How Good is My HTAP System?

Milkai, Elena; Chronis, Yannis; Gaffney, Kevin P.; Guo, Zhihan; Patel, Jignesh M.; Yu, Xiangyao

doi:10.1145/3514221.3526148

Cited by 7 publications

(4 citation statements)

References 25 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This approach is followed by [18,78,111,150,168]. Milkai et al extend this approach by visualizing the throughput frontier in a 2D chart [117]. In the chart, the xaxis represents OLTP throughput, and the y-axis represents OLAP throughput.…”

Section: Evaluation Methods Of Performance Isolationmentioning

confidence: 99%

“…To gauge data freshness, the metric should quantify how recently each analytical query sees the view (a.k.a. snapshot) of OLTP data in an HTAP system [117]. We summarize existing freshness metrics into two categories.…”

Section: Metrics Of Data Freshnessmentioning

confidence: 99%

“…For instance, PostgreSQL-SR [117] replicates data by replaying Write-Ahead Logging (WAL) records in a streaming manner (i.e., without waiting for the WAL to be filled), which makes the replica stay more up-to-date. To serve a hybrid workload, the primary is generally used for OLTP, and the replica serves read-only queries.…”

Section: Leveraging Replication Mechanismmentioning

confidence: 99%

“…HTAPBench [47] follows a similar schema design, confining the schema from TPC-C and TPC-E. Additionally, HTAPBench proposes a new evaluation metric: QpHpW. Instead of evaluating transactions and queries using inde- HTAPTrick [117] bridges TPC-C and TPC-H with a star schema (i.e., TPC-H SSB [128]). Star schema provides a clear division between dimension tables and fact tables, where dimension tables are most likely to join fact tables with their primary keys.…”

Section: Benchmarksmentioning

confidence: 99%

See 3 more Smart Citations

A survey on hybrid transactional and analytical processing

Song,

Zhou,

Cui

et al. 2024

The VLDB Journal

View full text Add to dashboard Cite

To provide applications with the ability to analyze fresh data and eliminate the time-consuming ETL workflow, hybrid transactional and analytical (HTAP) systems have been developed to serve online transaction processing and online analytical processing workloads in a single system. In recent years, HTAP systems have attracted considerable interest from both academia and industry. Several new architectures and technologies have been proposed. This paper provides a comprehensive overview of these HTAP systems. We review recently published papers and technical reports in this field and broadly classify existing HTAP systems into two categories based on their data formats: monolithic and hybrid HTAP. We further classify hybrid HTAP into four sub-categories based on their storage architecture: row-oriented, column-oriented, separated, and hybrid. Based on such a taxonomy, we outline each stream’s design challenges and performance issues (e.g., the contradictory format demand for monolithic HTAP). We then discuss potential solutions and their trade-offs by reviewing noteworthy research findings. Finally, we summarize emerging HTAP applications, benchmarks, future trends, and open problems.

show abstract