PolarDB Brings Alibaba Cloud SIGMOD Best Paper Award

On June 13, at SIGMOD 2024, the leading international academic conference on databases held in Chile, Alibaba Cloud's self-developed PolarDB architecture won the Best Paper Award of the Industry Track. This is the first time a Chinese company's independent output has received this top honor from SIGMOD. PolarDB, an innovative design by Alibaba Cloud, is a multi-primary database that utilizes disaggregated shared memory. It can achieve up to three times the performance of similar databases in the industry. PolarDB is currently widely used in commercial applications.

Alibaba Cloud is the Chinese technology company with the highest number of papers selected for SIGMOD 2024, with a total of 14.

Figure: PolarDB wins the SIGMOD 2024 Best Industrial Paper Award

ACM SIGMOD is the most influential top academic conference in the field of databases. One of the core technical directions it focuses on the most is database architecture. Recent years have seen a gradual shift from traditional primary-secondary architectures to multi-primary architectures and the emergence of various research directions including shared-nothing and shared storage. However, these architectures still face the problems of low performance and inefficient data fusion in high-concurrency scenarios.

Alibaba Cloud is the first in the industry to propose a multi-primary architecture based on disaggregated shared memory and storage, and has completed the world's first cloud-native multi-primary database practice and commercial deployment with PolarDB. Specifically, the new PolarDB architecture introduces the Polar Multi-Primary Fusion Server (PMFS), which is based on disaggregated shared memory and storage and efficiently supports multiple primary nodes, each capable of independently processing transactions and accessing all data. Test data shows that in a high-concurrency scenario with eight nodes, PolarDB delivers three times the throughput of similar databases in the industry.

Image: In a SysBench workload test with eight nodes, PolarDB delivers three times the throughput of Taurus-MM

The SIGMOD review committee commented, "PolarDB solves the challenges of conflict handling and data fusion in multi-primary architectures in an innovative way through disaggregated shared memory. The new architecture has been validated in production and wide application to have strong practical value."

The paper, PolarDB-MP: A Multi-Primary Cloud-Native Database via Disaggregated Shared Memory, written independently by the Alibaba Cloud ApsaraDB team, won the Best Paper Award of the Industry Track at SIGMOD 2024. This is another top honor for Alibaba Cloud after the Best Paper Award of the Research Track at SIGMOD 2023 and the Best Paper Award at ICDE 2024. The paper is the 17th award-winning paper among more than 330 papers published by Alibaba Cloud in top journals and conferences in recent years.

PolarDB has achieved a growth rate of 400% in the past three years and is widely used in the core business systems of more than 10,000 customers in various fields, including government, finance, telecommunications, logistics, and the Internet. Alibaba Cloud has been ranked first in database market share in China for years, and has been recognized as a leader in the Gartner Magic Quadrant™ for Cloud Database Management Systems for four consecutive years, becoming the first and only Chinese company to be named a leader in basic software.

Community

PolarDB Brings Alibaba Cloud SIGMOD Best Paper Award

Read previous post:

Read next post:

ApsaraDB

You may also like

Comments

ApsaraDB

Related Products

PolarDB for PostgreSQL

PolarDB for Xscale

PolarDB for MySQL

Database for FinTech Solution