SLURM: Simple Linux Utility for Resource Management

By A. Yoo, Morris A. Jette, Mark Grondona. Job Scheduling Strategies for Parallel Processing, 2003

1,811 citations recorded by the source corpus.

Subfields: Computer Networks and Communications, Hardware and Architecture

View Paper: https://www.semanticscholar.org/paper/e3b958f93eb28d8cb6780480501a414c11361206

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Computer Networks and Communications

In Hardware and Architecture

Citations in the corpus

HPEPDOCK: a web server for blind peptide–protein docking based on a hierarchical algorithm (2018, methodology): “The SLURM Workload Manager (39) is used as the job scheduler of HPEPDOCK server.”
Parallel Job Scheduling - A Status Report (2004, methodology): “In many Linux clusters SLURM (Simple Linux Utility for Resource Management) is especially used for low priority jobs [51].”
Fiber: A Platform for Efficient Development and Distributed Training for Reinforcement Learning and Population-Based Methods (2020, background): “Examples include Apache Mesos (Hindman et al. 2011), Kubernetes (Burns et al. 2016), Uber Peloton (Cai and Bansal 2019) and Slurm (Yoo, Jette, and Grondona 2003).”
GreenHadoop: leveraging green energy in data-processing frameworks (2012, background): “[8, 32], GreenHadoop does not have information about the jobs’ desired number of servers or expected running times, from which we could estimate their energy needs.”
CaImAn an open source tool for scalable calcium imaging data analysis (2019, methodology): “Unlike previous work (Freeman et al., 2014), CAIMAN BATCH assumes minimal computational infrastructure (down to a standard laptop computer), is not tied to a particular parallel computation framework, and is compatible with HPC scheduling systems like SLURM (Yoo et al., 2003).”
ZCCloud: Exploring Wasted Green Power for High-Performance Computing (2016, background): “Batch enables the schedulers to schedule processors, memory, and sometimes network and input/output to both minimize interapplication interference and maximize resource utilization [26], [27].”
Sparrow: distributed, low latency scheduling (2013, background): “, Mesos [8], YARN [16], SLURM [10]); meeting this design requirement would be difficult with a design that schedules and launches all tasks through a single node.”
Image Matching Across Wide Baselines: From Paper to Practice (2020, methodology): “3 The implementation relies on SLURM (Yoo et al. 2003) for scalable job scheduling, which is compatible with our supercomputer clusters—we also provide on-the-cloud, ready-to-go images.”
GreenSlot: Scheduling energy consumption in green datacenters (2011, methodology): “GreenSlot extends the popular SLURM scheduler for Linux with roughly 2800 uncommented lines of Python code.”
CPR: Understanding and Improving Failure Tolerant Training for Deep Learning Recommendation with Partial Recovery (2020, background): “Rescheduling overhead is the time spent for the cluster scheduler to ﬁnd alternative, available nodes to take over the role of the failed nodes (Basney & Livny, 2000; Yoo et al., 2003).”
The PATRIC Bioinformatics Resource Center: expanding data and analysis capabilities (2019, background): “The Genome Assembly Service has been rebuilt with a new job scheduler that enables a fairer jobqueuing process that prevents large jobs from creating bottlenecks (14).”
Autopilot: workload autoscaling at Google (2020, background): “In a somewhat different context, schedulers for HPC systems, such as Slurm [37], require each batch job to specify the number of machines.”
proovread: large-scale high-accuracy PacBio correction through iterative short read consensus (2014, background): “The queueing system Slurm was used to distribute jobs.”