Explore the Newly Implemented DROP DATABASE Capability | Greptime Biweekly Report

Summary ​

Together with our global community of contributors, GreptimeDB continues to evolve and flourish as a growing open-source project. We are grateful to each and every one of you.

Below are the highlights among recent commits:

• DROP DATABASE feature is now basically available, which provides a shortcut for deleting a database instance and releasing all underlying resources.

• Continuous polishing the new PartitionTreeMemtable, adding basic support for append mode to reduce the overhead when data deduplication is not required.

• Recent updates on Metric Engine provide the user interface based on SQL to access the physical tables.

Contributors ​

For the past two weeks, our community has been super active with a total of 54 PRs merged. 6 PRs from 5 individual contributors merged successfully and lots pending to be merged.

Congrats on becoming our most active contributors in the past 2 weeks:

• @dimbtp (#3582)

• @JetSquirrel (go#20)

• @xifyang (#2211)

• @xxxuuu (#3571 #3568)

• @YCCDSZXH (#3539)

👏 Welcome contributor @xifyang @xxxuuu @JetSquirrel @YCCDSZXH join to the community as the new individual contributor, and congratulations on successfully merging their first PR, more PRs are waiting to be merged.

A big THANK YOU to all our members and contributors! It is people like you who are making GreptimeDB a great product. Let's build an even greater community together.

Highlights of Recent PRs ​

#3494 Support Gzip-compressed data in InfluxDB line protocol HTTP API ​

This PR supports Gzip-compressed data ingestion by adding a decompress layer in axum routes.

#3501 Support partitioning data in memtable by time range ​

• By strictly partitioning data in memtable according to their timestamps, GreptimeDB can now ensure the SST files flushed to disk never contain overlapping time ranges hence boost the performance for ad-hoc queries with precise time range.

• Note: this feature is applicable to those tables created with compaction.twcs.time_window option.

#3527 Add maintenance mode ​

This PR adds maintenance mode for metasrv so that metasrv can now tolerate temporary downtime of datanodes and will not trigger region failover when, for example, a scheduled rolling update happens.

#3524 Support customize memtable options when creating tables ​

• With the introduction of new memtable, GreptimeDB now exposes more memtable options when creating tables to adapt to different workloads.

• Please also note that the new MergeTreeMemtable has now been renamed as PartitionTreeMemtable in #3547.

#3541 #3534 Basic implementation of DROP DATABASE statement ​

• DROP DATABASE is a shortcut for deleting all tables and resources in a database instance.

• DROP DATABASE is an integral part of GreptimeDB's feature roadmap, accelerated by user feedback. We warmly invite you to use GreptimeDB and share your valuable insights. Your feedback could hasten the implementation of our next feature.

#3508 Shared in-memory state store for Greptime Flow ​

• Some dataflow operators need to share state with other operators, so this PR provides a basic shared state store based on BTreeMap.

• For more updates of Greptime Flow, please check #3187.

Good First Issue ​

#3511 Support specifying time ranges in the COPY FROM statement to avoid importing unwanted data ​

Keyword: backup, restore, filtering, timestamp

Difficulty: medium

The COPY FROM statement is often used to restore data from backup files. If it can support filtering a specified time range, it can avoid importing unnecessary data and speed up the data recovery process.

#3477 Support converting UNIX epoch numbers to specified timezone in to_timezone function ​

Keyword: timezone, UNIX epoch, timestamp

Difficulty: easy

The to_timezone function is used to convert a timestamp to another specified time zone's timestamp. Currently, the to_timezone function only accepts timestamp string representation, such as "2024-03-27T11:01:50+08:00". This issue requires enabling to_timezone to support UNIX millisecond timestamp numerical values, like 1711508510000.