A field guide to performance and cost engineering: claim-checking with the scaling and queueing laws, measurement discipline, where GPU and CPU budgets leak, and how to turn a benchmark into a hardware decision.
Second in my benchmark best practices series. Little’s law checks that throughput, latency and concurrency describe the same system, exposes the load generator’s ceiling, and turns a latency budget into a sizing requirement.
GPT-6-Astra XHIGH sits alone on the DeepSWE cost-performance frontier, and an older model beats a flagship tier. Follow-up: DeepSeek V4.1 Flash hits 98% of Astra’s score at 1.4% of the cost.
Part 1 of my benchmark best practices series: how I turn a scaling claim into an implied serial fraction with Amdahl’s law, and what the Universal Scalability Law adds. Includes a measured pipeline where assigning more vCPUs per image cut global throughput.