Jump to contentJump to page navigation: previous page [access key p]/next page [access key n]
SUSE Linux Enterprise for High-Performance Computing 15 SP5

Release Notes

Abstract

SUSE Linux Enterprise for High-Performance Computing is a highly-scalable, high-performance open-source operating system designed to utilize the power of parallel computing. This document provides an overview of high-level general features, capabilities, and limitations of SUSE Linux Enterprise for High-Performance Computing 15 SP5 and important product updates.

These release notes are updated periodically. The latest version of these release notes is always available at https://www.suse.com/releasenotes. General documentation can be found at https://documentation.suse.com/sle-hpc/15-SP4.

Publication Date: 2023-07-10, Version: 15.500000000.20230710

1 About the release notes

These Release Notes are identical across all architectures, and the most recent version is always available online at https://www.suse.com/releasenotes.

Entries are only listed once but they can be referenced in several places if they are important and belong to more than one section.

Release notes usually only list changes that happened between two subsequent releases. Certain important entries from the release notes of previous product versions are repeated. To make these entries easier to identify, they contain a note to that effect.

However, repeated entries are provided as a courtesy only. Therefore, if you are skipping one or more service packs, check the release notes of the skipped service packs as well. If you are only reading the release notes of the current release, you could miss important changes.

2 SUSE Linux Enterprise for High-Performance Computing

SUSE Linux Enterprise for High-Performance Computing is a highly scalable, high performance open-source operating system designed to utilize the power of parallel computing for modeling, simulation and advanced analytics workloads.

SUSE Linux Enterprise for High-Performance Computing 15 SP5 provides tools and libraries related to High Performance Computing. This includes:

  • Workload manager

  • Remote and parallel shells

  • Performance monitoring and measuring tools

  • Serial console monitoring tool

  • Cluster power management tool

  • A tool for discovering the machine hardware topology

  • System monitoring

  • A tool for monitoring memory errors

  • A tool for determining the CPU model and its capabilities (x86-64 only)

  • User-extensible heap manager capable of distinguishing between different kinds of memory (x86-64 only)

  • Serial and parallel computational libraries providing the common standards BLAS, LAPACK, …​

  • Various MPI implementations

  • Serial and parallel libraries for the HDF5 file format

2.1 Hardware Platform Support

SUSE Linux Enterprise for High-Performance Computing 15 SP5 is available for the Intel 64/AMD64 (x86-64) and AArch64 platforms.

2.2 Important Sections of This Document

If you are upgrading from a previous SUSE Linux Enterprise for High-Performance Computing release, you should review at least the following sections:

2.3 Support and life cycle

SUSE Linux Enterprise for High-Performance Computing is backed by award-winning support from SUSE, an established technology leader with a proven history of delivering enterprise-quality support services.

SUSE Linux Enterprise for High-Performance Computing 15 has a 13-year life cycle, with 10 years of General Support and 3 years of Extended Support. The current version (SP5) will be fully maintained and supported until 6 months after the release of SUSE Linux Enterprise for High-Performance Computing 15 SP6.

Any release package is fully maintained and supported until the availability of the next release.

Extended Service Pack Overlay Support (ESPOS) and Long Term Service Pack Support (LTSS) are also available for this product. If you need additional time to design, validate and test your upgrade plans, Long Term Service Pack Support (LTSS) can extend the support you get by an additional 12 to 36 months in 12-month increments, providing a total of 3 to 5 years of support on any given Service Pack.

For more information, see:

2.4 Support statement for SUSE Linux Enterprise for High-Performance Computing

To receive support, you need an appropriate subscription with SUSE. For more information, see https://www.suse.com/support/programs/subscriptions/?id=SUSE_Linux_Enterprise_Server.

The following definitions apply:

L1

Problem determination, which means technical support designed to provide compatibility information, usage support, ongoing maintenance, information gathering and basic troubleshooting using available documentation.

L2

Problem isolation, which means technical support designed to analyze data, reproduce customer problems, isolate problem area and provide a resolution for problems not resolved by Level 1 or prepare for Level 3.

L3

Problem resolution, which means technical support designed to resolve problems by engaging engineering to resolve product defects which have been identified by Level 2 Support.

For contracted customers and partners, SUSE Linux Enterprise for High-Performance Computing is delivered with L3 support for all packages, except for the following:

SUSE will only support the usage of original packages. That is, packages that are unchanged and not recompiled.

2.4.1 Software requiring specific contracts

Certain software delivered as part of SUSE Linux Enterprise for High-Performance Computing may require an external contract. Check the support status of individual packages using the RPM metadata that can be viewed with rpm, zypper, or YaST.

2.4.2 Software under GNU AGPL

SUSE Linux Enterprise for High-Performance Computing 15 SP5 (and the SUSE Linux Enterprise modules) includes the following software that is shipped only under a GNU AGPL software license:

  • Ghostscript (including subpackages)

SUSE Linux Enterprise for High-Performance Computing 15 SP5 (and the SUSE Linux Enterprise modules) includes the following software that is shipped under multiple licenses that include a GNU AGPL software license:

  • MySpell dictionaries and LightProof

  • ArgyllCMS

2.5 Documentation and other information

2.5.1 Available on the product media

  • Read the READMEs on the media.

  • Get the detailed change log information about a particular package from the RPM (where FILENAME.rpm is the name of the RPM):

    rpm --changelog -qp FILENAME.rpm
  • Check the ChangeLog file in the top level of the installation medium for a chronological log of all changes made to the updated packages.

  • Find more information in the docu directory of the installation medium of SUSE Linux Enterprise for High-Performance Computing 15 SP5. This directory includes PDF versions of the SUSE Linux Enterprise for High-Performance Computing 15 SP5 Installation Quick Start Guide.

