September 13, 2026· ~4 min readLoad-Testing Production Without Blowing It Up: an 8 GB CEILING lesson Load tests feel safe because they are "just requests." No CI pipeline I know of warns you that the request you're blasting at production is itself th
May 28, 2026· ~6 min readTaming OOM on DGX Spark: Debugging Unified Memory Pressure in a 2-Node vLLM Cluster Your cluster loads the model. Weights land in GPU memory. Then — crash. `NV_ERR_NO_MEMORY`. The OOM killer takes down `VLLM::Worker_TP`. You restart.