{"type":"rich","version":"1.0","provider_name":"Transistor","provider_url":"https://transistor.fm","author_name":"Automatic","title":"GPU Scheduling: Herding Cores in the Cloud","html":"<iframe width=\"100%\" height=\"180\" frameborder=\"no\" scrolling=\"no\" seamless src=\"https://share.transistor.fm/e/b00d0c4a\"></iframe>","width":"100%","height":180,"duration":488,"description":"GPU scheduling in the cloud sounds straightforward until your cluster is half-idle and half-on-fire. This episode breaks down why matching workloads to GPU resources is genuinely hard — and what thoughtful scheduling actually looks like in practice.","thumbnail_url":"https://img.transistorcdn.com/geAvK8Lf2TpOu15wwN-7jYkVWQf54CoengvnTEp3AzQ/rs:fill:0:0:1/w:400/h:400/q:60/mb:500000/aHR0cHM6Ly9pbWct/dXBsb2FkLXByb2R1/Y3Rpb24udHJhbnNp/c3Rvci5mbS8zNDA3/NDliNDQwMTkxMzZi/MzA0YTM3NGQxZTc1/NTk4MC5wbmc.webp","thumbnail_width":300,"thumbnail_height":300}