Advanced

Request Networking: DNS, TCP and TLS

A model can be healthy while requests never reach it. Trace name resolution, connection setup, encryption and HTTP routing; use boundary evidence to separate a network failure from a slow inference engine.

Unlock the full curriculum — ₹2,000 / $25every concept + every answer · 6 months · no auto-renew

RELATED CONCEPTS

Inference Platform Architecture Latency Metrics: TTFT, TPOT and Goodput Control Plane and API Design for GPU Platforms RDMA, InfiniBand and RoCEv2

LESSONS THAT TEACH THIS

From Model to ServiceTrace a request before choosing a platform12 min→

PRACTICE THIS IN REAL QUESTIONS

Networking, Interconnects & StorageWhat is RDMA, and when does it help a training cluster more than TCP?→GPU & Accelerator ArchitectureExplain GPUDirect RDMA. What does it remove from the data path, and when does it change performance?→GPU & Accelerator ArchitectureDesign a non-blocking network fabric for 512 H100s. How many switches, how are they wired, and where does the NVLink domain end?→Networking, Interconnects & StorageWhat does NCCL actually do when you call all-reduce, and how does it decide which algorithm to use?→Networking, Interconnects & StorageName the four collectives a training job uses, say what each moves, and match them to the parallelism that needs them.→Networking, Interconnects & StorageInfiniBand or RoCE version 2 for a new GPU training cluster. Make the call and say what would reverse it.→