Question 1

Will our admins have to run something they didn't write inside SLURM?

Accepted Answer

No. Cedana does not fork or replace SLURM and does not change the interface; sbatch, srun, and squeue stay exactly as they are. It installs as a bootstrap script plus a plugin, typically in under an hour, transparent at the OS and kernel level. See the installation instructions.

Question 2

Our researchers aren't asking for checkpointing. Why deploy it?

Accepted Answer

The pitch is to the operations team, not the researchers. The gain lands on effective utilization, uptime, and cost per researcher served. Researchers benefit downstream without doing anything: their jobs survive interruptions and they stop over-allocating to hedge.

Question 3

We run a condo model and the lab politics are the hard part. Does this help?

Accepted Answer

Yes. The politics exist because preemption destroys work. Cedana lets a borrowed job checkpoint and resume elsewhere when the owner reclaims, with no progress lost, so sharing becomes a scheduling policy instead of a standing fight.

Question 4

We have air-gapped and classified workloads. Can it run without calling home?

Accepted Answer

Yes. Cedana supports on-premise and fully air-gapped deployment with no runtime dependency. To be precise, it is not FedRAMP authorized today, which is on the roadmap; what it provides now is system-level provable recovery entirely inside your boundary.

Question 5

Most of our science is multi-node distributed. Can you checkpoint that?

Accepted Answer

Yes, that is the case Cedana is built for. It takes a coordinated snapshot across the job, accounting for in-flight MPI and NCCL collectives, RDMA, and OS and network state, which single-GPU primitives like the NVIDIA driver checkpoint do not do.

Question 6

What does it cost in steady-state performance?

Accepted Answer

Low single digits in percent. Snapshots are taken on a roughly 60-90 second interval, and we measure the exact overhead against your workload mix in a PoC rather than quote a universal number.

More science from the GPUs you already paid for.