Customers choose SchedMD Slurm for:

• – Massive scalability to handle performance requirements for large cluster and Exascale supercomputer needs spanning HPC and AI workloads with proven reliability
• – First-class resource management for GPUs and ultra granular task allocation by specialized resources (core, GPUs, threads, etc.)
• – Industry-leading workload throughput that supports more than 12,000 jobs per minute, 17 million jobs per day, and 120 million jobs per week
• – Advanced policies, scheduling algorithms, and queues/partitions management that simplifies and optimizes management of complex workload mix to meet project and organizational priorities
• – Optimize utilization to get more value from existing HPC investments
• – Agile innovation and integration to meet needs, driven by a strong open source community and partnerships

SchedMD Slurm Workload Manager Support

SchedMD is the core developer and service provider for Slurm, offering support, consulting, configuration, development, and training services that accelerate workload results through proven best practices and innovation. SchedMD offers five to ten times more experience with complex HPC and AI scheduling, including experience with half of the largest systems in the TOP500. This expertise helps optimize speed, throughput, and resource utilization for each unique workload mix so organizations can achieve more results, faster and more easily. SchedMD expert services help organizations implement Slurm quickly, maximize throughput, manage complexity, and scale high-performance workloads on HPE HPC solutions.

Customers add SchedMD Slurm support for:

• – Consulting and implementation expertise that speeds custom configuration tuning to increase throughput and utilization efficiency on complex and large-scale systems
• – Support services that ensure cluster workload management configuration continually processes workloads at peak levels as mix and scale evolves for improved productivity
• – Unique HPC workload expertise that spans Slurm development, software & computer engineering, and systems administration for resolution speed and quality, without escalation delays
• – Best practices for optimal workload performance from an expert team with complex HPC scheduling experience, including half of the biggest systems in the TOP500
• – Slurm Proof of Concept plan to assist with seamless and smooth site migration. Full access to SchedMD support with a dedicated engineer and configuration review
• – Tailored Slurm expert training that empowers users on harnessing Slurm capabilities with HPE exascale technologies to speed projects and increase adoption
• – Cloud workload management expertise and proven best practices to ensure optimal performance and streamlined management for workloads across on-prem systems, public and private clouds

Distribution Media & Software Documentation

Slurm Workload Manager is available for download as free, open-source software, together with documentation and installation guides, at slurm.schedmd.com.

Customers may purchase SchedMD Slurm Workload Manager Support through HPE as part of their HPC solution. After order delivery, HPE Services will receive confirmation of SchedMD support access and electronic instructions for submitting support requests on the customer’s behalf. HPE Services must visit the specified SchedMD website to complete the support account setup and add HPE and customer users. Customers can find additional details about SchedMD Slurm support services at https://www.schedmd.com/services.php.

Slurm Workload Management for HPE HPC Systems

Slurm is part of the HPE HPC software portfolio and provides an open-source, agile, and modern option for highly scalable, fault-tolerant workload management, including policies, algorithms, and reporting. These capabilities help HPE customers process jobs faster, optimize utilization of specialized HPE system resources, and increase throughput in alignment with organizational priorities. Slurm delivers optimized workload performance for HPE Cray supercomputers, HPE Apollo density-optimized systems purpose-built for HPC and AI workloads, and HPE GreenLake cloud services.

Slurm capabilities for HPE HPC systems include:

• − Allocates and optimizes exclusive and nonexclusive access to granular resources: Slurm allocates resources to users and jobs for the duration of each workload. Supported controls include network topology, fair-share scheduling, advanced reservations, preemption, resource limits, and accounting factors.
• − Provides a framework for starting, executing, and monitoring workloads on allocated nodes or GPUs: Slurm capabilities include real-time task-level accounting, power-consumption monitoring, API usage tracking, and automatic job requeuing.
• − Simplifies management and arbitrates contention for resources: Slurm manages a queue of pending work according to organizational and work priorities.
• − Accelerates work processing and throughput: Slurm provides extensive policies and algorithms, including support for high-throughput computing workloads and thousands of job submissions per second.
• − Supports elastic computing and cloud bursting: Slurm natively supports capabilities that allow workloads to expand into cloud resources when additional capacity is required.

