We want to bring the LLM-D test engine from TOPSAIL, so that we can demonstrate FOURNOS multi job scheduling.
We want to launch jobs on two clusters (eg, CKS and PSAP-H200) so that they run concurrently.
On each clusters, we want to launch 4 fours that will run sequentially:
llama3.3-70b heterogeneousllama3.3-70b multiturngpt-oss-120b heterogeneousgpt-oss-120b multiturn
We want to bring the LLM-D test engine from TOPSAIL, so that we can demonstrate FOURNOS multi job scheduling.
We want to launch jobs on two clusters (eg, CKS and PSAP-H200) so that they run concurrently.
On each clusters, we want to launch 4 fours that will run sequentially:
llama3.3-70b heterogeneousllama3.3-70b multiturngpt-oss-120b heterogeneousgpt-oss-120b multiturn