.png){kind=icon}
返回
AnalyticDB-PG: A Cloud-Native High-Performance Data Warehouse in Alibaba Cloud
.png){kind=icon}
PRE
AI
DOI：10.14778/3750601.3750633.png)
.png){kind=icon}
原文求助.png){kind=icon}
分享.png){kind=icon}
收藏
摘要
En 中文
In the era of big data, the landscape of data management and ana-lytics has significantly transformed, presenting diverse challengesfor cloud platforms. Modern data warehouses face increasing chal-lenges in handling hybrid transactional and analytical processing(HTAP) workloads efficiently in cloud environments. Traditionalshared-nothing architectures provide high-performance query exe-cution but suffer from high storage costs and limited elasticity, whileshared-storage approaches improve scalability but often strugglewith query efficiency due to increased data movement and indexingoverhead. Furthermore, existing execution engines lack optimizedsupport for vectorized processing and real-time analytics, limitingtheir ability to handle large-scale workloads efficiently. To address these limitations, we introduce AnalyticDB-PG (ADB-PG), a cloud-native, high-performance data warehouse designedfor modern analytical workloads. It integrates a unified architec-ture supporting both Shared-Nothing and Shared-Storage modes,allowing flexible deployment and seamless elasticity. In ADB-PG,we introduce Beam, a hybrid storage engine that efficiently bal-ances row-based and columnar storage for real-time analytics, andLaser, an optimized execution engine leveraging vectorized execu-tion and Just-In-Time compilation to accelerate query processing.The system further incorporates advanced indexing mechanisms,adaptive runtime filtering, and dictionary encoding to enhanceperformance. Extensive evaluations on TPC-H and TPC-DS bench-marks demonstrate that ADB-PG achieves significant performanceimprovements while reducing storage and operational costs, mak-ing it a compelling solution for modern cloud-based data analytics
Keyword:
CHALLENGES
期刊
P
IF:
3.3
论文数:
563
被引数:
1.2W
机构
.png){kind=icon}
查看更多内容