checkpointing

Without it, every byte a GPU sends over the network crosses PCIe twice and passes through host memory. With it, the NIC reads and writes GPU memory directly. The PCIe arithmetic that shows why a 400G NIC cannot be fed the slow way, the NCCL line that tells you which path you are on, and the checkpoint case.

Meta's Llama 3 report counted 419 unexpected interruptions in 54 days on 16,384 GPUs, one every three hours. The failure-rate arithmetic, the loss as a function of the interval, the square-root formula that minimizes it, and why the write must be sharded and asynchronous first.

Every rank writes its own shard, a metadata file maps every chunk to a global tensor, and a loader intersects the old layout with the new. The bytes per rank for a 405B, the manifest that makes the write atomic, the metadata that makes resharding possible, and the host-memory staging that keeps the GPUs running.

Sixteen bytes per parameter is 6.5 TB per checkpoint; over a 100 GB/s filesystem that is about a minute, which at a 30-minute cadence is 3% of the run. The chain, the storage tiers that meet it, and why the write is made asynchronous.

The token stream is tens of megabytes a second and the checkpoints are tens of gigabytes a second; the two workloads differ by a thousand and want different tiers. The chain from tokens per second to bytes, the checkpoint burst, and the restore that sets the real spec.

Activations are the term the 16-bytes-per-parameter rule leaves out, and at 8k context they are bigger than the weights: about 34 × sequence × hidden bytes per layer, 2.3 GB, 182 GB per micro-batch across 80 layers. The chain, what FlashAttention removes, and what checkpointing does to the number and the compute.

Spot is 60% cheaper until the preemptions eat the discount. The expected-loss formula (waste per preemption = half the checkpoint interval plus restart), the break-even worked at three cadences, the checkpoint cost that pushes back the other way, and the gang arithmetic that says which jobs must never be on spot.

A hundred thousand jobs is a queue you cannot scan per tick, a preemption policy that has to know when each victim last checkpointed, and a state machine with one transition that most designs get wrong. The data model, the tick loop, the checkpoint-aware eviction and the arithmetic behind each.