Large-Scale Graph Building in Dynamic Environments: Low Latency and High Quality

2025-07-14

AI Summary

To address the challenge of balancing low latency and high graph quality in large-scale graph construction under continuously evolving dynamic data, this paper proposes Dynamic GUS—a novel system that uniquely integrates Grale’s high-fidelity graph construction with ScaNN’s millisecond-level approximate nearest neighbor (ANN) retrieval. Dynamic GUS supports multi-feature fusion and real-time graph updates via a dedicated dynamic graph update mechanism and multimodal feature learning, significantly enhancing ANN adaptability and accuracy over streaming data. The system has been deployed in over ten production applications within Google. In Android security and privacy use cases, it achieves a 4× speedup in malicious app identification while maintaining sub-50-ms per-request latency, demonstrating its effectiveness, scalability, and suitability for time-critical, multimodal industrial scenarios.

Technology Category

• Machine Learning: Graph-based Machine Learning
• Data Mining & Knowledge Management: Graph Mining, Social Network Analysis & Community
• Search and Optimization: Distributed Search

Application Category

• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs
• Search and Retrieval-Augmented AI: Efficiency and scalability of Web search engines
• Security and Privacy: Large-scale security measurements

Abstract

Learning and constructing large-scale graphs has attracted attention in recent decades, resulting in a rich literature that introduced various systems, tools, and algorithms. Grale is one of such tools that is designed for offline environments and is deployed in more than 50 different industrial settings at Google. Grale is widely applicable because of its ability to efficiently learn and construct a graph on datasets with multiple types of features. However, it is often the case that applications require the underlying data to evolve continuously and rapidly and the updated graph needs to be available with low latency. Such setting make the use of Grale prohibitive. While there are Approximate Nearest Neighbor (ANN) systems that handle dynamic updates with low latency, they are mostly limited to similarities over a single embedding. In this work, we introduce a system that inherits the advantages and the quality of Grale, and maintains a graph construction in a dynamic setting with tens of milliseconds of latency per request. We call the system Dynamic Grale Using ScaNN (Dynamic GUS). Our system has a wide range of applications with over 10 deployments at Google. One of the applications is in Android Security and Privacy, where Dynamic Grale Using ScaNN enables capturing harmful applications 4 times faster, before they can reach users.

Problem

• Handling dynamic graph updates with low latency
• Supporting multiple feature types in graph construction
• Improving real-time detection of harmful applications

Innovation

• Dynamic Grale Using ScaNN for low latency
• Handles multiple feature types dynamically
• Enables rapid graph updates in milliseconds