Edge Datastore for Low-Latency Event-Triggered Function Access

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Content delivery networks (CDNs) face latency issues due to the need for network transmission delays to retrieve data from origin databases, and edge location scripts have limited capabilities, including inability to make network calls or access databases for security reasons.

Innovation Solution

Implementing a system with edge datastores that allow low-latency query processing and data retrieval by translating queries and responses across different database formats, enabling edge locations to store and retrieve data locally without network calls, and allowing event-triggered functions to access data from origin databases without incurring high latency.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If data is retrieved from origin database at edge location, then data availability is improved, but network transmission delay increases

Engineering Contradiction:
Improvedata availabilityVSAvoidnetwork transmission delay
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent pre-loads data from the origin database into edge datastores before it is actually needed. Configuration data, client data, and origin database data are fetched and stored at edge locations in advance, so when a request arrives, the data is already available locally without requiring network transmission at request time.

Inventive Principle:
Principle #10Preliminary action

Solution Approach 2:

The patent divides the CDN system into multiple independent components: origin servers, edge datastores, and execution environments. Each edge location maintains its own local datastore and execution environment, allowing independent operation and reducing dependency on remote origin databases for routine operations.

Inventive Principle:
Principle #1Segmentation

2Adaptability or versatility

If scripts are given full database access at edge location, then function capability is improved, but security risk increases

Engineering Contradiction:
Improvefunction capabilityVSAvoidsecurity risk
Core Design Contradiction:
Adaptability or versatilityVSObject-affected harmful factors

Solution Approach 1:

The patent creates isolated execution environments at each edge location with controlled access to local datastores. Scripts run in sandboxed containers that have specific permissions granted to them, allowing them to access and modify data in the local edge datastore while preventing access to other parts of the system. This provides localized capability without system-wide security risks.

Inventive Principle:
Principle #3Local quality

3Speed

If data is cached at edge location, then data retrieval speed is improved, but data consistency with origin database deteriorates

Engineering Contradiction:
Improvedata retrieval speedVSAvoiddata consistency
Core Design Contradiction:
SpeedVSStability of the object's composition

Solution Approach 1:

The patent implements a configuration system that monitors and manages edge datastores. When origin database schemas change or new data becomes available, the configuration system detects these changes and automatically updates the edge datastores. This feedback mechanism ensures that cached data remains consistent with the origin database without requiring constant synchronization.

Inventive Principle:
Principle #23Feedback

Data Source

PatentUS11089136B1Low latency datastore access for event-triggered functions at the edge
Publication Date: 2021.08.10 AMAZON TECH INC
  • US11089136B1 patent drawing
  • US11089136B1 patent drawing
  • US11089136B1 patent drawing

AI summary

Edge functions at an edge location of a content delivery network (CDN) may access data from a back-end database without the need to make high-latency network calls to the back-end databases. When a client sends a query to the edge location, an edge function is triggered. The requested data may be retrieved from a read-only local table at the edge location (populated from a back-end database) and the retrieved data is provided to the edge function's memory for low-latency access during function execution.