Key Differences in Transaction Processing Flow Between TiDB Distributed Transactions and Traditional Database Transactions (e.g., Oracle): Is There a Significant Efficiency Gap?

Translated
1

Note:
This topic has been translated from a Chinese forum by GPT and might contain errors.

Original topic: TiDB分布式事务 vs 传统数据库事务（例如oracle）整个事务处理流程中主要区别点在哪里？效率上相差大吗？

| username: alfred

To improve efficiency, please provide the following information. Clear problem descriptions can be resolved faster:
[TiDB Usage Environment]

[Overview] Scenario + Problem Overview

Generally, traditional data transactions commit by first flushing the WAL log, and the memory simultaneously stores the inserted or modified data, such as Oracle’s database buffer cache. TiDB also first flushes the WAL log but requires the Leader’s raft log to replicate logs between nodes through the raft protocol. After the majority of followers in the cluster respond successfully, the log is then applied to the KV region, i.e., data is generated. Roughly speaking, there are three additional steps. How does this compare in terms of efficiency impact?
[Background] Operations performed

[Phenomenon] Business and database phenomena

[Problem] Current issues encountered

[Business Impact]

[TiDB Version]

[Application Software and Version]

[Attachments] Relevant logs and configuration information

  • TiUP Cluster Display information
  • TiUP Cluster Edit config information

Monitoring (https://metricstool.pingcap.com/)

  • TiDB-Overview Grafana monitoring
  • TiDB Grafana monitoring
  • TiKV Grafana monitoring
  • PD Grafana monitoring
  • Corresponding module logs (including logs one hour before and after the issue)

If the question is related to performance optimization or troubleshooting, please download the script and run it. Please select all and copy-paste the terminal output results for upload.

2
| username: ddhe9527 | Original post link

It’s unfair to compare TiDB’s distributed transactions across multiple nodes with Oracle’s single-node transactions. Obviously, single-node transactions have shorter paths and higher efficiency.

3
| username: alfred | Original post link

Distributed systems are obviously more advanced technology. Based on practical experience, I would like to learn more about some of the gaps.

4
| username: 胡杨树旁 | Original post link

In a distributed system, network latency needs to be considered, and the requirements for the network are relatively high, right?

5
| username: alfred | Original post link

It seems that it’s not just the network requirements that are high.

6
| username: system | Original post link

This topic will be automatically closed 60 days after the last reply. No new replies are allowed.