Kelly Greer on X: "We’ve spent the last couple of months deep in financing and coordinating our own B300 cluster deployment. Suffice to say buying, financing, deploying and monetizing GPUs is a wild process, so we figured we’d outsource our findings with ideas on improving this market. DMs open if https://t.co/OxwB8ea2zM" / X
To view keyboard shortcuts, press question mark View keyboard shortcuts Home Explore Notifications Chat Grok Communities Premium Profile More Post kristie! @kristiehuang Post See new posts Conversation Pinned Kelly Greer @kellyjgreer We’ve spent the last couple of months deep in financing and coordinating our own B300 cluster deployment. Suffice to say buying, financing, deploying and monetizing GPUs is a wild process, so we figured we’d outsource our findings with ideas on improving this market. DMs open if you’re working on any of this: 1. First important to address that compute is product led not asset led. Consumption experience is the product, so buying bare metal is only part of the equation of bring a cluster to market (also why 1GPUhr does not = 1GPUhr). Orchestration and networking are critical processes to maintain on an ongoing basis. H/t to the start ups doing this as part of the monetization (selling the compute via contracts or marketplace) process- @fluidstack , @sf
Kelly Greer @kellyjgreer We’ve spent the last couple of months deep in financing and coordinating our own B300 cluster deployment. Suffice to say buying, financing, deploying and monetizing GPUs is a wild process, so we figured we’d outsource our findings with ideas on improving this market. DMs open if you’re working on any of this: 1. First important to address that compute is product led not asset led. Consumption experience is the product, so buying bare metal is only part of the equation of bring a cluster to market (also why 1GPUhr does not = 1GPUhr). Orchestration and networking are cri
Explore this link on the map →related reading
- evan conrad on X: "What SF Compute does. When you finance a GPU cluster, you need to get an "offtake" agreement. Basically, someone has to agree to rent the cluster from you, typically for a 3+ year period. If that agreement falls through (x.com
- Nobody knows what a used GPU cluster is worthciphertalk.substack.com
- ClusterMAX™ 2.0: The Industry Standard GPU Cloud Rating Systemnewsletter.semianalysis.com
- Designing Compute Markets | Kavish Gargkavishgarg.com
- Nvidia H100 GPUs: Supply and Demand · GPU Utils ⚡️gpus.llm-utils.org
- Nvidia Envy: understanding the GPU gold rushblog.johnluttig.com
- Accelerate AI & Machine Learning Workflows | NVIDIA Run:airun.ai
- From bare metal to a 70B model: infrastructure set-up and scripts - Imbueimbue.com
- How to Rack 30 Petabytes of Storage | blogsi.inc
- Meta Compute: Everyone Wants To Be A Neocloudnewsletter.semianalysis.com
- The Financialization of the Data Center: GPU Clouds, Debt and PE fueling AI infrastructure supercycledatagravity.dev
- GPU Instances and Serverless Inference — Verda (formerly DataCrunch)verda.com