Shared queues allow multiple users to run jobs on the same node simultaneously. Proper resource management is crucial to avoid exceeding memory or CPU limits.

Shared Queues Overview

Shared queues are designed for users who do not need exclusive access to a node but still require compute resources. You must explicitly request memory to avoid exceeding available resources.

Available Shared Queues

CPU Nodes

Queue	CPU Architecture	Vector/Matrix Extension	CPU Cores per Node	Node Memory	Default Runtime	Max Runtime	Max Nodes
short-40core-shared	Intel Skylake	AVX512	40	192 GB	1 hr	4 hrs	4
long-40core-shared	Intel Skylake	AVX512	40	192 GB	8 hrs	24 hrs	3
extended-40core-shared	Intel Skylake	AVX512	40	192 GB	8 hrs	3.5 days	1
short-96core-shared	AMD EPYC Milan	AVX2	96	256 GB	1 hr	4 hrs	4
long-96core-shared	AMD EPYC Milan	AVX2	96	256 GB	8 hrs	24 hrs	3
extended-96core-shared	AMD EPYC Milan	AVX2	96	256 GB	8 hrs	3.5 days	1

GPU Nodes

Queue	CPU Architecture	Vector/Matrix Extension	CPU Cores per Node	GPUs per Node	Node Memory	Default Runtime	Max Runtime	Max Nodes	Max Simultaneous Jobs per User
a100	Intel Ice Lake	AVX512, DL Boost	64	4	256 GB	1 hr	8 hrs	2	2
a100-long	Intel Ice Lake	AVX512, DL Boost	64	4	256 GB	8 hrs	2 days	1	2
a100-large	Intel Ice Lake	AVX512, DL Boost	64	4	256 GB	1 hr	8 hrs	4	1

Instructions for Using Shared Queues

• Request memory explicitly: Use #SBATCH --mem=[amount] to avoid exceeding available memory on a shared node.
• Check resource limits: Know the total node memory and CPU cores and plan your job accordingly.
• Monitor usage: Use squeue -u $USER to check your jobs and top or htop to see memory/CPU usage.
• Benefit: Shared queues can have shorter wait times than exclusive nodes if you request exactly the resources you need.