GraphScope Blog

In today’s data analysis landscape, the optimization of complex graph queries has been a persistent challenge for the industry. GOpt is an optimizer framework designed for complex graph queries and was officially presented at SIGMOD Industry 2025. Its core objective is to provide an efficient and universal optimization solution that adapts to various graph query languages and execution engines. In this article, we will delve into the key innovations of GOpt in unified graph query optimization and analyze its optimization process through specific query cases from LDBC SNB.

The GraphScope team’s work on proposing RelGo for optimizing SQL/PGQ queries has been accepted by SIGMOD 2025. In this work, RelGo integrates GOpt into relational databases (using DuckDB as an example), enhancing its ability to optimize SQL/PGQ queries and providing better optimization results than DuckDB’s own optimizer. This article introduces the main content of RelGo.

From June 9th to June 15th, 2024, SIGMOD 2024 was held in Santiago, Chile. The GraphScope team presented their paper “GraphScope Flex: LEGO-like Graph Computing Stack” at the SIGMOD Industry Session. This article introduces the main content of that paper.

From June 9th to June 15th, 2024, SIGMOD 2024 was held in Santiago, Chile. Graph computing remains a hot topic at this conference, and it is also the area that received the most paper submissions. SIGMOD 2024 organized a panel discussion on graph computation titled “The Future of Graph Analytics”.

Recently, LDBC released the latest results for the LDBC SNB Interactive benchmark test, where GraphScope Flex leads the pack once again with a score exceeding 127,000 QPS (Queries Per Second), representing a more than 2.6 times improvement over the second place, which was the previous record holder!

In this blog, we introduce GOpt, which is a unified graph query optimization framework in GraphScope. GOpt enables the system to support multiple graph query languages while providing consistent and efficient query optimization. We also present two practical cases to demonstrate the effectiveness of our optimizer.

We are pleased to announce an array of improvements in the GraphScope 0.26.0 release. In this release, under the original GraphScope framework, the persistent storage Groot of the Graph Interactive Query Engine (GIE) allows users to launch a Secondary Instance in read-only mode, thereby enhancing the performance of reading graph data. Under the GraphScope Flex architecture, the graph query engine GraphScope Interactive, which is designed for high-concurrency scenarios, now supports running on macOS and has introduced compaction operations for graph data.

We are glad to announce a suite of upgrades in the latest GraphScope 0.25.0 release, bringing significant improvements to the platform. Starting with this version, our updates will be divided into two parts: one is the updates introduced under the original GraphScope framework (including the graph analytics engine GAE, graph interactive engine GIE, and graph learning engine GLE); the other is the latest product features built for the new GraphScope Flex architecture.

GraphAr is an open source, standard data file format for graph data storage and retrieval. It defines a standardized file format for graph data, and provides a set of interfaces for generating, accessing, and transforming these formatted files. This post is a quick guide that shows how to import and export graph data of Neo4j with GraphAr.

We are pleased to introduce a range of enhancements to GraphScope with the GraphScope 0.24.0 release. This release includes a graph query engine specifically designed for high-QPS (queries per second) scenarios. It also encompasses notable features and improvements in the Interactive Engine (GIE), Learning Engine (GLE), and deployment processes.

GraphAr is an open source, standard data file format for graph data storage and retrieval. It defines a standardized file format for graph data, and provides a set of interfaces for generating, accessing, and transforming these formatted files. This post is a quick guide that explains how to work with GraphAr, using the C++ SDK it provides.

In this post, we will introduce GraphAr, which is an open source, standard data file format for graph data storage and retrieval. It defines a standardized file format for graph data, independent of the computation/storage system, and provides a set of interfaces for generating, accessing, and transforming these formatted files.

In this post, we will introduce FLASH, which is a distributed programming model for programming a broad spectrum of graph algorithms, including clustering, centrality, traversal, matching, mining, etc. It makes diverse complex graph algorithms easy to write at the distributed runtime. The algorithms expressed in FLASH take only a few lines of code, and provide a satisfactory performance.

Graph algorithms serve as essential building blocks for a wide range of applications, such as social network analytics, routing, constructing protein network and De Bruijn graphs, and mining valuable information in RDF (Resource Description Framework) graphs. Generally, graph analytics involve propagating labels across edges or iteratively accumulating values from adjacent vertices. Existing engines in both academia and industry, like PowerGraph, Pregel, and GraphX, have paved the way. However, in the era of big data, the computational and storage complexity of sophisticated algorithms coupled with rapidly growing datasets have exhausted the limits of a single device.

Graph neural networks(GNNs) learn graph vertex representations by aggregating multi-hop neighbor information. Industrial applications often adopt mini-batch training to scale out GNNs on large-scale graphs, where neighbor sampling is used during both model training and inference. Since the structure and attributes of real-world graphs often change dynamically, it is imperative that the inferred vertex representation can accurately reflect these updates.