OpenLake

OpenLake, an open source storage system released today, delivers 8x higher throughput than conventional object stores, ~600 µs small object reads, and over a million concurrent IOPS for LLM inference …

OpenLake, an open-source storage engine for offloading LLM KV caches from GPU memory to RAM and NVMe, cuts GPU time by 48.2% for long-context inference, reducing a 1,169-second workload to 606 seconds…