2.5.2 Online documentation

4 Technology previews

Technology previews are packages, stacks, or features delivered by SUSE which are not supported. They may be functionally incomplete, unstable or in other ways not suitable for production use. They are included for your convenience and give you a chance to test new technologies within an enterprise environment.

Whether a technology preview becomes a fully supported technology later depends on customer and market feedback. Technology previews can be dropped at any time and SUSE does not commit to providing a supported version of such technologies in the future.

Give your SUSE representative feedback about technology previews, including your experience and use case.

4.1 64K page size kernel flavor has been added

SUSE Linux Enterprise for High-Performance Computing for Arm 12 SP2 and later kernels have used a page size of 4K. This offers the widest compatibility also for small systems with little RAM, allowing to use Transparent Huge Pages (THP) where large pages make sense.

As a technology preview, SUSE Linux Enterprise for High-Performance Computing for Arm 15 SP5 adds a kernel flavor 64kb, offering a page size of 64 KiB and physical/virtual address size of 52 bits. Same as the default kernel flavor, it does not use preemption.

Main purpose at this time is to allow for side-by-side benchmarking for High Performance Computing, Machine Learning and other Big Data use cases. Contact your SUSE representative if you notice performance gains for your specific workloads.

Important
Important: Swap needs to be re-initialized

After booting the 64K kernel, any swap partitions need to re-initialized to be usable. To do this, run the swapon command with the --fixpgsz parameter on the swap partition. Note that this process deletes data present in the swap partition (for example, suspend data). In this example, the swap partition is on /dev/sdc1:

swapon --fixpgsz /dev/sdc1
Important
Important: Btrfs file system uses page size as block size

It is currently not possible to use Btrfs file systems across page sizes. Block sizes below page size are not yet supported and block sizes above page size might never be supported.

During installation, change the default partitioning proposal and choose another file system, such as Ext4 or XFS, to allow rebooting from the default 4K page size kernel of the Installer into kernel-64kb and back.

See the Storage Guide for a discussion of supported file systems.

Warning
Warning: RAID 5 uses page size as stripe size

It is currently not yet possible to configure stripe size on volume creation. This will lead to sub-optimal performance if page size and block size differ.

Avoid RAID 5 volumes when benchmarking 64K vs. 4K page size kernels.

See the Storage Guide for more information on software RAID.

Note
Note: Cross-architecture compatibility considerations

The SUSE Linux Enterprise for High-Performance Computing 15 SP5 kernels on x86-64 use 4K page size.

The SUSE Linux Enterprise for High-Performance Computing for POWER 15 SP5 kernel uses 64K page size.

5 Modules

5.1 HPC module

The HPC module contains HPC specific packages. These include the workload manager Slurm, the node deployment tool clustduct, munge for user authentication, the remote shell mrsh, the parallel shell pdsh, as well as numerous HPC libraries and frameworks.

This module is available with the SUSE Linux Enterprise for High-Performance Computing only. It is selected by default during the installation. It can be added or removed using the YaST UI or the SUSEConnect CLI tool. Refer to the system administration guide for further details.

5.2 NVIDIA Compute Module

The NVIDIA Compute Module provides the NVIDIA CUDA repository for SUSE Linux Enterprise 15. Note that that any software within this repository is under a 3rd party EULA. For more information check https://docs.nvidia.com/cuda/eula/index.html.

This module is not selected for addition by default when installing SUSE Linux Enterprise for High-Performance Computing. It may be selected manually during installation from the Extension and Modules screen. You may also select it on an installed system using YaST. To do so, run from a shell as root yast registration, select: Select Extensions and search for NVIDIA Compute Module and press Next.

Important
Important

Do not attempt to add this module with the SUSEConnect CLI tool. This tool is not yet capable of handling 3rd party repositories.

Once you have selected this module you will be asked to confirm the 3rd party license and verify the repository signing key.

6 Changes affecting all architectures

Information in this section applies to all architectures supported by SUSE Linux Enterprise for High-Performance Computing 15 SP5.

6.1 Enriched system visibility in the SUSE Customer Center (SCC)

SUSE is committed to helping provide better insights into the consumption of SUSE subscriptions regardless of where they are running or how they are managed; physical or virtual, on-prem or in the cloud, connected to SCC or Repository Mirroring Tool (RMT), or managed by SUSE Manager. To help you identify or filter out systems in SCC that are no longer running or decommissioned, SUSEConnect now features a daily “ping”, which will update system information automatically.

For more details see the documentation at https://documentation.suse.com/subscription/suseconnect/single-html/SLE-suseconnect-visibility/.

6.2 Automatically opened ports

Installing the following packages automatically opens the following ports:

  • dolly - TCP ports 9997 and 9998

  • slurm - TCP ports 6817, 6818, and 6819

Important
Important

These release notes only document changes in SUSE Linux Enterprise for High-Performance Computing compared to the immediate previous service pack of SUSE Linux Enterprise for High-Performance Computing. The full changes and fixes can be found on the respective web site of the packages.

6.3 GNU compiler suite version 12

SLE HPC 15 SP5 now supports the GNU compiler suite version 12. To install the runtime environments (environment modules) for version 12, run: zypper install gnu12-compilers-hpc. To install all the packages required for development (C, C and Fortran compilers), run `zypper install gnu12-compilers-hpc-devel`. To load the environment, run 'module load gnu/12' in your shell. When the `-devel` package is installed, the compilers (`gcc-12`, `g-12`, gfortran-12) will become available in this shell under their standard names (gcc, g++, gfortran).

6.4 conman

conman has been updated to version 0.3.1:

  • Fixed username/password use in libipmiconsole.conf.

