dependabot[bot] opened a new pull request, #3705: URL: https://github.com/apache/iceberg-python/pull/3705
Bumps [ray](https://github.com/ray-project/ray) from 2.55.1 to 2.56.1. <details> <summary>Release notes</summary> <p><em>Sourced from <a href="https://github.com/ray-project/ray/releases";>ray's releases</a>.</em></p> <blockquote> <h2>Ray-2.56.1</h2> <h1>Highlights</h1> <ul> <li><strong>Ray Data</strong>: We added fixes for several <code>to_pandas</code> regressions introduced in 2.56: an opt-out flag (<code>RAY_DATA_ENABLE_ARROW_BACKED_PANDAS_CONVERSION</code>) for Arrow-backed conversion, an int64/<code>double[pyarrow]</code> overflow crash on concatenation, and a <code>TensorDtype.__from_arrow__</code> crash on empty tensor columns (<a href="https://redirect.github.com/ray-project/ray/issues/64793";>#64793</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64794";>#64794</a>).</li> <li><strong>Ray Core</strong>: We added early detection for system-slice memory pressure: the memory monitor now snapshots the user and system cgroup slices together and logs an error when the system slice exceeds reserved system memory, warning users to raise <code>--system-reserved-memory</code> before it causes node deaths (<a href="https://redirect.github.com/ray-project/ray/issues/64492";>#64492</a>).</li> <li><strong>Ray Serve</strong>: We added protobuf 7 compatibility and a routing fix for LLM direct streaming, so body-aware routers like <code>PrefixCacheAffinityRouter</code> no longer hang when <code>RAY_SERVE_LLM_ENABLE_DIRECT_STREAMING=1</code> (<a href="https://redirect.github.com/ray-project/ray/issues/64592";>#64592</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64488";>#64488</a>).</li> </ul> <h1>Ray Data</h1> <h2>🔨 Fixes</h2> <ul> <li>Fixed two Arrow-backed <code>to_pandas</code> regressions: added <code>DataContext.enable_arrow_backed_pandas_conversion</code> as an opt-out, and reconciled divergent numeric column types before concatenation to avoid int64/<code>double[pyarrow]</code> overflow crashes (<a href="https://redirect.github.com/ray-project/ray/issues/64793";>#64793</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64768";>#64768</a>).</li> <li>Fixed a <code>TensorDtype.__from_arrow__</code> crash on zero-size tensor elements by using an explicit row count instead of numpy's <code>-1</code> dimension inference (<a href="https://redirect.github.com/ray-project/ray/issues/64794";>#64794</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64767";>#64767</a>).</li> <li>Fixed a crash in hash partition caused by read-only hash arrays (<a href="https://redirect.github.com/ray-project/ray/issues/64584";>#64584</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64552";>#64552</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64559";>#64559</a>).</li> <li>Nullified <code>_input_dependencies</code> in <code>_get_args</code> so exporting operator args no longer triggers an exponential <code>sanitize_for_struct</code> call chain over fused operators (<a href="https://redirect.github.com/ray-project/ray/issues/64412";>#64412</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64316";>#64316</a>).</li> </ul> <h1>Ray Serve</h1> <h2>🔨 Fixes</h2> <ul> <li>Added protobuf <code>>=7</code> compatibility to <code>_proto_to_dict</code> by binding to <code>FieldDescriptor.is_repeated</code> when the deprecated <code>label</code> attribute is absent (<a href="https://redirect.github.com/ray-project/ray/issues/64592";>#64592</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64362";>#64362</a>).</li> </ul> <h1>Ray LLM</h1> <h2>🔨 Fixes</h2> <ul> <li>Fixed direct-streaming routing for body-aware routers: the ingress now parses the raw request body into a <code>SimpleNamespace</code> over routing-key fields (<code>messages</code>, <code>prompt</code>) so <code>choose_replica</code> receives the message body instead of raw bytes (<a href="https://redirect.github.com/ray-project/ray/issues/64488";>#64488</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64328";>#64328</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64326";>#64326</a>).</li> </ul> <h1>Ray RLlib</h1> <h2>🔨 Fixes</h2> <ul> <li>Upgraded the ONNX example from the retired MobileNet v1 to MobileNet v3 via <code>torchvision</code> and pinned <code>onnxscript</code> in the GPU/ml-build CI dep locks, fixing ONNX export failures (<a href="https://redirect.github.com/ray-project/ray/issues/64591";>#64591</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64028";>#64028</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64031";>#64031</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64590";>#64590</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64033";>#64033</a>).