Meta’s AI Storage Blueprint at Scale

Summary

Meta redesigned its BLOB-storage architecture on top of the foundational Tectonic block layer to eliminate GPU stalls and accelerate AI training workflows. The legacy storage system suffered from multi-layered metadata lookups, cross-region latency, and dataplane proxy bottlenecks that conflicted with the millisecond access requirements of flash-based AI clusters. To resolve these issues, the team collapsed disparate metadata into a single flat schema backed by ZippyDB for O(1) path lookups and introduced a fat client SDK capable of streaming data directly from storage servers. The updated stack also leverages spare GPU host memory as a distributed data cache alongside distributed read-plan caches to handle severe traffic spikes during checkpointing and model loading. These protocol and architectural changes deliver predictable latencies, reduce power consumption, and maintain high throughput during distributed training.

Takeaways

• Replacing multi-layer metadata lookups with a flat ZippyDB schema enables O(1) storage address resolution, while a fat client SDK removes dataplane proxies to stream bytes directly from Tectonic storage servers.
• Integrating Meta's Owl subsystem components directly into the client SDK turns spare GPU host memory into a distributed cache that achieves an average 80% hit rate and insulates storage from traffic spikes.
• The client SDK employs hedged reads to mitigate tail latencies from slow storage nodes and uses dynamic concurrency control to regulate egress spikes during checkpointing events.

Related reading

• Meta · Jul 15, 2026

  Exploring Hierarchical Interest Representation For Meta Ads Deep Funnel Optimization

  Meta developed Hierarchical Interest Representation as an upstream representation layer to optimize deep funnel ad ranking across its platforms. The architecture organizes interactions among users, advertisers, and products into a heterogeneous, time-decayed graph and projects sparse raw edges into multi-level super-graphs of latent interest primitives. To encode these structures, a transformer-based hierarchical encoder incorporates multimodal product semantics processed via language models alongside topology-aware attention biases. Computation scales across billions of entities by pairing deep hash ID embeddings with FlexAttention to evaluate graph biases on the fly without materializing full matrices. Self-supervised cross-view distillation trains student representations against broader teacher graph views to yield universal embeddings and interest tokens.

  By Yuhui Ouyang, Di Wang, Sreedal Menon, Jie Tian

• Meta · Aug 24, 2026

  MTIA 300: Meta’s First Training Chip with Built-in NICs and Communication-Offloading Engines

  Training recommendation models poses distinct network bottlenecks because massive embedding tables trigger frequent collective operations that compete with computation on standard GPUs. Meta designed the MTIA 300 accelerator to overcome this by embedding two network chiplets containing twelve 800 Gbps RDMA NICs directly inside the chip package. The architecture offloads communication from its compute grid to sixteen dedicated message engines equipped with RISC-V cores and near-memory compute reduction blocks. Working alongside the co-designed HCCL library, the chip autonomously compiles and executes collective communication subgraphs without host CPU intervention. On a 150-billion-parameter production recommendation model, MTIA 300 delivered communication speeds 3.9 times faster than an equivalent GPU cluster while keeping concurrent compute degradation under 0.5 percent.

  By Rajiv Krishnamurthy, Wes Bland

• Meta · Aug 24, 2026

  MetaRoCE: A New RDMA Transport Built for AI-Scale Ethernet

  Scaling frontier AI training and inference across massive GPU clusters places severe demands on network fabrics, yet conventional RoCE relies on Priority Flow Control to enforce in-order packet delivery. Meta introduced MetaRoCE, a clean-sheet RDMA transport protocol engineered specifically for lossy, multiplane commodity Ethernet without requiring PFC. By shifting intelligence to endpoint NICs, the protocol sprays packets across fine-grained logical paths, writing out-of-order data directly into destination memory without reorder buffers or head-of-line blocking. Congestion is managed via per-path ECN windows paired with receiver-driven rate hints returned in acknowledgments. In validation on a 64-node AMD GPU cluster running collective operations, MetaRoCE outperformed RoCEv2, sustaining roughly 86% throughput under 1% packet loss and achieving linear scaling across multiplane fabrics.

  By Arvind Srinivasan, Neil Spring, Omar Baldonado, Rajiv Krishnamurthy

• Meta · Jun 30, 2026

  10 Years of Meta’s Commitment to Python

  Meta has marked ten consecutive years as a sponsor of the Python Software Foundation (PSF) to ensure the language remains robust and sustainable. Python represents the most widely used programming language across Meta's infrastructure, powering applications like Instagram and Threads as well as internal AI research. Beyond financial backing, Meta engineers actively contribute core language features, author Python Enhancement Proposals (PEPs), and build open-source tools such as the Pyrefly type checker. Company sponsorship directly funds the PSF's Developer-in-Residence program, core infrastructure, and security enhancements for the Python Package Index (PyPI). Continued support of the PSF functions as a strategic investment in maintaining the stability of the broader open-source ecosystem.