  • Added -T command-line option to specify terminal emulator.

  • General move of files from /usr/lib/conman to /usr/share/conman.

6.5 dolly

dolly has been updated to version 0.64.2. The tool is less verbose by default and the dolly service can be activated through a socket.

6.6 imb

imb is shipped in version 2021.3:

  • Change default value for mem_alloc_type to device.

  • Added new IMB-MPI1-GPU benchmarks as technical preview.

  • Added -msg_pause option.

  • Changed default window_size from 64 to 256.

  • Added -window_size option for IMB-MPI1.

6.7 memkind

memkind has been updated to version 1.14.0. The full list of changes is available at http://memkind.github.io/memkind/.

6.8 openblas

openblas has been updated to version 0.3.21. It contains performance regression fixes and optimizations. For more information see https://github.com/xianyi/OpenBLAS/releases/tag/v0.3.21.

6.9 munge

munge has been updated to version 0.5.15:

  • Fixed systemd service unit configuration to wait until network is online.

  • Fixed sending repeated SIGTERMs to signal stop.

6.10 cpuid

cpuid has been updated to version 20221201.

This update includes:

  • Added and updated identification of many CPU models and variants.

  • Updated hypervisor support.

  • Improved synth information and u-architecture decoding.

6.11 lmod

lmod has been updated to version 8.7.17.

The user visible changes include:

  • Add option --miniConfig to report configuration differences from default.

  • Move cache file location from ~/.lmod.d/.cache/* to ~/.cache/lmod/*

  • Transitional support for using ~/.config/lmod for collections. Currently collect are written to both ~/.lmod.d/ and ~/.config/lmod.

  • setenv and pushenv change local environment when running spider (and avail).

  • Allow bash users to export SUPPORT_KSH=no so that they can avoid bash startup setting FPATH

  • Add --location option to show to write to stderr the file location.

  • Only rebuild spider caches if there are any loaded or pending modules. module avail <name1> <name2> …​ now only prints matching aliases. Search names are resolved.

  • Print dataT table when there is an Exception.

  • New command added: module overview.

  • Add spiderPathFilter hook so that sites can control what paths are kept or ignored.

  • Added $LMOD_SITE_MODULEPATH support to prepend to MODULEPATH

  • Add support for sh_to_modulefile to support zsh, ksh, bash and tcsh with aliases and shell functions

  • Support for source_sh added. Now support more than one shell script per modulefile.

6.12 PAPI

PAPI has been updated to version 7.0.0.

The highlights include:

  • Added "intel_gpu" component with monitoring capabilities support for Intel GPUs, including GPU hardware events and memory performance metrics.

  • Added "sysdetect" component for detecting a machine’s architectural details. Additionally, PAPI offers a new API that enables users to get "sysdetect" details from within their application.

  • A major redesign of the "rocm" component for advanced monitoring features for the latest AMD GPUs. The PAPI "rocm" component is now thread-safe.

  • Support for NVIDIA compute capability 7.0 and greater. This implies support for CUPTI’s new Profiling and Perfworks APIs.

  • Significant redesign of the "sde" component into two separate entities:

    1. a standalone library "libsde" with a new API for software developers to define software-based metrics from within their applications

    2. the PAPI "sde" component that enables monitoring of these new software-based events.

  • New C++ interface for "libsde," which enables software developers to define software-defined events from within their C++ applications.

  • New Counter Analysis Toolkit (CAT) benchmarks and refinements of PAPI’s CAT data analysis.

  • Support for FUGAKU’s A64FX Arm architecture, including monitoring capabilities for memory bandwidth and other node-wide metrics. For futher details check https://bitbucket.org/icl/papi/wiki/PAPI-Releases.md

6.13 warewulf4

warewulf4 is a popular SLE for HPC deployment tool whose latest version is a full rewrite in the Go programming language. It is applying lessons learned from its predecessors. It deploys minimal images which it obtains from container images stored in a registry and performs a minimal configuration for the image to be useful as a compute node image in a cluster. warewulf4 is deprecating the former deployment tool clustduct.

6.14 Creating containers from current HPC environment

Usually users use environment modules to adjust their environment (that is, environment variables like PATH, LD_LIBRARY_PATH, MANPATH etc.) to pick exactly the tools and libraries they need for their work. The same can be achieved with containers by including only those components in a container that are part of this environment. This functionality is now provided using the spack and singularity applications.

6.15 Spack

6.15.1 v0.19.1

6.15.1.1 Spack Bugfixes
  • buildcache create: make file exists less verbose

  • spack mirror create: don’t change paths to urls

  • Improve error message for requirements

  • uninstall: fix accidental cubic complexity

  • scons: fix signature for install_args

  • Fix combine_phase_logs text encoding issues

  • Use a module-like object to propagate changes in the MRO, when setting build env

  • PackageBase should not define builder legacy attributes

  • Forward lookup of the run_tests attribute

  • Bugfix for timers

  • Fix path handling in prefix inspections

  • Fix libtool filter for Fujitsu compilers

  • FileCache: delete the new cache file on exception

  • Propagate exceptions from Spack python console

  • Tests: Fix a bug/typo in a config_values.py fixture

  • Various CI fixes

  • Docs: remove monitors and analyzers, typos

  • bump release version for tutorial command

6.15.2 v0.19.0

v0.19.0 is a major feature release.

6.15.2.1 Major features in this release
  1. Package requirements

    Spack’s traditional package preferences are soft, but we’ve added hard requriements to packages.yaml and spack.yaml . Package requirements use the same syntax as specs:

    packages:
      libfabric:
        require: "@1.13.2"
      mpich:
        require:
        - one_of: ["+cuda", "+rocm"]

    More details in the docs.

