[None][feat] Add TRTLLM_NIXL_KVCACHE_BACKEND environment variable for NIXL backend selection #9075
Conversation
zackyoray
force-pushed
the
yorayz/nixl_backend
branch
from
62bd5be to
9c6c9ed
Compare
📝 WalkthroughWalkthroughAdds a thread-safe utility to read and cache the TRTLLM_NIXL_KVCACHE_BACKEND environment variable (default "UCX") and updates the NIXL transfer agent to select and validate the backend at runtime, replacing previously hardcoded UCX usage. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
actor Caller
participant TransferAgent
participant EnvUtils
participant NixlBackend as "NIXL Backend Factory"
Caller->>TransferAgent: initiate transfer setup
TransferAgent->>EnvUtils: getEnvNixlBackend()
note right of EnvUtils `#DDDDFF`: thread-safe\nlazy init (once_flag)
EnvUtils-->>TransferAgent: "UCX" (or env value)
TransferAgent->>TransferAgent: validate backend ∈ supportedBackends
alt valid backend
TransferAgent->>NixlBackend: createBackend(selectedBackend, params)
NixlBackend-->>TransferAgent: backend instance / success
TransferAgent-->>Caller: proceed
else unsupported backend
TransferAgent-->>Caller: log error / fallback or fail
end
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes
Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out. Comment |
zackyoray
changed the title
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
cpp/tensorrt_llm/common/envUtils.cpp(1 hunks)cpp/tensorrt_llm/common/envUtils.h(1 hunks)cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cpp(2 hunks)
🧰 Additional context used
📓 Path-based instructions (7)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh}: Namespace closing braces must include a trailing comment with the namespace name (e.g., '} // namespace foo').
Prefer const or constexpr variables over #define for constants.
Declare variables that are not modified after initialization as const.
Avoid magic literals in code; except for 0, nullptr, true, false. Use named constants for comparisons and logic.
Use Allman brace style for formatting.
Place the semicolon of an empty for/while loop on a new line.
Bodies of switch/while/do-while/for must be compound statements (brace-delimited), and if/else must always be followed by brace-delimited statements.
Type names (e.g., classes) must be CamelCase starting with an uppercase letter (e.g., FooBar).
Local variables, methods, and namespaces use lowerCamelCase (e.g., localFooBar).
Non-magic-number global variables that are non-static and not in an anonymous namespace must be lowerCamelCase prefixed with 'g' (e.g., gDontUseGlobalFoos).
Non-magic-number globals that are static or in an anonymous namespace use lowerCamelCase prefixed with 's' (e.g., sMutableStaticGlobal).
Locally visible static variables use lowerCamelCase with 's' prefix (e.g., static std::once_flag sFlag).
Private/protected member variables use 'm' prefix with CamelCase (e.g., mNbFooValues). Public members may omit, but 'm' is encouraged for clarity.
Constants (enums, global constants, static constants, and function-scope magic/literal constants) use uppercase SNAKE_CASE with 'k' prefix (e.g., kDIGIT_NUM).
Function-scope constants that are not magic numbers or literals are named like non-constant variables (e.g., bool const pass = a && b).
If macros are necessary, name them in UPPER_SNAKE_CASE (e.g., FOO_VERSION) and prefer constants over #define.
Use LLVM clang-format; wrap lines at a maximum of 120 columns; use '// clang-format off/on' sparingly with justification.
Use smart pointers for heap allocations; prefer unique_ptr for sole ownership, shared_ptr for shared...
Files:
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cppcpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.h
**/*.{cpp,cxx,cc,cu,h,hpp,hh,hxx,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
C++ filenames should be lowerCamelCase (first letter lowercase) and must be case-insensitive unique within a compilation target.
Files:
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cppcpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.h
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use only spaces, no tabs; indent with 4 spaces.
Files:
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cppcpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.h
**/*.{h,hpp,hh,hxx,cpp,cxx,cc}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc}: Prefer anonymous namespaces over 'static' for internal linkage of functions.
All templates (class/function/member/static) must be instantiated at least once; non-POD classes should have private data members.
Files:
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cppcpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.h
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).
Files:
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cppcpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.h
**/*.{h,hpp,hh,hxx}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Document new class interfaces and function prototypes with Doxygen; use //! for single-line and //!< for members.
Files:
cpp/tensorrt_llm/common/envUtils.h
**/*.{h,hpp,hh,hxx,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use include guards named 'TRTLLM_<FILE_NAME_IN_CAPS_WITH_UNDERSCORES>_H' (no leading or trailing underscore; directory names excluded).
Files:
cpp/tensorrt_llm/common/envUtils.h
🧬 Code graph analysis (2)
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cpp (1)
cpp/tensorrt_llm/common/envUtils.cpp (2)
getEnvNixlBackend(321-341)getEnvNixlBackend(321-321)
cpp/tensorrt_llm/common/envUtils.h (1)
cpp/tensorrt_llm/common/envUtils.cpp (2)
getEnvNixlBackend(321-341)getEnvNixlBackend(321-321)
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
- GitHub Check: Pre-commit Check
🔇 Additional comments (4)
cpp/tensorrt_llm/common/envUtils.h (1)
91-92: LGTM!The function declaration is well-placed and follows the established pattern for environment variable accessors in this header.
