NBI-Slurm: Simplified submission of Slurm jobs with energy saving mode

Andrea Telatin

arxiv: 2604.04558 · v1 · submitted 2026-04-06 · 💻 cs.DC

NBI-Slurm: Simplified submission of Slurm jobs with energy saving mode

Andrea Telatin This is my paper

Pith reviewed 2026-05-10 19:47 UTC · model grok-4.3

classification 💻 cs.DC

keywords nbi-slurmslurmhpc job submissionenergy-aware schedulingeco modetui job managementjob wrappersperl hpc tools

0 comments

The pith

NBI-Slurm simplifies job submission on SLURM clusters and adds an eco mode to defer flexible jobs to off-peak hours for energy savings.

A machine-rendered reading of the paper's core claim, the machinery that carries it, and where it could break.

The paper presents NBI-Slurm as a Perl package that gives users a simpler way to submit and manage jobs on SLURM-based high-performance computing systems. It combines a set of command-line tools and Perl modules to lower the effort needed compared to the standard SLURM commands. The package adds text-based interfaces for checking and stopping jobs, tools to create custom wrappers around bioinformatics programs, and an energy-saving feature called eco mode. In eco mode the system identifies jobs that can wait and runs them during times of lower demand, cutting the overall energy use and carbon output of the cluster. Readers would care because this tackles both the difficulty of using HPC resources and the environmental impact of large-scale computing in research.

Core claim

NBI-Slurm is a Perl package that provides a simplified, user-friendly interface for submitting and managing jobs on SLURM high-performance computing clusters. It offers both a library of Perl modules for programmatic job management and a suite of command-line tools designed to reduce the cognitive overhead of SLURM's native interface. Distinctive features of NBI-Slurm are TUI applications to view and cancel jobs, the possibility to generate tool-specific wrappers for bioinformatic tools and an energy-aware scheduling mode called eco mode that automatically defers flexible jobs to off-peak periods, helping research institutions reduce their computational carbon footprint without requiring the

What carries the argument

The NBI-Slurm Perl package, which integrates TUI applications for job viewing and cancellation, a generator for tool-specific wrappers, and an eco-mode scheduler that defers flexible jobs to off-peak periods.

If this is right

Institutions can lower computational carbon footprint by shifting work automatically without requiring users to choose submission times.
Users can view and cancel jobs through dedicated TUI applications rather than relying solely on command-line syntax.
Bioinformatics researchers can generate custom wrappers for specific analysis tools to reduce submission errors.
Programmers gain a Perl library for embedding job submission and management into larger scripts.
Research groups avoid manual planning of job timing while still benefiting from energy-aware scheduling.

Where Pith is reading between the lines

These are editorial extensions of the paper, not claims the author makes directly.

The separation of job flexibility classification from user decisions could extend to other workload managers beyond SLURM.
Wider adoption might prompt clusters to maintain explicit off-peak capacity as a design choice rather than an afterthought.
The wrapper generator could be extended to support pipelines that mix multiple tools, reducing configuration overhead further.
Energy savings from deferred jobs might compound if combined with real-time monitoring of cluster power draw.

Load-bearing premise

That the eco mode can accurately classify jobs as flexible without user input and that sufficient off-peak cluster capacity exists to absorb the deferred work without unacceptable delays to research.

What would settle it

A controlled test submitting known flexible jobs through eco mode and verifying whether execution occurs only during off-peak windows while tracking any change in overall job completion times or user-reported delays.

Figures

Figures reproduced from arXiv: 2604.04558 by Andrea Telatin.

**Figure 1.** Figure 1: Interactive TUI of viewjobs, showing job navigation, multi-column display, and bulk-cancel workflow. The image is AI generated from a real screenshot (Google NanoBanana) Energy consumption in research computing is a growing concern [Lannelongue et al., 2021]. Most researchers have no practical mechanism to shift flexible jobs to periods when grid electricity is cheaper or cleaner. NBI-Slurm addresses this… view at source ↗

read the original abstract

NBI-Slurm is a Perl package that provides a simplified, user-friendly interface for submitting and managing jobs on SLURM high-performance computing (HPC) clusters. It offers both a library of Perl modules for programmatic job management and a suite of command-line tools designed to reduce the cognitive overhead of SLURM's native interface. Distinctive features of NBI-Slurm are (a) TUI applications to view and cancel jobs, (b) the possibility to generate tool-specific wrappers for (bioinformatic) tools and (c) an energy-aware scheduling mode -- "eco mode" -- that automatically defers flexible jobs to off-peak periods, helping research institutions reduce their computational carbon footprint without requiring users to manually plan submission times.

Editorial analysis

A structured set of objections, weighed in public.

Desk editor's note, referee report, simulated authors' rebuttal, and a circularity audit. Tearing a paper down is the easy half of reading it; the pith above is the substance, this is the friction.