  2. Environment UI Improvements

    • Fewer surprising modifications to spack.yaml :

      • spack install in an environment will no longer add to the specs: list; you’ll need to either use spack add <spec> or spack install --add <spec>.

      • Similarly, spack uninstall will not remove from your environment’s specs: list; you’ll need to use spack remove or spack uninstall --remove.

        This will make it easier to manage an environment, as there is clear separation between the stack to be installed (spack.yaml/spack.lock) and which parts of it should be installed (spack install / spack uninstall).

    • concretizer:unify:true is now the default mode for new environments

      We see more users creating unify:true environments now. Users who need unify:false can add it to their environment to get the old behavior. This will concretize every spec in the environment independently.

    • Include environment configuration from URLs ( docs)

      You can now include configuration in your environment directly from a URL:

      spack:
        include:
        - https://github.com/path/to/raw/config/compilers.yaml
  3. Compiler and variant propagation

    Currently, compiler flags and variants are inconsistent: compiler flags set for a package are inherited by its dependencies, while variants are not. We should have these be consistent by allowing for inheritance to be enabled or disabled for both variants and compiler flags.

    Example syntax: * package ++variant: enabled variant that will be propagated to dependencies * package +variant: enabled variant that will NOT be propagated to dependencies * package ~~variant: disabled variant that will be propagated to dependencies * package ~variant: disabled variant that will NOT be propagated to dependencies * package cflags==-g: cflags will be propagated to dependencies * package cflags=-g: cflags will NOT be propagated to dependencies

    + Syntax for non-boolan variants is similar to compiler flags. More in the docs for variants and compiler flags.

  4. Enhancements to git version specifiers

    • v0.18.0 added the ability to use git commits as versions. You can now use the git. prefix to specify git tags or branches as versions. All of these are valid git versions in v0.19 :

      foo@abcdef1234abcdef1234abcdef1234abcdef1234      # raw commit
      foo@git.abcdef1234abcdef1234abcdef1234abcdef1234  # commit with git prefix
      foo@git.develop                                   # the develop branch
      foo@git.0.19                                      # use the 0.19 tag
    • v0.19 also gives you more control over how Spack interprets git versions, in case Spack cannot detect the version from the git repository. You can suffix a git version with =<version> to force Spack to concretize it as a particular version :

      # use mybranch, but treat it as version 3.2 for version comparison
      foo@git.mybranch=3.2
      
      # use the given commit, but treat it as develop for version comparison
      foo@git.abcdef1234abcdef1234abcdef1234abcdef1234=develop

      More in the docs

  5. Changes to Cray EX Support

    Cray machines have historically had their own platform within Spack, because we needed to go through the module system to leverage compilers and MPI installations on these machines. The Cray EX programming environment now provides standalone craycc executables and proper mpicc wrappers, so Spack can treat EX machines like Linux with extra packages .

    We expect this to greatly reduce bugs, as external packages and compilers can now be used by prefix instead of through modules. We will also no longer be subject to reproducibility issues when modules change from Cray PE release to release and from site to site. This also simplifies dealing with the underlying Linux OS on cray systems, as Spack will properly model the machine’s OS as either SuSE or RHEL.

  6. Improvements to tests and testing in CI

    • spack ci generate --tests will generate a .gitlab-ci.yml file that not only does builds but also runs tests for built packages . Public GitHub pipelines now also run tests in CI.

    • spack test run --explicit will only run tests for packages that are explicitly installed, instead of all packages.

  7. Experimental binding link model

    You can add a new option to config.yaml to make Spack embed absolute paths to needed shared libraries in ELF executables and shared libraries on Linux ( docs):

    config:
      shared_linking:
        type: rpath
        bind: true

    This can improve launch time at scale for parallel applications, and it can make installations less susceptible to environment variables like LD_LIBRARY_PATH, even especially when dealing with external libraries that use RUNPATH. You can think of this as a faster, even higher-precedence version of RPATH.

6.15.2.2 Other new features of note
  • spack spec prints dependencies more legibly. Dependencies in the output now appear at the earliest level of indentation possible

  • You can override package.py attributes like url, directly in packages.yaml ( docs)

  • There are a number of new architecture-related format strings you can use in Spack configuration files to specify paths ( docs)

6.15.2.3 Performance Improvements
  • Major performance improvements for installation from binary caches

  • Test suite can now be parallelized using xdist (used in GitHub Actions)

  • Reduce lock contention for parallel builds in environments

6.15.2.4 New binary caches and stacks
  • We now build nearly all of E4S with oneapi in our buildcache

  • Added 3 new machine learning-centric stacks to binary cache: x86_64_v3, CUDA, ROCm

6.15.2.5 Removals and Deprecations
  • Support for Python 3.5 is dropped . Only Python 2.7 and 3.6+ are officially supported.

  • This is the last Spack release that will support Python 2 . Spack v0.19 will emit a deprecation warning if you run it with Python 2, and Python 2 support will soon be removed from the develop branch.

  • LD_LIBRARY_PATH is no longer set by default by spack load or module loads.

    Setting LD_LIBRARY_PATH in Spack environments/modules can cause binaries from outside of Spack to crash, and Spack’s own builds use RPATH and do not need LD_LIBRARY_PATH set in order to run. If you still want the old behavior, you can run these commands to configure Spack to set LD_LIBRARY_PATH:

    spack config add modules:prefix_inspections:lib64:[LD_LIBRARY_PATH]
    spack config add modules:prefix_inspections:lib:[LD_LIBRARY_PATH]
  • The spack:concretization:[together|separately] has been removed after being deprecated in v0.18. Use concretizer:unify:[true|false].

  • config:module_roots is no longer supported after being deprecated in v0.18. Use configuration in module sets instead ( docs).

