SYSTEM Cited by 1 source
Netflix Model Scoring Service (MSS)¶
Definition¶
Model Scoring Service (MSS) is Netflix's shared GPU inference backend — the component that handles model loading, batching, and GPU scheduling for models too large to run in-process on CPU. It sits behind the unified JVM-based serving system and supports XGBoost, TensorFlow, PyTorch, and LLMs behind a single interface, with NVIDIA Triton Inference Server as the compute layer managing GPU resources.
Architecture¶
MSS is structured as:
- Triton Inference Server (compute layer) — manages model loading, continuous batching, GPU scheduling, and multi-model concurrency.
- Java control plane (on top of Triton) — handles deployment orchestration, versioning, health checking, autoscaling, and multi-region rollout.
- vLLM (paved-path LLM engine) — integrated as the inference engine for LLM workloads within Triton, replacing TensorRT-LLM as of 2025.
Model authors package artifacts and configure deployment; the control plane provisions GPU instances, configures Triton, and orchestrates zero-downtime upgrades.
Access paths¶
Two paths reach MSS:
- gRPC (existing) — via the JVM serving system, which handles routing, A/B, feature fetching, pre/post-processing, and logging.
- Direct HTTP (newer) — OpenAI-compatible API for LLM-driven applications, exposed via Triton's OpenAI-compatible frontend (FastAPI + embedded Triton).
Deployment strategies¶
- Red-Black — new version alongside old, phased traffic shift, atomic rollback on failure. Default for stable interfaces.
- Versioned — independent per-(modelId, modelVersion) deployment for breaking interface changes. Temporary GPU cost overhead accepted for correctness.
Operational details¶
- Model caching on FSx — models materialized on Amazon FSx at announcement time to avoid S3 download latency at boot.
- Version pinning — Triton and vLLM versions pinned in the service image to prevent API-surface drift.
- Unified metrics proxy — merges Triton and vLLM Prometheus metrics into a single
/metricsendpoint (Triton's bridge only exposes 9 of 40+ vLLM metrics).
Seen in¶
- sources/2026-07-17-netflix-in-house-llm-serving — primary architecture source
- sources/2026-05-01-netflix-state-of-routing-in-model-serving — routing layer above MSS (via systems/netflix-model-serving-platform)