Y
HN Search
Hacker News Search
new
|
comments
|
top
|
jobs
arnav__1
searching PlanetScale…
1.
▲
2.
▲
3.
▲
4.
▲
5.
▲
6.
▲
3 ms
·
1.
▲
Show HN: We Beat MLPerf: Modern Storage for KV Offload and LLM Training
(theopenlake.com)
35 points
by
arnav__1
3mo ago
|
1 comments
2.
▲
by
arnav__1
3mo ago
Hey HN, Today we are launching OpenLake, a KV offloading solution for modern LLM workloads. OpenLake achieved 1M+ iops in 1ms on a 96 node cluster. We achieved this through rust compio (thread per core design) with on GPU compression. In ML
3.
▲
Show HN: ExANS – Lossless KV cache compression at 622 GB/s on H100
(theopenlake.com)
15 points
by
arnav__1
4mo ago
|
0 comments
4.
▲
Show HN: Cuts Long Horizon Inference Costs by 50% via external KV Cache Offload
(github.com)
22 points
by
arnav__1
4mo ago
|
0 comments