Nikita Taranov

Extend protocol with query parameters

1

### Changelog category (leave one): - Improvement ### Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md): Parameters are now transferred in `Query` packets right after...

### Changelog category (leave one): - Not for changelog (changelog entry is not required) Fixing memory limit doesn't help (

### Changelog category (leave one): - Not for changelog (changelog entry is not required)

Control block size in aggregator

1

### Changelog category (leave one): - Performance Improvement ### Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md): Account `max_block_size` value while producing single-level aggregation results....

Support complex names

1

* input size is the last part of the name * label is everything before the last slash

S3Express support

1

### Changelog category (leave one): - Not for changelog (changelog entry is not required) ### Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md): todo ---...

More parallel insert-select pipeline

1

### Changelog category (leave one): - Not for changelog (changelog entry is not required) ### Changelog entry (a user-readable short description of the changes that goes to CHANGELOG.md): Currently the...

Build full-fledged plan for local replica

3

as we do for distributed queries. i.e. instead of ``` sql ┌─explain────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────────┐ 1. │ Expression (Project names) │ 2. │ ReadFromRemoteParallelReplicas (Replica: 127.0.0.1, Replica: 127.0.0.2, Replica: 127.0.0.3, Replica: 127.0.0.4, Replica:...

Noticeable when there are not so many rows with heavy aggregate function states. Because we create as many AggregatingTransforms as there are input reading streams we deteriorate performance (more granular...

1. need to refactor PrefetchedReadPool * tasks should be created on demand (instead of creating all at once) * we should prefetch next task as soon as another one from...