feat: Add runner for local distributed execution #11609
Closed
Conversation
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
facebook-github-bot
added
the
CLA Signed
✅ Deploy Preview for meta-velox canceled.
|
|
@oerling has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
|
@oerling has imported this pull request. If you are a Meta employee, you can view this diff on Phabricator. |
…1609) Summary: Adds LocalRunner and LocalSchema for running distributed Velox plans in-process with multiple fragments and multiple tasks per fragment. This introduces the abstractions for distributed plans, their options, schema and split sources. Adds an extension of PlanBuilder for building plan trees with shuffles. These get partitioned into multiple plan trees for execution with LocalRunner. Adds LocalRunnerTestBase, a test base class with utilities for generating multiple tables each with multiple files of test data. These are then returned as a LocalSchema that can be used to produce splits for LocalRunner. Adds error propagation for TaskCursor since distributed plans may have to produce errors from any fragment that are consumed by the application via TaskCursor. This will be generalized to distributed execution on a cluster with Velox coordination and shuffles. Differential Revision: D66261119 Pulled By: oerling
|
This pull request was exported from Phabricator. Differential Revision: D66261119 |
assignUser
changed the title
assignUser
approved these changes
Nov 21, 2024
|
Conbench analyzed the 1 benchmark run on commit There were no benchmark performance regressions. 🎉 The full Conbench report has more details. |
facebook-github-bot
pushed a commit
that referenced
this pull request
Nov 29, 2024
Summary: A minor CMake fix for #11609, to address the linking issue when `VELOX_BUILD_TESTING` is off. ```C++ 20:52:36 #10 2.268 CMake Error at velox/velox/runner/tests/CMakeLists.txt:19 (target_link_libraries): 20:52:36 #10 2.268 Target "velox_local_runner_test" links to: 20:52:36 #10 2.268 20:52:36 #10 2.268 GTest::gtest 20:52:36 #10 2.268 20:52:36 #10 2.268 but the target was not found. Possible reasons include: 20:52:36 #10 2.268 20:52:36 #10 2.268 * There is a typo in the target name. 20:52:36 #10 2.268 * A find_package call is missing for an IMPORTED target. 20:52:36 #10 2.268 * An ALIAS target is missing. 20:52:36 #10 2.268 20:52:36 #10 2.268 20:52:36 #10 2.268 20:52:36 #10 2.273 CMake Generate step failed. Build files cannot be regenerated correctly. 20:52:36 #10 2.302 make: *** [Makefile:96: cmake-and-build] Error 1 20:52:36 #10 2.302 make: Leaving directory '/prestissimo' 20:52:36 #10 ERROR: process "/bin/sh -c EXTRA_CMAKE_FLAGS=${EXTRA_CMAKE_FLAGS} NUM_THREADS=${NUM_THREADS} make --directory=\"/prestissimo/\" cmake-and-build BUILD_TYPE=${BUILD_TYPE} BUILD_DIR=${BUILD_DIR} BUILD_BASE_DIR=${BUILD_BASE_DIR}" did not complete successfully: exit code: 2 ``` Pull Request resolved: #11669 Reviewed By: zacw7 Differential Revision: D66560342 Pulled By: xiaoxmeng fbshipit-source-id: 9f6b74c5a64f6e248cb8d86ec213ae76eb894770
TongWei1105
pushed a commit
to TongWei1105/velox
that referenced
this pull request
Dec 3, 2024
…kincubator#11669) Summary: A minor CMake fix for facebookincubator#11609, to address the linking issue when `VELOX_BUILD_TESTING` is off. ```C++ 20:52:36 facebookincubator#10 2.268 CMake Error at velox/velox/runner/tests/CMakeLists.txt:19 (target_link_libraries): 20:52:36 facebookincubator#10 2.268 Target "velox_local_runner_test" links to: 20:52:36 facebookincubator#10 2.268 20:52:36 facebookincubator#10 2.268 GTest::gtest 20:52:36 facebookincubator#10 2.268 20:52:36 facebookincubator#10 2.268 but the target was not found. Possible reasons include: 20:52:36 facebookincubator#10 2.268 20:52:36 facebookincubator#10 2.268 * There is a typo in the target name. 20:52:36 facebookincubator#10 2.268 * A find_package call is missing for an IMPORTED target. 20:52:36 facebookincubator#10 2.268 * An ALIAS target is missing. 20:52:36 facebookincubator#10 2.268 20:52:36 facebookincubator#10 2.268 20:52:36 facebookincubator#10 2.268 20:52:36 facebookincubator#10 2.273 CMake Generate step failed. Build files cannot be regenerated correctly. 20:52:36 facebookincubator#10 2.302 make: *** [Makefile:96: cmake-and-build] Error 1 20:52:36 facebookincubator#10 2.302 make: Leaving directory '/prestissimo' 20:52:36 facebookincubator#10 ERROR: process "/bin/sh -c EXTRA_CMAKE_FLAGS=${EXTRA_CMAKE_FLAGS} NUM_THREADS=${NUM_THREADS} make --directory=\"/prestissimo/\" cmake-and-build BUILD_TYPE=${BUILD_TYPE} BUILD_DIR=${BUILD_DIR} BUILD_BASE_DIR=${BUILD_BASE_DIR}" did not complete successfully: exit code: 2 ``` Pull Request resolved: facebookincubator#11669 Reviewed By: zacw7 Differential Revision: D66560342 Pulled By: xiaoxmeng fbshipit-source-id: 9f6b74c5a64f6e248cb8d86ec213ae76eb894770
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Adds LocalRunner and LocalSchema for running distributed Velox plans in-process with multiple fragments and multiple tasks per fragment. This introduces the abstractions for distributed plans, their options, schema and split sources.
Adds an extension of PlanBuilder for building plan trees with shuffles. These get partitioned into multiple plan trees for execution with LocalRunner.
Adds LocalRunnerTestBase, a test base class with utilities for generating multiple tables each with multiple files of test data. These are then returned as a LocalSchema that can be used to produce splits for LocalRunner.
Adds error propagation for TaskCursor since distributed plans may have to produce errors from any fragment that are consumed by the application via TaskCursor.
This will be generalized to distributed execution on a cluster with Velox coordination and shuffles.