# Agent Quickstart, DataFusion 53 & Dashboard Updates | Greptime Biweekly Report - No. 83

Source: https://greptime.com/blogs/2026-05-17-greptimedb-biweekly-report

## Summary

*Development period: 2026-05-04 - 2026-05-17*

Here are the highlights from recent commits:

- Queries with constant expressions run faster due to pre-casting optimizations
- DataFusion engine upgraded to version 53 with latest query execution improvements
- Storage engine removes `PartitionTreeMemtable` component in favor of simplified architecture
- Documentation refreshed with a large batch of stale-content fixes

Our [website](https://greptime.com/product/db) and [docs](https://docs.greptime.com/getting-started/overview/) now ship an Agent Quickstart prompt — one click and your AI agent knows how to install and use GreptimeDB. Give it a try!

![Agent Quickstart](/blogs/2026-05-17-greptimedb-biweekly-report/agent-quickstart.png)

We encourage users on older versions to upgrade for these fixes and improvements.

## Contributors

Over the past two weeks, **15 contributors** merged a total of **46 PRs**. Among them, **4 community contributors** contributed **5 PRs**.
Welcome to our new contributors: **[@kimjune01](https://github.com/kimjune01)**, **[@Detachm](https://github.com/Detachm)**!

Thanks to our community contributors:

- **[@QuakeWang](https://github.com/QuakeWang)** *[db#8094](https://github.com/GreptimeTeam/greptimedb/pull/8094)* *[db#8060](https://github.com/GreptimeTeam/greptimedb/pull/8060)*
- **[@BootstrapperSBL](https://github.com/BootstrapperSBL)** *[db#8018](https://github.com/GreptimeTeam/greptimedb/pull/8018)*
- **[@Detachm](https://github.com/Detachm)** *[db#8067](https://github.com/GreptimeTeam/greptimedb/pull/8067)*

## Highlights of Recent PRs

### [db#8098](https://github.com/GreptimeTeam/greptimedb/pull/8098), [db#8106](https://github.com/GreptimeTeam/greptimedb/pull/8106) chore: update dashboards

The [pre-defined Grafana dashboards](https://github.com/GreptimeTeam/greptimedb/tree/main/grafana) add two new sections. The `Hotspot` section provides region distribution and datanode load analysis panels, letting operators diagnose data skew and write hotspots directly without custom queries. The `Autopilot` section displays metrics for region balancer operations, auto-repartitioning actions, and P99 latency. Datanode labels in region statistics are also simplified from the full address format to `datanode-<peer_id>`.

### [db#8080](https://github.com/GreptimeTeam/greptimedb/pull/8080) refactor(mito2)!: remove PartitionTreeMemtable

The mito2 storage engine removes the `PartitionTreeMemtable` implementation as part of internal memory table refactoring. This change simplifies the memtable architecture without affecting query behavior or data compatibility.

### [db#7926](https://github.com/GreptimeTeam/greptimedb/pull/7926) feat: pre-cast constants

Query execution is faster for filters with constants that don't match column types, as GreptimeDB now pre-casts the constants at planning time instead of casting entire columns during execution. This change also enables filter pushdown in cases that were previously blocked by type mismatches.

### [db#8107](https://github.com/GreptimeTeam/greptimedb/pull/8107) feat: bump datafusion to 53

GreptimeDB upgrades to DataFusion 53.1, Arrow 58.3, and Parquet 58.3, bringing the latest query engine improvements and file format optimizations to the database core.


## Good First Issue

### [Issue#7755](https://github.com/GreptimeTeam/greptimedb/issues/7755) Feature request: Support EKS Pod Identity / ECS container credentials endpoint for S3 object storage

Update GreptimeDB's OpenDAL dependency to support AWS container credentials endpoint for EKS Pod Identity and ECS task roles authentication.

* Keywords: Storage, Dependencies
* Difficulty: Easy

### [Issue#7987](https://github.com/GreptimeTeam/greptimedb/issues/7987) feat: add flow_statistics system table and SHOW FLOW STATUS for flow runtime observability

Add a `flow_statistics` system table and `SHOW FLOW STATUS` SQL command to display flow runtime data like start time, uptime, errors, and data processing statistics.

* Keywords: SQL, Observability
* Difficulty: Medium
