The repaste worked. The VRAM got 12 °C hotter.
Follows on from Putting an RTX 3090 in an HP ProLiant ML350 Gen9. The card is a second-hand blower-style 3090 in an enterprise tower, power-capped to 275 W.
Read entryTopic
6 observations in this topic
Follows on from Putting an RTX 3090 in an HP ProLiant ML350 Gen9. The card is a second-hand blower-style 3090 in an enterprise tower, power-capped to 275 W.
Read entryI run semantic search over a few thousand recorded meetings. The single thing that makes it work is not the embedding model or the vector store — it is a paragraph of context an LLM writes for every chunk before I embed it. That enrichment step roughly doubled retrieval quality. It is also the slowest, most expensive part of the pipeline, so this summer I tried to make it cheaper by swapping the model that writes the context.
Read entryThe castle stays put. You only get a small bag of extra bricks.
Read entryFourth in a series. Follows the DeepSeek-V4-Flash single-3090 note. Living document; numbers update as the campaign runs. Last updated 2026-08-16. A revision history is at the bottom, and it is worth reading: this note has retracted one headline and reversed one verdict since rev 1.
Read entryFollows on from Putting an RTX 3090 in an HP ProLiant ML350 Gen9. Living document; numbers update as the campaign runs. Last updated 2026-08-13.
Read entryWe have been running low-bit quantization experiments on CPU only. It works, and it is slow enough that the feedback loop hurts — a context-depth sweep is most of an afternoon. The GPUs we already own are doing production work, so borrowing one was not an option. What we did have spare was an older dual-socket ML350 Gen9 tower with 512 GB of RAM, which is a good shape for MoE expert offload and had no GPU at all. So: a second-hand RTX 3090, blower-style, into an enterprise tower that was never designed for one.
Read entry