</li> </ul> <h1>Ray Core</h1> <h2>💫 Enhancements</h2> <ul> <li>The threshold memory monitor now snapshots both the user and system cgroup slices and logs an error when system-slice usage exceeds the reserved system memory, prompting users to raise <code>--system-reserved-memory</code> (<a href="https://redirect.github.com/ray-project/ray/issues/64492";>#64492</a>).</li> </ul> <h2>🔨 Fixes</h2> <ul> <li>Disabled NCCL <code>cuMem</code> host buffer registration in CI pytests to stabilize GPU test runs (<a href="https://redirect.github.com/ray-project/ray/issues/64580";>#64580</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64146";>#64146</a>).</li> </ul> <h1>Documentation</h1> <h2>📖 Documentation</h2> <ul> <li>Repointed the ASHA Tune example links to the renamed <code>README</code> document, fixing the <code>fail_on_warning</code> ReadTheDocs build on the release line (<a href="https://redirect.github.com/ray-project/ray/issues/64761";>#64761</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64630";>#64630</a>).</li> <li>Updated the Python 3.10 CPU <code>pip freeze</code> dependency list for the Ray 2.56.0 release (<a href="https://redirect.github.com/ray-project/ray/issues/64447";>#64447</a>, <a href="https://redirect.github.com/ray-project/ray/issues/64357";>#64357</a>).</li> </ul> <h2>Ray-2.56.0</h2> <h1>Highlights</h1> <ul> <li><strong>Ray Data Stability:</strong> In this Ray release, we've added a variety of stability improvements, including running multiple datasets in a cluster, adding automatic batch size selection to CPU-based map-batches, and default logical memory configuration to prevent OOMs. We've also tightened <code>iter_batches</code> stability by reducing hidden buffering and shutting down the executor when consumers exit early (<a href="https://redirect.github.com/ray-project/ray/issues/63660";>#63660</a>, <a href="https://redirect.github.com/ray-project/ray/issues/63682";>#63682</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62949";>#62949</a>). This reduces object-store spilling for common training workloads</li> <li><strong>Ray Serve:</strong> We re-architected Ray Serve LLM by decoupling request handling from token streaming response path (<a href="https://redirect.github.com/ray-project/ray/issues/62667";>#62667</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62680";>#62680</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62668";>#62668</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62669";>#62669</a>, <a href="https://redirect.github.com/ray-project/ray/issues/63167";>#63167</a>), resulting in significant LLM serving performance improvements. We've also introduced new routing policies such as session-sticky routing via consistent hashing with <code>ConsistentHashRouter</code> (<a href="https://redirect.github.com/ray-project/ray/issues/62905";>#62905</a>, <a href="https://redirect.github.com/ray-project/ray/issues/63096";>#63096</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62906";>#62906</a>) and <code>CapacityQueueRouter</ code> (<a href="https://redirect.github.com/ray-project/ray/issues/62323";>#62323</a>) which is beneficial for supply-constrained workloads.</li> <li><strong>Ray Core:</strong> We've added GPU-domain-aware placement groups using label locality (<a href="https://redirect.github.com/ray-project/ray/issues/61442";>#61442</a>, <a href="https://redirect.github.com/ray-project/ray/issues/61614";>#61614</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62487";>#62487</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62533";>#62533</a>). This enables placement groups to pack bundles onto nodes that share a <code>ray.io/gpu-domain</code> label instead of only packing at the single-node level. We've also added initial Kubernetes in-place pod resizing support for Autoscaler v2 (<a href="https://redirect.github.com/ray-project/ray/issues/55961";>#55961</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62369";>#62369</a>, <a href="https://redirect.github.com/ray-project/ray/issues/62215";>#62215</a>), enabling Ray clusters to resize CPU and memory on existing worker pods before scaling out new pods. </li> </ul> <!