  • spack activate and spack deactivate are no longer supported, having been deprecated in v0.18. Use an environment with a view instead of activating/deactivating (docs).

  • The old YAML format for buildcaches is now deprecated . If you are using an old buildcache with YAML metadata you will need to regenerate it with JSON metadata.

  • spack bootstrap trust and spack bootstrap untrust are deprecated in favor of spack bootstrap enable and spack bootstrap disable and will be removed in v0.20.

  • The graviton2 architecture has been renamed to neoverse_n1, and graviton3 is now neoverse_v1. Buildcaches using the old architecture names will need to be rebuilt.

  • The terms blacklist and whitelist have been replaced with include and exclude in all configuration files . You can use spack config update to automatically fix your configuration files.

6.15.2.6 Notable Bugfixes
  • Permission setting on installation now handles effective uid properly

  • buildable:true for an MPI implementation now overrides buildable:false for mpi

  • Improved error messages when attempting to use an unconfigured compiler

  • Do not punish explicitly requested compiler mismatches in the solver

  • spack stage: add missing –fresh and –reuse

  • Fixes for adding build system executables like cmake to package scope

  • Bugfix for binary relocation with aliased strings produced by newer binutils

6.15.3 v0.18.1

6.15.3.1 Spack Bugfixes
  • Fix several bugs related to bootstrapping

  • Fix a regression that was causing spec hashes to differ between Python 2 and Python 3

  • Fixed compiler flags for oneAPI and DPC++

  • Fixed several issues related to concretization

  • Improved support for Cray manifest file and spack external find

  • Assign a version to openSUSE Tumbleweed according to the GLIBC version in the system

  • Improved Dockerfile generation for spack containerize

  • Fixed a few bugs related to concurrent execution of commands

6.15.3.2 Package updates
  • WarpX: add v22.06, fixed libs property

  • openPMD: add v0.14.5, update recipe for @develop

6.15.4 v0.18.0

v0.18.0 is a major feature release.

6.15.4.1 Major features in this release
  1. Concretizer now reuses by default

    spack install --reuse was introduced in v0.17.0, and --reuse is now the default concretization mode. Spack will try hard to resolve dependencies using installed packages or binaries .

    To avoid reuse and to use the latest package configurations, (the old default), you can use spack install --fresh, or add configuration like this to your environment or concretizer.yaml:

    concretizer:
        reuse: false
  2. Finer-grained hashes

    Spack hashes now include link, run, and build dependencies, as well as a canonical hash of package recipes. Previously, hashes only included link and run dependencies (though build dependencies were stored by environments). We coarsened the hash to reduce churn in user installations, but the new default concretizer behavior mitigates this concern and gets us reuse and provenance. You will be able to see the build dependencies of new installations with spack find. Old installations will not change and their hashes will not be affected.

  3. Improved error messages

    Error handling with the new concretizer is now done with optimization criteria rather than with unsatisfiable cores, and Spack reports many more details about conflicting constraints.

  4. Unify environments when possible

    Environments have thus far supported concretization: together or concretization: separately. These have been replaced by a new preference in concretizer.yaml:

    concretizer:
        unify: [true|false|when_possible]

    concretizer:unify:when_possible will try to resolve a fully unified environment, but if it cannot, it will create multiple configurations of some packages where it has to. For large environments that previously had to be concretized separately, this can result in a huge speedup (40-50x).

  5. Automatically find externals on Cray machines

    Spack can now automatically discover installed packages in the Cray Programming Environment by running spack external find (or spack external read-cray-manifest to only query the PE). Packages from the PE (e.g., cray-mpich are added to the database with full dependency information, and compilers from the PE are added to compilers.yaml. Available with the June 2022 release of the Cray Programming Environment.

  6. New binary format and hardened signing

    Spack now has an updated binary format, with improvements for security. The new format has a detached signature file, and Spack verifies the signature before untarring or decompressing the binary package. The previous format embedded the signature in a tar file, which required the client to run tar before verifying . Spack can still install from build caches using the old format, but we encourage users to switch to the new format going forward.

    Production GitLab pipelines have been hardened to securely sign binaries. There is now a separate signing stage so that signing keys are never exposed to build system code, and signing keys are ephemeral and only live as long as the signing pipeline stage.

  7. Bootstrap mirror generation

    The spack bootstrap mirror command can automatically create a mirror for bootstrapping the concretizer and other needed dependencies in an air-gapped environment.

  8. Makefile generation

    spack env depfile can be used to generate a Makefile from an environment, which can be used to build packages the environment in parallel on a single node. e.g.:

    spack -e myenv env depfile > Makefile
    make

    Spack propagates gmake jobserver information to builds so that their jobs can share cores.

  9. New variant features

    In addition to being conditional themselves, variants can now have conditional values that are only possible for certain configurations of a package.

    Variants can be declared sticky, which prevents them from being enabled or disabled by the concretizer. Sticky variants must be set explicitly by users on the command line or in packages.yaml.

    • Allow conditional possible values in variants

    • Add a sticky property to variants

6.15.4.2 Other new features of note
  • Environment views can optionally link only run dependencies with link:run

  • spack external find --all finds library-only packages in addition to build dependencies

  • Customizable config:license_dir option

  • spack external find --path PATH takes a custom search path

  • spack spec has a new --format argument like spack find

  • spack concretize --quiet skips printing concretized specs

  • spack info now has cleaner output and displays test info

  • Package-level submodule option for git commit versions

  • Using /hash syntax to refer to concrete specs in an environment now works even if /hash is not installed.

