Following up from #26436, this PR adds a distributed benchmark test for Tensorflow FashionMNIST training. It compares training with Ray AIR with training with vanilla PyTorch.
Signed-off-by: Kai Fricke <kai@anyscale.com>
This adds a nightly release test that asserts that autoscaling a cluster up and down in a Ray Tune run works.
Signed-off-by: Kai Fricke <kai@anyscale.com>
This PR adds a distributed benchmark test for Pytorch MNIST training. It compares training with Ray AIR with training with vanilla PyTorch.
In both cases, the same training loop is used. For Ray AIR, we use a TorchTrainer with 4 CPU workers. For vanilla PyTorch, we upload a training script and kick it off (using Ray tasks) in subprocesses on each node. In both cases, we collect the end to end runtime.
Signed-off-by: Kai Fricke <kai@anyscale.com>
See #23676 for context. This is another attempt at that as I figured out what's going wrong in `bazel test`. Supersedes #24828.
Now that there are Python 3.10 wheels for Ray 1.13 and this is no longer a blocker for supporting Python 3.10, I still want to make `bazel test //python/ray/tests/...` work for developing in a 3.10 env, and make it easier to add Python 3.10 tests to CI in future.
The change contains three commits with rather descriptive commit message, which I repeat here:
Pass deps to py_test in py_test_module_list
Bazel macro py_test_module_list takes a `deps` argument, but completely
ignores it instead of passes it to `native.py_test`. Fixing that as we
are going to use deps of py_test_module_list in BUILD in later changes.
cpp/BUILD.bazel depends on the broken behaviour: it deps-on a cc_library
from a py_test, which isn't working, see upstream issue:
https://github.com/bazelbuild/bazel/issues/701.
This is fixed by simply removing the (non-working) deps.
Depend on conftest and data files in Python tests BUILD files
Bazel requires that all the files used in a test run should be
represented in the transitive dependencies specified for the test
target. For py_test, it means srcs, deps and data.
Bazel enforces this constraint by creating a "runfiles" directory,
symbolic links files in the dependency closure and run the test in the
"runfiles" directory, so that the test shouldn't see files not in the
dependency graph.
Unfortunately, the constraint does not apply for a large number of
Python tests, due to pytest (>=3.9.0, <6.0) resolving these symbolic
links during test collection and effectively "breaks out" of the
runfiles tree.
pytest >= 6.0 introduces a breaking change and removed the symbolic link
resolving behaviour, see pytest pull request
https://github.com/pytest-dev/pytest/pull/6523 for more context.
Currently, we are underspecifying dependencies in a lot of BUILD files
and thus blocking us from updating to newer pytest (for Python 3.10
support). This change hopefully fixes all of them, and at least those in
CI, by adding data or source dependencies (mostly for conftest.py-s)
where needed.
Bump pytest version from 5.4.3 to 7.0.1
We want at least pytest 6.2.5 for Python 3.10 support, but not past
7.1.0 since it drops Python 3.6 support (which Ray still supports), thus
the version constraint is set to <7.1.
Updating pytest, combined with earlier BUILD fixes, changed the ground
truth of a few error message based unit test, these tests are updated to
reflect the change.
There are also two small drive-by changes for making test_traceback and
test_cli pass under Python 3.10. These are discovered while debugging CI
failures (on earlier Python) with a Python 3.10 install locally. Expect
more such issues when adding Python 3.10 to CI.
The old user-facing TrialCheckpoint class has been deprecated in favor of `ray.ml.Checkpoint` and will be removed with this PR.
The main change in this PR is to delete the old `TrialCheckpoint` class and replace remaining API calls (e.g. `checkpoint.local_path`) with the correct AIR equivalents.
One issue that comes up is that with Ray client usage, checkpoint directories are not available on the local node (the client). Thus, we can't construct `Checkpoint` objects easily. (Previously, the TrialCheckpoint object held a reference to the location, even if it is not locally available). There are ongoing discussions on how to resolve this in the future. For now, we print an error when such a checkpoint is requested.
Depends on #25805
Signed-off-by: Kai Fricke <kai@anyscale.com>
Uses the new AIR Train API for examples and tests.
The `Result` object gets a new attribute - `log_dir`, pointing to the Trial's `logdir` allowing users to access tensorboard logs and artifacts of other loggers.
This PR only deals with "low hanging fruit" - tests that need substantial rewriting or Train user guide are not touched. Those will be updated in followup PRs.
Tests and examples that concern deprecated features or which are duplicated in AIR have been removed or disabled.
Requires https://github.com/ray-project/ray/pull/25943 to be merged in first
Revert back to using nightly base images instead of pinning to 1.12.1. Pinning the docker image had led to uncaught errors in the past. Instead, we should be using nightly to make sure release tests will work on the most up to date versions of docker/cluster envs. If there are any test failures, the underlying issues should be fixed rather than pinning the docker image.
Co-authored-by: Kai Fricke <kai@anyscale.com>
Adds a CI test for 100TB shuffle.
There is a custom config for this nightly test to: (1) make sure each node gets 4TB of storage, (2) head node has 0 CPUs, (3) worker nodes have half their actual vCPU count.
Related issue number
Closes#24480.
Fixes a bug in wait_cluster where we count the total number of nodes ever in the cluster rather than the alive nodes. This has causes infra/autoscaler failures (e.g. #26138) to be mislabeled as test failures (and probably messes with timing too).
Co-authored-by: Alex Wu <alex@anyscale.com>
This adds "environments" to the release package that can be used to configure some environment variables. These variables will be loaded either by an `--env` argument or a `env` definition in the test definition and can be used to e.g. run release tests on staging.
There are mysterious memory usage growth in Ray clusters that disappear when running with jemalloc. Before we are able to figure out the root cause, it seems using jemalloc by default can be a good walkaround. Because of its efficiency, using jemalloc by default can be beneficial, but we need to run more benchmarks to verify.
Enable checking of the ray core module, excluding serve, workflows, and tune, in ./ci/lint/check_api_annotations.py. This required moving many files to ray._private and associated fixes.
The error message in #25638 indicates we should use protobuf>3.19.0 to generated code so that we can work with python protobuf >= 4.21.1. Try generating wheels to see if this works.
Error message suggests:
Wait timeout after 30 seconds for key(s): 0. You may want to increase the timeout via HOROVOD_GLOO_TIMEOUT_SECONDS
Bumped up to 120 seconds.
Tests run successfully: https://buildkite.com/ray-project/release-tests-pr/builds/6906
m5.16xlarge instances have 64 CPU and 256GB memory, which are overkill for scheduling tests that do not have a lot of computations. Use smaller instance m5.4xlarge to save cost and make allocating instances easier.
