@Greptime: Most of GreptimeDB's June work came down to one idea: a filter is no use if it can't reach the data. In a distributed q…
Summary
GreptimeDB improved distributed query performance by enabling remote dynamic filters to push down to datanode scans at runtime and optimizing the optimizer to run before MergeScan wraps remote plans, ensuring filters reach the data. JSON v2 columns now support type hints.
View Cached Full Text
Cached at: 07/01/26, 04:14 PM
Most of GreptimeDB’s June work came down to one idea: a filter is no use if it can’t reach the data.
In a distributed query, predicates used to get stranded above the scan boundary. Two changes fixed that:
- Remote dynamic filters (from hash joins) now push down to datanode scans at runtime, so nodes prune before the full plan finishes
- The optimizer runs before MergeScan wraps the remote plan, so scan-local filters reach the datanode TableScan for region pruning
Also landed: JSON v2 columns take type hints now.
June report (now monthly):
Similar Articles
@Greptime: GreptimeDB's flat-format queries can now prefilter on any column — tags, fields, timestamps — not just primary keys. Wh…
GreptimeDB's flat-format queries now support prefiltering on any column (tags, fields, timestamps), not just primary keys, delivering up to 4.5x faster performance. Additionally, the mito2 storage engine removed its legacy scan path, cleaning up about 1,800 lines of code.
@Greptime: DataFusion shipped dynamic filter pushdown upstream last September. GreptimeDB v1.0 wires it into the Mito scan layer. …
GreptimeDB v1.0 integrates DataFusion's dynamic filter pushdown to accelerate TopK queries by pushing runtime bounds to the scan layer, reducing query time from 29 seconds to 0.21 seconds on a 5B-row trace table.
@Greptime: GreptimeDB v1.1.0 is out PromQL rate/increase up to 97% faster, 20–40% lower query time overall Up to 4.5x faster on TS…
GreptimeDB v1.1.0 is released, offering up to 97% faster PromQL queries, 20-40% lower overall query times, and up to 4.5x improvement on TSBS scan-heavy queries, along with online repartitioning for existing tables.
@Greptime: We refreshed our deep-dive on Mito2, the storage engine inside GreptimeDB. LSM-tree write path. Columnar Parquet SSTs. …
GreptimeDB's Mito2 storage engine uses an LSM-tree design with columnar Parquet SSTs, three-level scan pruning, and TWCS compaction. The blog post provides a full walkthrough of its architecture.
@Greptime: GreptimeDB v1.1: you no longer have to lock in a table's partition layout when you create it. Before, only tables creat…
GreptimeDB v1.1 introduces online repartitioning for existing tables, incremental Flow reads, a semantic layer for LLMs, and stability improvements.