Desk Editor's Note private letter to a colleague

NBI-Slurm is a practical Perl wrapper for SLURM that adds a TUI and an eco-mode deferral feature, but the paper is only a feature list with no code, benchmarks, or validation.

read the letter

This paper describes NBI-Slurm, a Perl package meant to make SLURM job submission easier. It includes TUI apps for checking and canceling jobs, a way to auto-generate wrappers for specific bioinformatics tools, and an eco mode that pushes flexible jobs to off-peak hours to lower energy use and carbon output. Those are the concrete additions over plain SLURM commands. The eco-mode idea is sensible for places that want to cut power costs without forcing users to schedule manually, and the TUI plus wrapper generator could reduce friction for people who run lots of short or batch jobs. The description is clear about what the tools are supposed to do. The main weakness is that the manuscript gives no evidence the features work as claimed. There are no usage examples, no timing or energy measurements, no tests of how the deferral logic decides what counts as flexible, and no comparison to existing SLURM wrappers or schedulers. Without the actual code or any data, it is impossible to judge whether the eco mode avoids delaying important work or whether the TUI is reliable. The paper is a straightforward software announcement rather than a research claim with testable results. It would be useful to someone who already runs a SLURM cluster and wants to try a new set of scripts, but it does not contain enough substance for a methods or systems paper. I would not send it for peer review in a standard journal; it belongs on GitHub or in a tools-oriented venue if the code is released and documented.

Referee Report

1 major / 1 minor

Summary. The manuscript describes NBI-Slurm, a Perl package providing a simplified interface for submitting and managing jobs on SLURM HPC clusters. It includes Perl modules for programmatic use, command-line tools, TUI applications for viewing and canceling jobs, generation of tool-specific wrappers for bioinformatic tools, and an 'eco mode' that automatically defers flexible jobs to off-peak periods to reduce carbon footprint without manual user scheduling.

Significance. If the features are implemented as described, NBI-Slurm offers a practical reduction in cognitive overhead for SLURM users and a mechanism for institutions to lower the carbon impact of flexible HPC workloads. The combination of TUI interfaces, wrapper generation, and automated energy-aware scheduling addresses real usability and sustainability needs in bioinformatics and general research computing. The work is a straightforward software description rather than a novel algorithmic contribution.

major comments (1)

[Abstract] Abstract (eco mode description): the claim that eco mode 'automatically defers flexible jobs to off-peak periods' without user input rests on an unspecified mechanism for classifying jobs as flexible and for ensuring off-peak capacity exists; this is load-bearing for the central promise of carbon-footprint reduction without delaying critical work, yet no decision criteria, fallback behavior, or cluster-capacity assumptions are stated.

minor comments (1)

The manuscript would benefit from a short usage example (e.g., a command-line invocation of wrapper generation or eco-mode submission) to illustrate the claimed simplification over native SLURM.

Simulated Author's Rebuttal

1 responses · 0 unresolved

We thank the referee for the positive evaluation and the recommendation of minor revision. We have addressed the single major comment by revising the abstract and expanding the relevant section of the manuscript to provide the requested details on the eco-mode mechanism.

read point-by-point responses

Referee: [Abstract] Abstract (eco mode description): the claim that eco mode 'automatically defers flexible jobs to off-peak periods' without user input rests on an unspecified mechanism for classifying jobs as flexible and for ensuring off-peak capacity exists; this is load-bearing for the central promise of carbon-footprint reduction without delaying critical work, yet no decision criteria, fallback behavior, or cluster-capacity assumptions are stated.

Authors: We agree that the original abstract was too concise and did not convey the classification and scheduling logic. In the revised manuscript we have updated the abstract to state that users explicitly mark jobs as flexible at submission time via a command-line option, after which the system monitors SLURM-reported cluster load and defers execution only when utilization falls below a configurable threshold; a user-specified maximum deferral window serves as fallback to prevent indefinite postponement. We have also added a short dedicated paragraph in the main text that spells out the decision criteria, the fallback rule, and the operating assumption that the target cluster exhibits predictable diurnal load variation. These changes make the central claim self-contained while remaining faithful to the existing implementation. revision: yes

Circularity Check

0 steps flagged

No significant circularity

full rationale

The manuscript is a software announcement describing a Perl package and its command-line/TUI interface for SLURM job management. There are no mathematical derivations, equations, fitted parameters, predictions, uniqueness theorems, or self-citations that form any load-bearing chain. The central claims simply assert that listed features (TUI viewers, wrapper generation, and an 'eco mode' scheduler) exist in the released code, which can be verified directly by inspecting the implementation. No circularity is present.

Axiom & Free-Parameter Ledger

0 free parameters · 0 axioms · 0 invented entities

No free parameters, axioms, or invented physical entities are required; the contribution is an implementation of user-interface and scheduling logic on top of the existing SLURM system.

pith-pipeline@v0.9.0 · 5418 in / 1093 out tokens · 33629 ms · 2026-05-10T19:47:58.390479+00:00 · methodology

discussion (0)

Lean theorems connected to this paper

Citations machine-checked in the Pith Canon. Every link opens the source theorem in the public Lean library.

