Kellogg Linux Cluster#
The Kellogg Linux Cluster (KLC) is a set of high-memory Linux servers available to Kellogg researchers for interactive computing, large-scale data analysis, and batch jobs. KLC is part of Northwestern’s Quest cluster, sharing its file systems and software module library. You can submit a SLURM script from a KLC node to run jobs on Quest, but everyday KLC work (interactive sessions, scripts, batch jobs) doesn’t require SLURM at all. Each node has 1.5–2 TB of RAM and up to 64 CPU cores.
The Getting Started guide walks you through getting an account, connecting, setting up your environment, and running your first job.
The User Guide covers all KLC topics in depth: connecting, storing data, transferring files, scheduling jobs, accessing GPUs, and more.
KLC Reserve lets you use the SLURM job scheduler to reserve resources beyond what shared KLC main nodes provide — more than 24 cores on a single node, high-memory nodes, GPU nodes, or job arrays.
KLC and KLC Reserve#
KLC Reserve uses the same Northwestern NetID, KLC account, and Kellogg file systems as standard KLC login access. Reserve is where you submit SLURM jobs to the kellogg partition from any KLC node to access a KLC GPU node, high memory node, or full-node core counts that are not available on shared login nodes.
For the decision between direct login and scheduled jobs, see When to Use KLC Reserve. For Reserve overview, resource specs, and your first submission, see Kellogg Linux Cluster Reserve.
Quest Documentation#
KLC is part of Quest. Northwestern RCDS maintains comprehensive documentation for Quest that applies to KLC as well:
Quest User Guide — connecting, storage, data transfer, software modules, SLURM, GPUs
GPUs on Quest — General Access GPU jobs on the
gengpupartition