SchedMD Slurm support service capabilities for HPE HPC systems include:

• − Level 3 Support: High-performance systems must perform at high utilization and performance to meet end users and management return on the investment expectations. Customers covered by a support contract can reach out to SchedMD Slurm experts to help resolve complex workload management issues and questions quicker, including detailed answers directly from the Slurm Development team, instead of taking weeks or even months to try to resolve them in-house.
• − Configuration assistance: Valuable assistance when the customer system is initially being configured to use Slurm or being modified as requirements change. Customers can review cluster requirements, operating environment, and organizational goals with a Slurm engineer who will work with them to optimize the configuration to achieve their needs.
• − Optional tailored Slurm training: SchedMD can provide customized Slurm training for HPE customers. HPE representatives or customers can contact Jess Arrington at SchedMD at jess@schedmd.com to request a training quote. A customer scoping call before remote or on-site training helps ensure that the hands-on lab workshop addresses specific use cases and requirements. The in-depth technical training covers Slurm best practices, site-specific use cases, and configuration optimization.

License Management in Slurm

Slurm can help manage software licenses by assigning available licenses to jobs at scheduling time. Administrators can dynamically modify the license count to synchronize it with license server availability. If licenses are unavailable, jobs remain pending until licenses become available. In Slurm, licenses are shared resources: they are configured for the entire cluster rather than tied to a specific host. Licenses can also be configured as preemptible resources.

Licenses in Slurm can be configured in two ways:

• – Local Licenses: Local licenses are local to the cluster using the slurm.conf in which they are configured.
• – Remote Licenses: Remote licenses are served by the database and are configured using the sacctmgr command. Remote licenses are dynamic in nature as upon running the sacctmgr command, the slurmdbd updates all clusters the licenses are assigned to.

Operating System Compatibility

Slurm Workload Manager has been thoroughly tested on popular Linux distributions using Arm64 (AArch64), ppc64, and x86-64 architectures. Some features are limited to recent releases and newer Linux kernel versions. Supported distributions include:

• – Cray Linux Environment 6
• – Cray Linux Environment 7
• – Debian (jessie, stretch, buster, and newer)
• – RedHat Enterprise Linux 6 (RHEL6), CentOS 6, Scientific Linux 6 (deprecated, and will be removed in a future release)
• – RedHat Enterprise Linux 7 (RHEL7), CentOS 7, Scientific Linux 7
• – Red Hat Enterprise Linux 8 (RHEL 8), CentOS 8, and newer versions
• – SUSE Linux Enterprise Server (SLES) 12 (and newer)
• – Ubuntu (16.04, 18.04, and newer)

Support

Support for SchedMD Slurm is provided by a combination of HPE Services and direct-to-vendor support. Contact HPE Services for assistance and SchedMD will partner with them, as needed, to deliver expedient resolution.

For the most up-to-date information on HPE Services, please refer to the HPE Services – Supplemental QuickSpecs, which provides a comprehensive and regularly updated overview of available services.

SchedMD Slurm support is configured according to the number of sockets in the system. The first tier, 1–100 sockets, is a flat-rate tier: the fee is the same whether support is purchased for one socket or 100 sockets. For systems with up to 100 sockets, configure a quantity of 1 for the applicable SKU.

All other tiers are based on a per socket basis being charged a unit price per the total number of sockets in the system. Note that in a configuration the total sockets sold for a deal, even if they are sold across multiple systems, should determine the tier to use for the correct SKU.

Example: Single customer with two main system deployments (760 sockets each) and a test and development system (48 sockets) to be supported for four years. Configure 1,568 of SKU S1R70AAE.

Have feedback on QuickSpecs? We’re listening