As the summer tiptoed in, we were cheerful about everything it brings to us: sunshine, ice cream, and a busy but fulfilling May!
Let’s take a look at what we have accomplished.
News
The first CNCF ambassador spotlight goes to Queeny Jin, who has been spreading the word about TiKV to English speakers since its inception in 2016. We appreciate Queeny’s contribution and “together we can go further!” The full blog is available here.
2 GSoC proposals and 1 Community Bridge project idea from TiKV project were accepted. Congratulations to accepted mentees and we look forward to working with them this summer.
- GSoC projects
- Cloud-Native KV-service
- Mentor: Wei Liu (@Little-Wallace)
- Student: Devdutt Shenoi (@de-sh)
- Versioned rawKV
- Mentor: Fred Chen (@fredchenbj) & Yi Wu (@yiwu-arbug)
- Student: Hyungsuk Kang (@hskang9)
- Cloud-Native KV-service
- Community bridge project
- Full Chunk-based Computing
- Mentor: Wish Shi (@breeswish) & Tianyi Zhuang (@TennyZhuang)
- Mentee: Chi Zhang (@skyzh)
- Full Chunk-based Computing
KubeCon + CloudNativeCon Europe 2020 is now an online experience taking place on Aug. 17- 20 and TiKV speaking sessions on KubeCon EU are scheduled on Aug. 19-20. More information about TiKV speaking sessions is available here.
The TiKV community meeting time was moved to 07.00 p.m. PST (Time zone converter) on every 4th Wednesday of every month to involve more people in our monthly discussion.
TiKV logo was updated with “Ti” added before KV.
Releases
This month our team made 2 releases!
You can review the changelogs here:
- 3.0.14
- New features
- Improve the performance when many write conflicts and
BatchRollbackcondition occur in optimistic transactions in #7650
- Improve the performance when many write conflicts and
- Bug fixes
- New features
- 4.0.0-rc.2
- Compatibility Changes
- Move the encryption-related configuration to the security-related configuration in #7810
- New features
- Support encryption debugging for tikv-ctl in #7698
- Support encrypting the lock column family in snapshots in #7712
- Support heatmap in the Grafana dashboard for Raftstore latency summary to better diagnose the jitter issue in #7717
- Support setting the upper limit for the size of the gRPC message in #7824
- Add in Grafana dashboard the encryption-related monitoring metrics in #7827
- Support Application-Layer Protocol Negotiation (ALPN) in #7825
- Support using the task ID provided by the client as the identifier in the unified read pool in #7814
- Improve the performance of the batch insert request in #7718
- Bug fixes
- Compatibility Changes
Reading materials
In Implement Raft in Rust, @siddontang shares the design of raft-rs, a Raft implementation in Rust designed by TiKV community, and how to develop your own app using it.
In TiKV in JD Cloud & AI, Can Cui, an Infrastructure Specialist at JD Cloud & AI, introduces how TiKV empowered JD Cloud & AI to manage huge amounts of OSS metadata with a simple and horizontally scalable architecture.
@nick_r_cameron wrote Building, Running, and Benchmarking TiKV and TiDB to introduce how to build and run your own TiDB or TiKV, and how to run some benchmarks on those databases.
Notable PRs
- @Little-Wallace created #6683 to support batching multiple write requests into an entry.
- @nrc opened #7676 to enable TiKV to split a region automatically if it contains many locks, which can improve the performance of the large transaction if it continues writing to a single region.
- @hicqu found that index-read requests can be out-of-order during configuration change, and fixed it in #363.
Notable issues
Call for participation
@cofyc proposed to add --advertise-status-addr flag to specify the status address to advertise in #7920. Many comments under this issue show support to this proposal. What do you think?
@zhangjinpeng1987 suggested using RocksDB range deletion to truncate a continuously huge range of data in #7803. @siddontang and @ajkr showed some concerns over this suggestion. What’s your opinion?
New Contributors
We’d like to welcome the following new contributors to TiKV and thank them for their work!
If you’d like to get involved, we’d love to help you get started. You might be interested in tackling one of these issues. If you don’t know how to begin, please leave a comment and somebody will help you out. We’re also very keen for people to contribute to documentation, tests, optimizations, benchmarks, refactoring, or other useful things.