From the Research Universities edition of September 18, 2026
RASER user-space Slurm agent runtime cuts makespan ~39% vs static partitioning
TU Dresden / TU Chemnitz collaborators (arXiv preprint) · University research computing / HPC agent runtimes · Germany (author institutions); transferable to Slurm campuses with limits
- Publisher
- arXiv
- Original publication
- September 3, 2026 (arXiv v1)
- Source retrieved
- 2026-09-19
What happened
RASER extends Slurm with agentic job arrays, filesystem work-stealing queues, application-level checkpointing plus requeue, and Apptainer isolation without image changes. Authors report nearly 39% makespan reduction versus static partitioning (38.8% in the detailed scenario) with near-full CPU utilization and estimated NVMe checkpoint/restore under 30 seconds, exercising resilience with injected preemption.
Why it matters
Why now: Prior research editions covered capacity awards and validation-gated agents; this still-unarchived Sep 2026 runtime paper supplies measured HPC scheduling benefits/constraints for deploying agentic scientific workflows on existing campus clusters.
Evidence and measured results
Comparative makespan vs static partitioning; CPU utilization claims; checkpoint overhead estimate; fault injection via scancel SIGTERM; user-space/no-kernel-privilege design.
Limitations and uncertainty
Scheduling metrics ≠ scientific outcomes; authors flag FS lock contention at higher concurrency; site Apptainer/Slurm assumptions vary.
Put this evidence to work
Lighthouse Advisory interpretation, grounded in this source. Enriched 2026-09-19; this does not change the original publication date. Each role follows the same checklist so takeaways compare across sources.
Sales
Role takeaway
- Customer problem
- Agentic research workflows fit poorly on static Slurm partitions, wasting capacity and failing under preemption.
- Stakeholders
- HPC center leadership
- Research computing architects
- Faculty running agent workflows
- Security/container owners
- Discovery questions
- Is Apptainer approved?
- How often are jobs preempted?
- What shared FS backs queues?
- What static array inefficiencies exist today?
- Value hypothesis
- User-space work-stealing agent arrays with checkpoint/requeue can cut makespan ~39% vs static partitioning when CPU utilization is the bottleneck—without claiming discovery quality gains.
- Potential engagement
- Non-production partition pilot comparing static arrays vs work-stealing agent arrays with fault injection.
- Unsupported claims
- Scientific accuracy improvements
- Guaranteed performance under extreme FS lock contention
- Works without Apptainer/Slurm
Pre-sales engineering
Role takeaway
- Fit
- User-space Slurm extension for dynamic agentic jobs with work stealing and checkpoint/requeue.
- Architecture and integration
- Wrap agents in Apptainer; use shared FS queues; serialize agent state; requeue via Slurm primitives.
- Prerequisites
- Slurm
- Apptainer
- Shared filesystem
- Agent state serialization API
- Constraints
- FS lock contention at high concurrency
- Site preemption policies
- Not a science-result validator
- Security
- Review container bind mounts and direct hardware access; enforce project isolation on shared queues.
- Proof of value
- Makespan vs static partitioning
- CPU utilization
- Checkpoint restore time
- Survival under injected preemption
Delivery
Role takeaway
- Work
- Deploy pilot plugins, run comparative workloads, inject preemption, and publish ops runbooks.
- Dependencies
- HPC center change window
- Apptainer support
- Research group agent workloads
- Ownership
- Research computing owns the runtime; PIs own agent application logic.
- Skills and adoption
- Train users to submit agent arrays and recover from requeue rather than one-shot jobs.
- Governance checkpoints
- Preemption drill
- FS contention review
- Container policy check
- Acceptance criteria
- Documented makespan improvement
- Restore within agreed seconds
- Written non-goal: scientific correctness
- Risks
- Shared FS hotspots
- Misread scheduling metrics as science progress
- Container policy rejection
Implementation considerations
Lighthouse Advisory interpretation across the operating dimensions a public-sector buyer must settle before this evidence becomes a design. Each note answers the question under its heading for this specific source.
Architecture and integration
What must connect, and where does the AI sit in the workflow?
Campus agent platforms should assume Slurm-native user-space orchestration with shared-FS queues and container isolation rather than bespoke privileged schedulers.
Governance
Who approves, reviews and stays accountable for outcomes?
HPC centers must define who may run immortal agent arrays and how preemption/accounting apply.
Security and privacy
What data, permissions and controls need testing?
Apptainer bind mounts with direct hardware access require project isolation and mount review.
Accessibility and workforce
Who is affected, and what skills or accommodations follow?
User-space design aims to democratize agentic HPC without admin privileges, but still needs Apptainer-fluent support staff.
Procurement
What should contracts, pricing and exit terms secure?
Prefer solutions that integrate with existing Slurm/Apptainer estates over net-new privileged schedulers.
Operating model
Which teams own the service once it runs?
Moves agent workflow ops into research-computing runtimes while researchers retain agent logic ownership.
Publication history
- 2026-09-18Research Universities · Issue 134 resources
Stable resource ID: raser-slurm-agentic-workstealing-makespan-20260903