Can TiKV support user-defined column families in transaction mode?

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

Original topic: TiKV transaction模式下能否支持用户自定义column family

| username: jwongz

[TiDB Usage Environment] Production Environment / Testing / PoC
[TiDB Version]
[Reproduction Path] What operations were performed that led to the issue
[Encountered Issue: Problem Phenomenon and Impact]

Only using the TiKV component, with approximately 3.5 million rows of data, among which 3 rows have special prefixes.

These 3 rows take 25 seconds to scan, while scanning the entire 3.5 million rows takes only 180 seconds.

  1. Is there a way to quickly read these 3 rows of data? Scanning is obviously too slow. Is there a way to optimize the scan?

  2. Is there a way to store these data independently, similar to a custom column family?

  3. Can we write to PD through the etcd client?

[Resource Configuration] Go to TiDB Dashboard - Cluster Info - Hosts and take a screenshot of this page
[Attachments: Screenshots / Logs / Monitoring]

| username: jwongz | Original post link

By adding a termination condition to optimize performance.

| username: dba远航 | Original post link


| username: 春风十里 | Original post link

I didn’t understand this. How was it optimized? Can someone explain it in detail?

| username: 哈喽沃德 | Original post link

In TiKV’s transaction mode, user-defined column families are currently not supported.

| username: linnana | Original post link

Are there any specific usage examples?

| username: system | Original post link

This topic was automatically closed 60 days after the last reply. New replies are no longer allowed.