VRAM GB →GiB / MiB
80 GB H100. nvidia-smi reports 74.5 GiB. Your 78 GB model just crashed with CUDA OOM. The gap is base-10 vs base-2 —not a firmware bug.
GB-to-GiB Converter →TFLOPS →PFLOPS
8× H100 nodes = 15.8 PFLOPS FP8. Your architecture review deck still says TFLOPS. Convert before the VP asks why the number looks small.
TFLOPS →PFLOPS Scaler →Mbps →MB/s Throughput
ISP quotes 1,000 Mbps. Your NVMe array benchmarks in MB/s. Divide by 8. Subtract 5—2% for protocol overhead. Now size the pipe.
Mbps →MB/s Converter →Gbps →TB / Day Egress
10 Gbps sustained = 108 TB/day. At AWS egress rates that's $9,180/day. Model the pipe before the CFO forwards you the bill with a question mark.
Gbps →TB/day Egress Meter →