apache / datafusion-ballista

Apache DataFusion Ballista Distributed Query Engine

https://datafusion.apache.org/ballista

apache/datafusion-ballista Issues

[Problem] How to deploy multiple schedulers on standalone mode but not docker
Updated a year ago17
Data quality framework
Updated 9 months ago1
Allow configuring statistics collection during scan
Closed a year ago
The decoded execution plan at the executor side cannot be directly shared.
Closed a year ago
Remove redundant session_registry
Closed a year ago
Refine the ExecuteQuery grpc interface
Closed a year ago
Remove the prometheus-metrics from the default feature
Closed a year ago
The last_seen_ts_threshold for getting alive executor at the scheduler side should be larger than the heartbeat time interval
Closed a year ago
Disable the ansi feature of tracing-subscriber
Closed a year ago
Refine create_datafusion_context()
Closed a year ago
Remove output_partitioning for task definition
Closed a year ago
Refine the interface of QueryStageExecutor
Closed a year ago
Error, message length too large: found 7666438 bytes, the limit is: 4194304 bytes
Closed 7 months ago7
Bug with initialising ClusterStorageConfig
Closed a year ago
Fix plan metrics length and stage metrics length not match
Closed a year ago
[Improvement] refactor the offer_reservation avoid wait result back
Closed a year ago
Critical bug for flight sql on version 0.11.0
Closed a year ago2
[fea] Avoid multithreaded write lock conflicts in event queue.
Closed a year ago
Upgrade DataFusion to 23.0.0
Closed a year ago
Save session data
Updated a year ago
[LOG] Print long running task status.
Closed a year ago
[Improvement] Increase the throughput in push-staged TaskSchedulingPolicy
Updated a year ago2
Ballista should be able to load Partitioned data frame using ListingOptions
Closed 5 months ago
Decode `LaunchMultiTaskParams` execution plan only once
Closed a year ago
[Document] return type not correct
Closed a year ago1
[feature] support shuffle read with retry when facing IO error.
Closed a year ago
[Shuffle] Support cache remote shuffle reader client in executor.
Updated a year ago
Remove redundant fields in ExecutorManager
Closed a year ago
Starting scheduler with docker-compose on `main` throws error: Error: An unknown argument '--config-backend' was specified.
Closed a year ago1
Building Docker images on main branch fails
Closed a year ago4
Submitting same DF to ballista is slow
Updated a year ago1
Change the task assignment philosophy from executor first to task first
Closed a year ago
Fix for error message during testing
Closed a year ago
Improve GitHub Workflows
Updated a year ago
Distributed Execution
Closed a year ago1
Make `task_manager` module pub
Closed a year ago
Make `with_task_launcher` pub
Closed a year ago
Expose SchedulerState as `pub`
Closed a year ago
Support executing query stages in execution engines other than DataFusion
Closed 6 months ago5
Broken build in CI: yarn failing to build
Closed a year ago8
Implement a disk manager with a LRU cache to maintain caches on local disk
Updated a year ago
TPC-H q16 fails to run
Closed 7 months ago2
Make task reset policy configurable
Updated a year ago
Rust / Check Cargo.toml formatting (amd64, stable) (pull_request) Failing
Closed a year ago5
Proposal for more efficient disk-based shuffle mechanism
Updated a year ago6
Issues with lost executor handling
Closed a year ago3
Remove `python` since it has been moved to its own repo, `arrow-ballista-python`
Closed a year ago
Introduce a cache model with LRU properties
Closed a year ago
Add Accumulator API
Updated a year ago2
Add executor self-registration mechanism in the heartbeat service
Closed a year ago