IndisputableMonolith/Foundation/BranchSelection.lean branch_selection unclear

?

unclear
Relation between the paper passage and the cited Recognition theorem.

an energy-aware scheduling mode -- 'eco mode' -- that automatically defers flexible jobs to off-peak periods

What do these tags mean?

matches: The paper's claim is directly supported by a theorem in the formal canon.
supports: The theorem supports part of the paper's argument, but the paper may add assumptions or extra steps.
extends: The paper goes beyond the formal theorem; the theorem is a base layer rather than the whole result.
uses: The paper appears to rely on the theorem as machinery.
contradicts: The paper's claim conflicts with a theorem or certificate in the canon.
unclear: Pith found a possible connection, but the passage is too broad, indirect, or ambiguous to say the theorem truly supports the claim.

Reference graph

Works this paper leans on

5 extracted references · 5 canonical work pages

[1]

Loïc Lannelongue, Jason Grealey, and Michael Inouye

doi: 10.1038/nbt.3820. Loïc Lannelongue, Jason Grealey, and Michael Inouye. Green algorithms: Quantifying the carbon footprint of computation. Advanced Science , 8 (12), May

work page doi:10.1038/nbt.3820
[2]

Advanced Science 8(12), 2100707 (2021)

ISSN 2198-3844. doi: 10.1002/advs.202100707. URL http://dx.doi.org/10.1002/advs.202100707. 6 Felix Mölder, Kim Philipp Jablonski, Brice Letcher, Michael B. Hall, Christo- pher H. Tomkins-Tinch, Vanessa Sochat, Jan Forster, Soohyun Lee, Sven O. Twardziok, Alexander Kanitz, Andreas Wilm, Manuel Holtgrewe, Sven Rah- mann, Sven Nahnsen, and Johannes Köster. S...

work page doi:10.1002/advs.202100707
[3]

F1000Research (2021)

ISSN 2046-1402. doi: 10.12688/f1000research.29032.2. URL http://dx.doi.org/10.12688/f1000res earch.29032.2. Bo Wang, Zhiguang Chen, and Nong Xiao. A survey of system scheduling for hpc and big data. In Proceedings of the 2020 4th International Conference on High Performance Compilation, Computing and Communications , HP3C 2020, page 178–183. ACM, June

work page doi:10.12688/f1000research.29032.2 2046
[4]

URL http://dx.doi.org/10.1145/3407947.3407977

doi: 10.1145/3407947.3407977. URL http://dx.doi.org/10.1145/3407947.3407977. Andy B. Yoo, Morris A. Jette, and Mark Grondona. Slurm: Simple linux utility for resource management. In Job Scheduling Strategies for Parallel Processing , page 44–60. Springer Berlin Heidelberg,

work page doi:10.1145/3407947.3407977
[5]

Jette, C

ISBN 9783540397274. doi: 10.1007/10968987_3. URL http://dx.doi.org/10.1007/10968987_3. 7

work page doi:10.1007/10968987_3

[1] [1]

Loïc Lannelongue, Jason Grealey, and Michael Inouye

doi: 10.1038/nbt.3820. Loïc Lannelongue, Jason Grealey, and Michael Inouye. Green algorithms: Quantifying the carbon footprint of computation. Advanced Science , 8 (12), May

work page doi:10.1038/nbt.3820

[2] [2]

Advanced Science 8(12), 2100707 (2021)

ISSN 2198-3844. doi: 10.1002/advs.202100707. URL http://dx.doi.org/10.1002/advs.202100707. 6 Felix Mölder, Kim Philipp Jablonski, Brice Letcher, Michael B. Hall, Christo- pher H. Tomkins-Tinch, Vanessa Sochat, Jan Forster, Soohyun Lee, Sven O. Twardziok, Alexander Kanitz, Andreas Wilm, Manuel Holtgrewe, Sven Rah- mann, Sven Nahnsen, and Johannes Köster. S...

work page doi:10.1002/advs.202100707

[3] [3]

F1000Research (2021)

ISSN 2046-1402. doi: 10.12688/f1000research.29032.2. URL http://dx.doi.org/10.12688/f1000res earch.29032.2. Bo Wang, Zhiguang Chen, and Nong Xiao. A survey of system scheduling for hpc and big data. In Proceedings of the 2020 4th International Conference on High Performance Compilation, Computing and Communications , HP3C 2020, page 178–183. ACM, June

work page doi:10.12688/f1000research.29032.2 2046

[4] [4]

URL http://dx.doi.org/10.1145/3407947.3407977

doi: 10.1145/3407947.3407977. URL http://dx.doi.org/10.1145/3407947.3407977. Andy B. Yoo, Morris A. Jette, and Mark Grondona. Slurm: Simple linux utility for resource management. In Job Scheduling Strategies for Parallel Processing , page 44–60. Springer Berlin Heidelberg,

work page doi:10.1145/3407947.3407977

[5] [5]

Jette, C

ISBN 9783540397274. doi: 10.1007/10968987_3. URL http://dx.doi.org/10.1007/10968987_3. 7

work page doi:10.1007/10968987_3