6.15.4.3 Major internal refactors
  • full hash (see above)

  • new develop versioning scheme 0.19.0-dev0

  • Allow for multiple dependencies/dependents from the same package

  • Splice differing virtual packages

6.15.4.4 Performance Improvements
  • Concretization of large environments with unify: when_possible is much faster than concretizing separately (see above)

  • Single-pass view generation algorithm is 2.6x faster

6.15.4.5 Archspec improvements
  • oneapi and dpcpp flag support

  • better support for M1 and a64fx

6.15.4.6 Removals and Deprecations
  • Spack no longer supports Python 2.6

  • Removed deprecated --run-tests option of spack install; use spack test

  • Removed deprecated spack flake8; use spack style

  • Deprecate spack:concretization config option; use concretizer:unify

  • Deprecate top-level module configuration; use module sets

  • spack activate and spack deactivate are deprecated in favor of environments; will be removed in 0.19.0

6.15.4.7 Notable Bugfixes
  • Fix bug that broke locks with many parallel builds

  • Many bugfixes and consistency improvements for the new concretizer and --reuse

6.15.4.8 Packages
  • CMakePackage uses CMAKE_INSTALL_RPATH_USE_LINK_PATH

  • Refactored lua support: lua-lang virtual supports both lua and luajit via new LuaPackage build system

  • PythonPackage: now installs packages with pip

  • Python: improve site_packages_dir handling

  • Extends: support spec, not just package name

  • Use stable URLs and ?full_index=1 for all github patches

6.16 Slurm 23.02

6.16.1 Important Notes on Upgrading Slurm from a Previous Version

If using the slurmdbd (Slurm DataBase Daemon) you must update this first.

If using a backup DBD you must start the primary first to do any database conversion, the backup will not start until this has happened.

The 23.02 slurmdbd will work with Slurm daemons of version 21.08 and above. You will not need to update all clusters at the same time, but it is very important to update slurmdbd first and having it running before updating any other clusters making use of it.

Slurm can be upgraded from version 22.05 to version 23.02 without loss of jobs or other state information. Upgrading directly from an earlier version of Slurm will result in loss of state information.

All SPANK plugins must be recompiled when upgrading from any Slurm version prior to 23.02.

Note
Note

PMIx v1.x is no longer supported.

6.16.2 Highlights

  • slurmctld - Add new RPC rate limiting feature. This is enabled through SlurmctldParameters=rl_enable, otherwise disabled by default.

  • Make scontrol reconfigure and sending a SIGHUP to the slurmctld behave the same. If you were using SIGHUP as a 'lighter' scontrol reconfigure to rotate logs please update your scripts to use SIGUSR2 instead.

  • Change cloud nodes to show by default. PrivateData=cloud is no longer needed.

  • sreport - Count planned (FKA reserved) time for jobs running in IGNORE_JOBS reservations. Previously was lumped into IDLE time.

  • job_container/tmpfs - Support running with an arbitrary list of private mount points (/tmp and /dev/shm are the default, but not required).

  • job_container/tmpfs - Set more environment variables in InitScript.

  • Make all cgroup directories created by Slurm owned by root. This was the behavior in cgroup/v2 but not in cgroup/v1 where by default the step directories ownership were set to the user and group of the job.

  • accounting_storage/mysql - change purge/archive to calculate record ages based on end time, rather than start or submission times.

  • job_submit/lua - add support for log_user() from slurm_job_modify().

  • Run the following scripts in slurmscriptd instead of slurmctld: ResumeProgram, ResumeFailProgram, SuspendProgram, ResvProlog, ResvEpilog, and RebootProgram (only with SlurmctldParameters=reboot_from_controller).

  • Only permit changing log levels with srun --slurmd-debug by root or SlurmUser.

  • slurmctld will fatal() when reconfiguring the job_submit plugin fails.

  • Add PowerDownOnIdle partition option to power down nodes after nodes become idle.

  • Add [jobid.stepid] prefix from slurmstepd and slurmscriptd prefix from slurmcriptd to Syslog logging. Previously was only happening when logging to a file.

  • Add purge and archive functionality for job environment and job batch script records.

  • Extend support for Include files to all "configless" client commands.

  • Make node weight usable for powered down and rebooting nodes.

  • Removed launch plugin.

  • Add Extra field to job to store extra information other than a comment.

  • Add usage gathering for AMD (requires ROCM 5.5+) and NVIDIA gpus.

  • Add job’s allocated nodes, features, oversubscribe, partition, and reservation to SLURM_RESUME_FILE output for power saving.

  • Automatically create directories for stdout/stderr output files. Paths may use %j and related substitution characters as well.

  • Add --tres-per-task to salloc/sbatch/srun.

  • Allow nodefeatures plugin features to work with cloud nodes. e.g. - Powered down nodes have no active changeable features.

    • Nodes can’t be changed to other active features until powered down.

    • Active changeable features are reset/cleared on power down.

  • Make slurmstepd cgroups constrained by total configured memory from slurm.conf (NodeName=<> RealMemory=#) instead of total physical memory.

  • node_features/helpers - add support for the OR and parentheses operators in a --constraint expression.

  • slurmctld will fatal() when [Prolog|Epilog]Slurmctld are defined but are not executable.

  • Validate node registered active features are a super set of node’s currently active changeable features.

  • On clusters without any PrologFlags options, batch jobs with failed prologs nolonger generate an output file.

  • Add SLURM_JOB_START_TIME and SLURM_JOB_END_TIME environment variables.

  • Add SuspendExcStates option to slurm.conf to avoid suspending/powering down specific node states.

  • Add support for DCMI power readings in IPMI plugin.

  • slurmrestd served /slurm/v0.0.39 and /slurmdb/v0.0.39 endpoints had major changes from prior versions. Almost all schemas have been renamed and modified. Sites using OpenAPI Generator clients are highly suggested to upgrade to to using atleast version 6.x due to limitations with prior versions.

