cache can independently serve multiple users and handle various operations, improving throughput through parallelism. The universal design reduces coordination complexity because all caches follow the same protocols and interfaces.

Inventive Principle:

Principle #6Universality (Multi-functionality)

Data Source

PatentUS10432706B2Low-latency high-throughput scalable data caching

Publication Date: 2019.10.01 CADENT LLC

AI summary

Low-latency high-throughput scalable data caching is disclosed. For example, a data source is connected over a network to a load balancer server with data cache. A load balancer service and a data cache service execute on processors on the first load balancer server to receive, by the load balancer service, a request from a client device over the network. The load balancer service requests a data entry associated with the request from the first data cache service. The data cache service retrieves the first data entry from the first data cache, which stores a first plurality of data entries that is a subset of a second plurality of data entries stored in the data source. The load balancer service modifies the request with the data entry. The load balancer service sends a modified request to a plurality of receivers.