-- raw HTML omitted --> </blockquote> <p>... (truncated)</p> </details> <details> <summary>Commits</summary> <ul> <li><a href="https://github.com/ray-project/ray/commit/936f0d7d49d9da8ac1a9f04cc8a89faf2cb3c42a";><code>936f0d7</code></a> [docker] Update latest Docker dependencies for 2.56.1 release (<a href="https://redirect.github.com/ray-project/ray/issues/64849";>#64849</a>)</li> <li><a href="https://github.com/ray-project/ray/commit/ea47a0f64e9501a89f92d834ea908c9329bc15bc";><code>ea47a0f</code></a> [Data] Enable memory-safety best practices in video object detection benchmar...</li> <li><a href="https://github.com/ray-project/ray/commit/06729817574615e25afe8f20902b37660c2e6962";><code>0672981</code></a> [Data][2.56.1] Fix Arrow-backed to_pandas regressions: opt-out flag + int/flo...</li> <li><a href="https://github.com/ray-project/ray/commit/6d4f5e67a36077d18c5328a5974de9b7027f07a2";><code>6d4f5e6</code></a> [Data][2.56.1] Fix TensorDtype.<strong>from_arrow</strong> crash on empty tensor columns (...</li> <li><a href="https://github.com/ray-project/ray/commit/e4063cc1311582257a96fe6b6869d300d05b4e19";><code>e4063cc</code></a> [Data] Enable spilling and OOM check for remaining multimodal release tests (...</li> <li><a href="https://github.com/ray-project/ray/commit/d7babce6b09a9f0d7090216403f543eea9f1d465";><code>d7babce</code></a> [Data] Add memory hint for reads to <code>image_embedding_from_jsonl</code> (<a href="https://redirect.github.com/ray-project/ray/issues/64717";>#64717</a>) (ch...</li> <li><a href="https://github.com/ray-project/ray/commit/b747a252fc5e4862005abff343d53a7cebe24a36";><code>b747a25</code></a> [doc] Repoint ASHA Tune example links to README (fixes docs build) (<a href="https://redirect.github.com/ray-project/ray/issues/64630";>#64630</a>) (...</li> <li><a href="https://github.com/ray-project/ray/commit/3ba205c1b0ec84fd038d3c82a8d1a78ef5fbf13a";><code>3ba205c</code></a> [release] Update shipped dependencies for Ray 2.56.1 (<a href="https://redirect.github.com/ray-project/ray/issues/64728";>#64728</a>)</li> <li><a href="https://github.com/ray-project/ray/commit/851cf153844c826ecb7def04e4bd13fff47da05f";><code>851cf15</code></a> [cherry-pick] Cherry-pick 2.56.1 mobilenet 64028 64031 (<a href="https://redirect.github.com/ray-project/ray/issues/64591";>#64591</a>)</li> <li><a href="https://github.com/ray-project/ray/commit/e7092663e0ab562418ebee2ca38302b4d847c793";><code>e709266</code></a> [deps][cherry-pick] cherry-pick 2.56.1 onnxscript <a href="https://redirect.github.com/ray-project/ray/issues/64033";>#64033</a> (<a href="https://redirect.github.com/ray-project/ray/issues/64590";>#64590</a>)</li> <li>Additional commits viewable in <a href="https://github.com/ray-project/ray/compare/ray-2.55.1...ray-2.56.1";>compare view</a></li> </ul> </details> <br /> [](https://docs.github.com/en/github/managing-security-vulnerabilities/about-dependabot-security-updates#about-compatibility-scores) Dependabot will resolve any conflicts with this PR as long as you don't alter it yourself. You can also trigger a rebase manually by commenting `@dependabot rebase`. [//]: # (dependabot-automerge-start) [//]: # (dependabot-automerge-end) --- <details> <summary>Dependabot commands and options</summary> <br /> You can trigger Dependabot actions by commenting on this PR: - `@dependabot rebase` will rebase this PR - `@dependabot recreate` will recreate this PR, overwriting any edits that have been made to it - `@dependabot show <dependency name> ignore conditions` will show all of the ignore conditions of the specified dependency - `@dependabot ignore this major version` will close this PR and stop Dependabot creating any more for this major version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this minor version` will close this PR and stop Dependabot creating any more for this minor version (unless you reopen the PR or upgrade to it yourself) - `@dependabot ignore this dependency` will close this PR and stop Dependabot creating any more for this dependency (unless you reopen the PR or upgrade to it yourself) </details> -- This is an automated message from the Apache Git Service. To respond to the message, please log on to GitHub and use the URL above to go to the specific comment. To unsubscribe, e-mail: [email protected] For queries about this service, please contact Infrastructure at: [email protected] --------------------------------------------------------------------- To unsubscribe, e-mail: [email protected] For additional commands, e-mail: [email protected]
