High Performance Distributed OLAP on Property Graphs with Grasper

Chen, Hongzhi; Wu, Bowen; Deng, Shiyuan; Huang, Chenghuan; Li, Changji; Li, Yichao; Cheng, James

doi:10.1145/3318464.3384685

Cited by 10 publications

(4 citation statements)

References 2 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This work focuses on graph storage, and could potentially be combined with research on execution models, such as those presented in Tao [25] and Grasper [26] .…”

Section: Related Workmentioning

confidence: 99%

Building a High-Performance Graph Storage on Top of Tree-Structured Key-Value Stores

Lin,

Wang,

et al. 2024

Big Data Min. Anal.

View full text Add to dashboard Cite

Graph databases have gained widespread adoption in various industries and have been utilized in a range of applications, including financial risk assessment, commodity recommendation, and data lineage tracking. While the principles and design of these databases have been the subject of some investigation, there remains a lack of comprehensive examination of aspects such as storage layout, query language, and deployment. The present study focuses on the design and implementation of graph storage layout, with a particular emphasis on tree-structured key-value stores. We also examine different design choices in the graph storage layer and present our findings through the development of TuGraph, a highly efficient single-machine graph database that significantly outperforms well-known Graph DataBase Management System (GDBMS).Additionally, TuGraph demonstrates superior performance in the Linked Data Benchmark Council (LDBC) Social Network Benchmark (SNB) interactive benchmark.

show abstract

“…This work focuses on graph storage, and could potentially be combined with research on execution models, such as those presented in Tao [25] and Grasper [26] .…”

Section: Related Workmentioning

confidence: 99%

Building a High-Performance Graph Storage on Top of Tree-Structured Key-Value Stores

Lin,

Wang,

et al. 2024

Big Data Min. Anal.

View full text Add to dashboard Cite

show abstract

“…If one operator has a relatively higher load than the others, G-Tran will use multiple threads to process this operation in parallel in order to achieve a shorter completion time. We follow the state-of-the-art MPP implementation from Grasper [15] while also take into consideration the sideeffects of NUMA architecture [28,47]. As shown in Figure 4, each thread in G-Tran's worker thread pool is bound with one CPU core and further divided into several logical thread regions.…”

Section: Distributed Transaction Processing With Mppmentioning

confidence: 99%

“…However, they have no MPP-based distributed transaction processing and most of them support only low isolation level (e.g., snapshot, read committed). Both Grasper [15,18] and TigerGraph [23] target at massive graph queries with native graph store, but Grasper focuses only on OLAP workload instead of OLTP and TigerGraph is not designed for high performance. A1 [12] is an in-memory graph database built upon FaRM [26,27], while it has no graph native design on memory management and transactional optimizations.…”

Section: Related Workmentioning

confidence: 99%

G-Tran: Making Distributed Graph Transactions Fast

Chen,

Li,

Zheng

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

Graph transaction processing raises many unique challenges such as random data access due to the irregularity of graph structures, low throughput and high abort rate due to the relatively large read/write sets in graph transactions. To address these challenges, we present G-Tran -an RDMA-enabled distributed in-memory graph database with serializable and snapshot isolation support. First, we propose a graph-native data store to achieve good data locality and fast data access for transactional updates and queries. Second, G-Tran adopts a fully decentralized architecture that leverages RDMA to process distributed transactions with the MPP model, which can achieve high performance by utilizing all computing resources. In addition, we propose a new MV-OCC implementation with two optimizations to address the issue of large read/write sets in graph transactions. Extensive experiments show that G-Tran achieves competitive performance compared with other popular graph databases on benchmark workloads.

show abstract

“…Such data not only need to be efficiently stored but also efficiently analyzed. Therefore, some OLAP-like analysis approaches from graph data have been recently proposed, e.g., Chen et al (2020), Ghrab et al (2018), Ghrab et al (2021), and Schuetz et al (2021). Thus, combining graph and OLAP technologies offers ways of analyzing graphs in a manner already well accepted by the industry (Richardson et al, 2021).…”

mentioning

confidence: 99%

Advances on Data Management and Information Systems

Darmont

Novikov²,

Wrembel

et al. 2022

Inf Syst Front

View full text Add to dashboard Cite

The research and technological area of data management encompasses various concepts, techniques, algorithms and technologies, including data modeling, data integration and ingestion, transactional data management, query languages, query optimization, physical data storage, data structures, analytical techniques (including On-Line Analytical Processing -OLAP), as well as service creation and orchestration (Garcia-Molina et al., 2009). Data management technologies are core components of every information system, either centralized or distributed, deployed in an onpremise hardware architecture or in a cloud ecosystem. Data management technologies have been used in commercial, mature products for decades. They were originally developed for managing structured data (mainly expressed in the relational data model).Yet, the ubiquitous big data (Azzini et al., 2021) require development of new data management techniques, suitable for the variety of data formats (from structured, through semi-structured, to unstructured), overwhelming data volumes and velocity of big data generation. These new techniques draw upon the concepts applied to managing relational data.

show abstract

High Performance Distributed OLAP on Property Graphs with Grasper

Cited by 10 publications

References 2 publications

Building a High-Performance Graph Storage on Top of Tree-Structured Key-Value Stores

Building a High-Performance Graph Storage on Top of Tree-Structured Key-Value Stores

G-Tran: Making Distributed Graph Transactions Fast

Advances on Data Management and Information Systems

Contact Info

Product

Resources

About