  • Allow for --nodelist to contain more nodes than required by --nodes.

  • Rename nodes to nodes_resume in SLURM_RESUME_FILE job output.

  • Rename all_nodes to all_nodes_resume in SLURM_RESUME_FILE output.

  • Add jobcomp/kafka plugin.

  • Add new PreemptParameters=reclaim_licenses option which will allow higher priority jobs to preempt jobs to free up used licenses. (This is only enabled for with PreemptModes of CANCEL and REQUEUE, as Slurm cannot guarantee suspended jobs will release licenses correctly.)

  • hpe/slingshot - add support for the instant-on feature.

  • Add ability to update SuspendExc* parameters with scontrol.

  • Add ability to restore SuspendExc* parameters on restart with slurmctld -R option.

  • Add ability to clear a GRES specification by setting it to "0" via scontrol update job.

  • Add SLURM_JOB_OVERSUBSCRIBE environment variable for Epilog, Prolog, EpilogSlurmctld, PrologSlurmctld, and mail ouput.

  • System node down reasons are appended to existing reasons, separated by ':'.

  • New command scrun has been added. scrun acts as an Open Container Initiative (OCI) runtime proxy to run containers seamlessly via Slurm.

  • Fixed GpuFreqDef option. When set in slurm.conf, it will be used if --gpu-freq was not explicitly set by the job step.

6.16.3 Configuration File Changes (see appropriate man page for details)

  • job_container.conf - Added Dirs option to list desired private mount points.

  • node_features plugins - invalid users specified for AllowUserBoot will now result in fatal() rather than just an error.

  • Deprecate AllowedKmemSpace, ConstrainKmemSpace, MaxKmemPercent, and MinKmemSpace.

  • Allow jobs to queue even if the user is not in AllowGroups when EnforcePartLimits=no is set. This ensures consistency for all the Partition access controls, and matches the documented behavior for EnforcePartLimits.

  • Add InfluxDBTimeout parameter to acct_gather.conf.

  • job_container/tmpfs - add support for expanding %h and %n in BasePath.

  • slurm.conf - Removed SlurmctldPlugstack option.

  • Add new SlurmctldParameters=validate_nodeaddr_threads=<number> option to allow concurrent hostname resolution at slurmctld startup.

  • Add new AccountingStoreFlags=job_extra option to store a job’s extra field in the database.

  • Add new defer_batch option to SchedulerParameters to only defer scheduling for batch jobs.

  • Add new DebugFlags option JobComp to replace Elasticsearch.

  • Add configurable job requeue limit parameter - MaxBatchRequeue - in slurm.conf to permit changes from the old hard-coded value of 5.

  • helpers.conf - Allow specification of node specific features.

  • helpers.conf - Allow many features to one helper script.

  • job_container/tmpfs - Add Shared option to support shared namespaces. This allows autofs to work with the job_container/tmpfs plugin when enabled.

  • acct_gather.conf - Added EnergyIPMIPowerSensors=Node=DCMI and Node=DCMI_ENHANCED.

  • Add new getnameinfo_cache_timeout=<number> option to CommunicationParameters to adjust or disable caching the results of getnameinfo().

  • Add new PrologFlags=ForceRequeueOnFail option to automatically requeue batch jobs on Prolog failures regardless of the job --requeue setting.

  • Add HealthCheckNodeState=NONDRAINED_IDLE option.

  • Add explicit to Flags in gres.conf. This makes it so the gres is not automatically added to a job’s allocation when --exclusive is used. Note that this is a per-node flag.

  • Moved the preempt_ options from SchedulerParameters to PreemptParameters, and dropped the prefix from the option names. (The old options will still be parsed for backwards compatibility, but are now undocumented.)

  • Add LaunchParameters=ulimit_pam_adopt, which enables setting RLIMIT_RSS in adopted processes.

  • Update SwitchParameters=job_vni to enable/disable creating job VNIs for all jobs, or when a user requests them.

  • Update SwitchParameters=single_node_vni to enable/disable creating single node VNIs for all jobs, or when a user requests them.

  • Add ability to preserve SuspendExc* parameters on reconfig with ReconfigFlags=KeepPowerSaveSettings.

  • slurmdbd.conf - Add new AllResourcesAbsolute to force all new resources to be created with the Absolute flag.

  • topology/tree - Add new TopologyParam=SwitchAsNodeRank option to reorder nodes based on switch layout. This can be useful if the naming convention for the nodes does not natually map to the network topology.

  • Removed the default setting for GpuFreqDef. If unset, no attempt to change the GPU frequency will be made if --gpu-freq is not set for the step.

6.16.4 Command Changes (see man pages for details)

  • sacctmgr - no longer force updates to the AdminComment, Comment, or SystemComment to lower-case.

  • sinfo - Add -F/--future option to sinfo to display future nodes.

  • sacct - Rename Reserved field to Planned to match sreport and the nomenclature of the 'Planned' node.

  • scontrol - advanced reservation flag MAINT will no longer replace nodes, similar to STATIC_ALLOC

  • sbatch - add parsing for #PBS -d and #PBS -w.

  • scontrol show assoc_mgr will show username(uid) instead of uid in QoS section.

  • Add strigger --draining and -R/--resume options.

  • Change --oversubscribe and --exclusive to be mutually exclusive for job submission. Job submission commands will now fatal if both are set. Previously, these options would override each other, with the last one in the job submission command taking effect.

  • scontrol - Requested TRES and allocated TRES will now always be printed when showing jobs, instead of one TRES output that was either the requested or allocated.