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cpp (2)
31-31: LGTM!The
<set>include is correctly placed and necessary for thesupportedBackendsset.
362-372: LGTM!The runtime-selected backend is correctly used in NIXL API calls, and error handling appropriately includes the backend name in the failure message.
cpp/tensorrt_llm/common/envUtils.cpp (1)
321-342: LGTM!The implementation correctly follows the thread-safe lazy initialization pattern used throughout this file. The use of
std::call_onceensures safe initialization, and the default value of "UCX" is appropriate.
cpp/tensorrt_llm/executor/cache_transmission/nixl_utils/transferAgent.cpp
Show resolved
Hide resolved
zackyoray
force-pushed
the
yorayz/nixl_backend
branch
from
9c6c9ed to
23e7743
Compare
zackyoray
changed the title
zackyoray
force-pushed
the
yorayz/nixl_backend
branch
2 times, most recently
from
6ffdb7d to
a54795c
Compare
|
|
||
| std::string nixlBackend = common::getEnvNixlBackend(); | ||
| // List of supported backends - extend this list as new backends are added | ||
| static const std::set<std::string> supportedBackends = {"UCX"}; |
There was a problem hiding this comment.
Any reason we only support UCX in this change and not other backends (like libfabric)?
There was a problem hiding this comment.
NIXL LIBFABRIC plugin is supported from NIXL 0.6.1.
Current NIXL version supported in TRTLLM is 0.5.0, there is an open PR to upgrade NIXL version to 0.7.1 #9055
We can add this as incremental feature later. or wait and rebase.
In addition, current TRTLLM container doesnt include LIBFABRIC we might want to address that as well.
Regarding other NIXL plugins, i am not if any other plugin can be fit for kvcache transfer for the current PD disagg serving.
There was a problem hiding this comment.
Thanks for explaining. My personal perspective is that we should rebase and add the NIXL option. The reason is that if we're still on NIXL 0.5.0, there is no real selection you can do, so adding the env var is not meaningful, and once we're on >=0.6.1 it is more meaningful to do the change.
In terms of the container not containing libfabric, is this something that needs to be compiled in or can be injected later (e.g. by mounting the plugin into the container)?
There was a problem hiding this comment.
The easiest will have libfabric as dependency, and installed like UCX and NIXL (docker/common/install_ucx.sh, docker/common/install_nixl.sh)
AFAIU from @Shixiaowei02 adding this dependency is reasonable.
I think the best is to have another followup PR (after this and #9055 will be merged) to support EFA and discuss there the full approach.
@ebarilanM what do you think?
zackyoray
force-pushed
the
yorayz/nixl_backend
branch
from
a54795c to
dd9b017
Compare
… NIXL backend selection Allows users to specify which NIXL backend to use for KV cache transmission. Usage: TRTLLM_NIXL_KVCACHE_BACKEND=<backend> Currently supports only UCX backend. Unsupported backend types will fallback to Default (UCX). Signed-off-by: Yoray Zack <62789610+zackyoray@users.noreply.github.com>
zackyoray
force-pushed
the
yorayz/nixl_backend
branch
from
dd9b017 to
b630a4e
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #24426 [ run ] triggered by Bot. Commit: |
|
PR_Github #24426 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #24571 [ run ] triggered by Bot. Commit: |
|
PR_Github #24571 [ run ] completed with state |
5 participants
Add TRTLLM_NIXL_KVCACHE_BACKEND environment variable for NIXL backend selection
Allows users to specify which NIXL backend to use for KV cache transmission.
Usage: TRTLLM_NIXL_KVCACHE_BACKEND=
Currently supports only UCX backend. Unsupported backend types will fallback to Default (UCX).
Summary by CodeRabbit
Description
Test Coverage
PR Checklist
Please review the following before submitting your PR:
PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.
PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.
Test cases are provided for new code paths (see test instructions)
Any new dependencies have been scanned for license and vulnerabilities
CODEOWNERS updated if ownership changes
Documentation updated as needed
Update tava architecture diagram if there is a significant design change in PR.
The reviewers assigned automatically/manually are appropriate for the PR.
Please check this after reviewing the above items as appropriate for this PR.
GitHub Bot Help
/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...Provide a user friendly way for developers to interact with a Jenkins server.
Run
/bot [-h|--help]to print this help message.See details below for each supported subcommand.
run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]Launch build/test pipelines. All previously running jobs will be killed.
--reuse-test (optional)pipeline-id(OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.--disable-reuse-test(OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.--disable-fail-fast(OPTIONAL) : Disable fail fast on build/tests/infra failures.--skip-test(OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.--stage-list "A10-PyTorch-1, xxx"(OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.--gpu-type "A30, H100_PCIe"(OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.--test-backend "pytorch, cpp"(OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.--only-multi-gpu-test(OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.--disable-multi-gpu-test(OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.--add-multi-gpu-test(OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.--post-merge(OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx"(OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".--detailed-log(OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.--debug(OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in thestage-listparameter to access the appropriate container environment. Note: Does NOT update GitHub check status.For guidance on mapping tests to stage names, see
docs/source/reference/ci-overview.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip testing for latest commit on pull request.
--comment "Reason for skipping build/test"is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.reuse-pipeline
reuse-pipelineReuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.