TiDB Weekly

News & blog post

One of the fundamental tradeoffs of programming language design is run-time performance vs. compile-time performance. The Rust team nearly always (if not always) chose run-time over compile-time.

Last week, we published a post that discusses the following topics as the first episode in a series of posts:

• The spectre of poor Rust compile times at PingCAP
• The TiKV compile-time adventure so far
• Rust's designs for poor compilation time
  • Bootstrapping Rust
  • (Un)virtuous cycles
  • Early decisions that favored run-time over compile-time
• Recent work on Rust compile times

The full post is here:

The Rust Compilation Model Calamity

Community

New contributors

tidb:

• ldnvnbl
• ZiheLiu

tikv:

• silathdiir

parser:

• rubb1sh

Call for participation

• More issues to solve for TiDB
• More issues to solve for TiKV

Updates of the week

Progress

Last two weeks, we landed 25 PRs in the TiDB repository and 7 PRs in the TiKV and PD repositories.

Critical PRs

TiDB:

• Use the 8-byte MysqlTime format
• Handle events of adding or truncating partitions
• Fix the bug that the table's existence is not checked when granting privileges on this table to users
• Enhance the rule of partition pruning
• Fix the issue that the RowKVClient scan method cannot correctly break the loop

TiKV and PD:

• Ensure that the peer with pending reads does not become hibernate
• Support using S3 as the external storage
• Enable the default protocol feature flag
• Switch the Chunk time to use the new 8-byte format