  • srun --ntasks-per-core now applies to job and step allocations. Now, use of --ntasks-per-core=1 implies --cpu-bind=cores and --ntasks-per-core>1 implies --cpu-bind=threads.

  • salloc/sbatch/srun - Check and abort if ntasks-per-core > threads-per-core.

  • scontrol - Add ResumeAfter=<secs> option to scontrol update nodename=.

  • Add a new nodes= argument to scontrol setdebug to allow the debug level on the slurmd processes to be temporarily altered.

  • Add a new nodes= argument to scontrol setdebugflags as well.

  • Make it so scrontab prints client-side the job_submit() error messsage (which can be set i.e. by using the log_user() function for the lua plugin).

  • scontrol - Reservations will not be allowed to have STATIC_ALLOC or MAINT flags and REPLACE[_DOWN] flags simultaneously.

  • scontrol - Reservations will only accept one reoccurring flag when being created or updated.

  • scontrol - A reservation cannot be updated to be reoccurring if it is already a floating reservation.

  • squeue - removed unused %s and SelectJobInfo formats.

  • squeue - align print format for exit and derived codes with that of other components (<exit_status>:<signal_number>).

  • sacct - Add --array option to expand job arrays and display array tasks on separate lines.

  • Partial support for --json and --yaml formated outputs have been implemented for sacctmgr, sdiag, sinfo, squeue, and scontrol. The resultant data ouput will be filtered by normal command arguments. Formatting arguments will continue to be ignored.

  • salloc/sbatch/srun - extended the --nodes syntax to allow for a list of valid node counts to be allocated to the job. This also supports a "step count" value (e.g., --nodes=20-100:20 is equivalent to --nodes=20,40,60,80,100) which can simplify the syntax when the job needs to scale by a certain "chunk" size.

  • srun - add user requestible vnis with --network=job_vni option.

  • srun - add user requestible single node VNIs with the --network=single_node_vni option.

6.16.5 API Changes

  • job_container plugins - container_p_stepd_create() function signature replaced uint32_t uid with stepd_step_rec_t* step.

  • gres plugins - gres_g_get_devices() function signature replaced pid_t pid with stepd_step_rec_t* step.

  • cgroup plugins - task_cgroup_devices_constrain() function signature removed pid_t pid.

  • task plugins - replace task_p_pre_set_affinity(), task_p_set_affinity(), and task_p_post_set_affinity() with task_p_pre_launch_priv() like it was back in slurm 20.11.

  • Allow for concurrent processing of job_submit_g_submit() and job_submit_g_modify() calls. If your plugin is not capable of concurrent operation you must add additional locking within your plugin.

  • Removed return value from slurm_list_append().

  • The List and ListIterator types have been removed in favor of list_t and list_itr_t respectively.

  • burst buffer plugins - add bb_g_build_het_job_script(). bb_g_get_status() - added authenticated UID and GID. bb_g_run_script() - added job_info argument.

  • burst_buffer.lua - Pass UID and GID to most hooks. Pass job_info (detailed job information) to many hooks. See etc/burst_buffer.lua.example for a complete list of changes. WARNING: Backwards compatibility is broken for slurm_bb_get_status: UID and GID are passed before the variadic arguments. If UID and GID are not explicitly listed as arguments to slurm_bb_get_status(), then they will be included in the variadic arguments. Backwards compatibility is maintained for all other hooks because the new arguments are passed after the existing arguments.

  • node_features plugins - node_features_p_reboot_weight() function removed. node_features_p_job_valid() - added parameter feature_list. node_features_p_job_xlate() - added parameters feature_list and job_node_bitmap.

  • New data_parser interface with v0.0.39 plugin.

6.16.6 Known Issues

  • The --uid option for the srun command is broken: its use may lead to the error message job <job ID> queued and waiting for resources. This does not, however, affect the sbatch command.

7 Removed and deprecated features and packages

This section lists features and packages that were removed from SUSE Linux Enterprise for High-Performance Computing or will be removed in upcoming versions.

  • dapl, rds-tools, and imgen are being deprecated due to lack of upstream activity.

  • openmpi 2 and openmpi 3 are being deprecated due to being replaced by openmpi 4.

7.1 Removed features and packages

The following features and packages have been removed in this release.

  • Python 2 bindings for genders has been removed. These are now provided for Python 3.

  • Ganglia is not supported anymore in 15 SP5. It has been replaced with Grafana (https://grafana.com/)

  • Due to a lack of usage by customers, some library packages have been removed from the HPC module in SLE HPC 15 SP5. On SUSE Linux Enterprise you can build your own library using spack. These libraries will continue to be available through SUSE Package Hub. The following libraries have been removed:

    • boost

    • adios

    • gsl

    • fftw3

    • hypre

    • metis

    • mumps

    • netcdf

    • ocr

    • petsc

    • ptscotch

    • scalapack

    • superlu

    • trilinos

7.2 Deprecated features and packages

The following features and packages are deprecated and will be removed in a future version of SUSE Linux Enterprise for High-Performance Computing.

  • clustduct is deprecated and will be removed in SUSE Linux Enterprise for High-Performance Computing 15 SP6. With SLE HPC 15 SP5, warewulf4 has been introduced as cluster deployment tool, and users are advised to migrate to it.

8 Obtaining source code

This SUSE product includes materials licensed to SUSE under the GNU General Public License (GPL). The GPL requires SUSE to provide the source code that corresponds to the GPL-licensed material. The source code is available for download at https://www.suse.com/download/sle-hpc/ on Medium 2. For up to three years after distribution of the SUSE product, upon request, SUSE will mail a copy of the source code. Send requests by e-mail to sle_source_request@suse.com. SUSE may charge a reasonable fee to recover distribution costs.

Print this page