MetaRoCE for AI Infrastructure on Ethernet
MetaRoCE for Ethernet-based AI infrastructure: how Meta moves intelligence into the NIC, eliminates PFC, and creates a loss-resilient transport for million-GPU scale.
Architecture and Infra on ThecoreGrid covers the foundations of designing and operating scalable, reliable systems at BigTech level. This category brings together system design and infrastructure practices: distributed architectures, highload patterns, cloud-native platforms, and core layers such as compute, networking, and storage. We focus on real engineering decisions — how to balance reliability, performance, cost, and long-term system evolution. Topics include Infrastructure as Code, Kubernetes, multi-region deployments, traffic management, and platform design. Content is grounded in production experience: incident post-mortems, large-scale migrations, and lessons from operating infrastructure under heavy load. Instead of abstract theory, you get practical trade-offs, proven patterns, and insights drawn from real-world systems. Architecture & Infra is built for architects, backend and platform engineers, DevOps teams, and SREs responsible for complex distributed systems and mission-critical infrastructure.
MetaRoCE for Ethernet-based AI infrastructure: how Meta moves intelligence into the NIC, eliminates PFC, and creates a loss-resilient transport for million-GPU scale.
AI Security in Production at Roblox: how sandboxing, guardrails, and exemplars help safely lead AI agents from prompt to production.
A central gateway for full telemetry: sizing, load testing, HPA, WAL, and GOMEMLIMIT in a production scenario—with no blind spots.
Automated RCA in microservices turns metrics, logs, and traces into ranked root cause hypotheses for faster incident validation and cleaner diagnosis
Data replication for AI agents: how Aurora, DynamoDB, and Keyspaces help avoid stale reads, race conditions, and context errors.
remote Spectre attack on Cloudflare Workers: how DyPrIs, V8 Sandbox, and in-process isolation reduced the risk of memory leakage in production.
Git fetch in CI became a bottleneck. An analysis of the gitretriever architecture at Datadog, CPU reductions, relay models, and APIs for code queries.
Container security at Swift: how zero-touch automation, central scanning, and regression testing reduced noise and simplified protection for hundreds of clusters.
eBPF for AI API in Kubernetes: how to intercept traffic, limit AI agents, and control behavior without code changes and container restarts.
Container security at Swift: how centralized architecture, automated scans, and regression testing reduced manual noise in hybrid cloud.
Controls: ← → to move, ↑ to rotate, ↓ to drop.
Mobile: use buttons below.