LoginSchedule demo
cedana / blog · field reports & engineering deep-dives

Writing from the
cedana team.

Engineering notes, benchmarks, and field reports covering how we build the automation layer for AI factories.
Featured
Blueprint schematic of an eight-GPU node accumulating session state, with a state gauge rising toward the VRAM capacity lineBlueprint schematic of an eight-GPU node accumulating session state, with a state gauge rising toward the VRAM capacity line
FeaturedEngineering

The Era of Stateful Inference: How to Improve Cost per Token.

We are entering the stateful inference era, driven by frontier models with longer context windows, longer in-flight sessions, and single instances spanning 8, 16, or more GPUs.

Updated · Cedana Editorial
read →
All posts · 08

insights from the team.

Field reports, engineering deep-dives, and benchmarks from the Cedana team.

TopicAll topics

Blueprint timeline of two jobs crossing a 24-hour wall-time boundary: one is terminated and restarts from zero, the other checkpoints and continues into the next allocationBlueprint timeline of two jobs crossing a 24-hour wall-time boundary: one is terminated and restarts from zero, the other checkpoints and continues into the next allocation
· Cedana Editorial

The Wall-Time Limit Forces an Expensive Tradeoff in HPC.

Jobs reach their wall-time limits and lose hours or days of in-memory progress. However, the limit itself is not the problem.

read →
Blueprint roofline plot: performance rises along the memory-bandwidth slope to a ridge point, then flattens at the compute ceilingBlueprint roofline plot: performance rises along the memory-bandwidth slope to a ridge point, then flattens at the compute ceiling
Updated · Cedana Editorial

Roofline Analysis and the Inference Value Chain

Once you stop renting intelligence, you own performance.

read →
Blueprint schematic of an eight-GPU node accumulating session state, with a state gauge rising toward the VRAM capacity lineBlueprint schematic of an eight-GPU node accumulating session state, with a state gauge rising toward the VRAM capacity line
Updated · Cedana Editorial

The Era of Stateful Inference: How to Improve Cost per Token.

We are entering the stateful inference era, driven by frontier models with longer context windows, longer in-flight sessions, and single instances spanning 8, 16, or more GPUs.

read →
Blueprint fleet of GPU cells with only a third lit, next to a utilization gauge stuck at a dashed 30 percent ceiling and a second gauge breaking through itBlueprint fleet of GPU cells with only a third lit, next to a utilization gauge stuck at a dashed 30 percent ceiling and a second gauge breaking through it
Updated · Cedana Editorial

The Utilization Ceiling: Why AI and HPC Schedulers Hit 30% and How to Fix This

GPU utilization across AI and HPC workloads is fundamentally capped at 30% because schedulers cannot migrate running jobs. Cedana's CPU and GPU migration capability surpasses this limitation, unlocking near-full utilization.

read →
Blueprint chart of three trajectories diverging from a point, with an accent curve rising steeply off the chartBlueprint chart of three trajectories diverging from a point, with an accent curve rising steeply off the chart
· Niranjan Ravichandra

The Fork in the Road Post GPT-5

On GPT-5 being underwhelming, scaling laws and where we can get reasonable progress from. Also, should AGI still be the goal?

read →
Blueprint reinforcement-learning loop: an agent block and an environment block joined by circular arrows, with reward pips along the return pathBlueprint reinforcement-learning loop: an agent block and an environment block joined by circular arrows, with reward pips along the return path
· Niranjan Ravichandra

Exploring new frontiers of Reinforcement Learning with Cedana

Cedana <3 RL !

read →
Blueprint molecule graph connected by an arrow to a four-GPU compute nodeBlueprint molecule graph connected by an arrow to a four-GPU compute node
Updated · Niranjan Ravichandra

Supercharging Scientific Computing with Cedana

How we leverage open-source tech (like Kueue) to push scientific compute and accelerate time to insights.

read →
Blueprint of live migration: a dashed reclaimed spot instance, an accent state block in flight, and a healthy instance holding the restored stateBlueprint of live migration: a dashed reclaimed spot instance, an accent state block in flight, and a healthy instance holding the restored state
Updated · Cedana Editorial

Using Cedana to Live-Migrate Stateful Workloads Between Spot Instances

Save, migrate and resume a running XGBoost workload in six steps.

read →

Showing 8 of 8 posts