Blog

HPC

BeeGFS 7.2.6 Now Available

March 17, 2022
5 min read
EXX-Blog-BGFS-update.jpg

BeeGFS 7.2.6 Released

ThinkParQ announced a new version of BeeGFS is now available. Before we take a look at what's new, let's take a quick look at what BeeGFS is.

What is BeeGFS?

Developed with a strong focus on performance and designed for ease of use, simple installation, and management, BeeGFS is one of the leading parallel file systems that continues to grow and gain significant popularity in the community. BeeGFS has evolved into a world-wide valued filesystem offering maximum performance, scalability, high flexibility, and robustness.

BeeGFS is available free of charge for end-users. For enterprise systems, professional commercial support is also available through our BeeGFS Parallel Storage Solutions.

Key Benefits

  • Distributed File Contents and Metadata - One of the most fundamental concepts of BeeGFS is the strict avoidance of architectural bottlenecks. Striping file contents across multiple storage servers is only one part of this concept. Another important aspect is the distribution of file system metadata (e.g., directory information) across multiple metadata servers. Large systems and metadata intensive applications, in general, can greatly profit from the latter feature.
  • HPC Technologies - BeeGFS is built on highly efficient and scalable multithreaded core components with native RDMA support. File system nodes can serve RDMA (InfiniBand, Omni-Path, RoCE) and TCP/IP network connections at the same time and automatically switch to a redundant connection path in case any of them fails.
  • Easy to Use - BeeGFS requires no kernel patches (the client is a patchless kernel module, the server components are userspace daemons), comes with graphical cluster installation tools and allows you to add more clients and servers to the running system whenever you want it.
  • Optimized for Highly Concurrent Access - Simple remote file systems like NFS do not only have serious performance problems in case of highly concurrent access, they can even corrupt data when multiple clients write to the same shared file, which is a typical use-case for cluster applications. BeeGFS was specifically designed with such use-cases in mind to deliver optimal robustness and performance in situations of high I/O load.
  • Client and Server on any Machine - No specific enterprise Linux distribution or other special environment is required to run BeeGFS. BeeGFS client and servers can even run on the same machine to enable performance increases for small clusters or networks. BeeGFS requires no dedicated file system partition on the servers - It uses existing partitions, formatted with any of the standard Linux file systems, e.g., XFS, ext4 or ZFS. For larger networks, it is also possible to create several distinct BeeGFS file system partitions with different configurations.

Need a turnkey storage appliance built on validated hardware for demanding HPC workloads?
Learn more about Exxact's BeeGFS solutions


BeeGFS v7.2.6

The BeeGFS 7.2.6 release includes small bugfixes and QoL improvements.

General Changes/Improvements

  • Added CMake configuration files to enable builds with CMake.
  • Updated client module code to work with RHEL 8.5 and OFED 5.5.
  • Introduced an experimental client side option to disable session checks for read, write and fsync operations on files that were changed in that session. This can be useful to continue reading or writing after server crashes, at the expense of not being able to detect crashes and therefore potential loss of caches on the servers any more. This option should only be used after careful consideration and must not be used unless measures have been taken to prevent cache loss on the servers. Please see the documentation in the client configuration file for more information.
  • Some smaller changes to the documentation (quota, fsck).

Fixes

  • Fixed an integer overflow in the inode parsing code that could lead to miscalculations of stripe set sizes for very large chunksizes and numbers of targets.
  • Fixed a compile time issue with newer versions of GCC. The code will now build with GCC versions > 9.
  • Removed informational messages about RDMA connection initiation and teardown from the client log in standard log levels.
  • The fsck will now properly ignore the tuneENOENTCacheValidityMS introduced in 7.2.5.

Known Issues and Limitations

  • The Metadata daemon does not work reliably on RHEL/CentOS 8 and SLES 15.1 and 15.2 due to a problem in the versions of glibc. The problem was fixed in RHEL/CentOS 8.1 and SLES 15.3.
  • The client module might hang if the IBV device(s) used are taken offline during operation.
  • The client module does not compile on SLES 15.2 with Mellanox OFED 5.2
  • Attempting to mount the filesystem with RDMA over RoCE will crash the kernel on RHEL 8.5 (and Rocky, Alma 8.5). RDMA over IB with and without Mellanox OFED works fine.

Supported Linux Distributions and Kernels

Packages are provided for the x86_64 architecture and the following distributions:

  • RHEL 7 and 8
  • SLES 15
  • Debian 9 and 10
  • Ubuntu 18
  • Ubuntu 20
  • Rocky Linux and AlmaLinux should now be configured to use the RHEL 8 repository (note the known issue with RoCE and RHEL 8.5 described above)

The following Mellanox OFED driver versions are supported: 5.5, 5.4, 5.3, 5.2, 5.1, 5.0, 4.9

The full integration test suite was run on Rocky 8.5, RHEL 8.4, Rocky 8.4, Alma 8.4, CentOS 8.3, CentOS 7.9, OpenSUSE 15 SP3 and Debian 9 and 10.

Client build testing:

  • RHEL 7.9: no OFED, OFED 4.9, 5.0, 5.2, 5.3, 5.4
  • RHEL 8.3: no OFED, OFED 4.9, 5.0, 5.1, 5.2, 5.3, 5.4
  • AlmaLinux 8.4: no OFED, OFED 5.3, 5.4
  • AlmaLinux 8.5: no OFED, OFED 5.3, 5.4, 5.5
  • Rocky Linux 8.4: no OFED, OFED 5.3, 5.4
  • Rocky Linux 8.5: no OFED, OFED 5.5
  • SLES 15.1: no OFED, OFED 5.0
  • SLES 15.2: no OFED, OFED 5.1, 5.4
  • SLES 15.3: no OFED, OFED 5.4, 5,5
  • Debian 9: no OFED, 5.4
  • Debian 10: no OFED, OFED 5.2, 5.3, 5.4
  • Ubuntu 18: no OFED
  • Ubuntu 20: no OFED, OFED 5.4

Version Interoperability

BeeGFS v7.2.6 is compatible with BeeGFS v7.2.x.

Upgrading from Older Versions

To upgrade from an older version, please refer to the Upgrade Guide.


Have any questions about BeeGFS or other storage solutions? Contact Exxact Today


Free Resources

Browse our whitepapers, e-books, case studies, and reference architecture.

Explore
EXX-Blog-BGFS-update.jpg
HPC

BeeGFS 7.2.6 Now Available

March 17, 2022 5 min read

BeeGFS 7.2.6 Released

ThinkParQ announced a new version of BeeGFS is now available. Before we take a look at what's new, let's take a quick look at what BeeGFS is.

What is BeeGFS?

Developed with a strong focus on performance and designed for ease of use, simple installation, and management, BeeGFS is one of the leading parallel file systems that continues to grow and gain significant popularity in the community. BeeGFS has evolved into a world-wide valued filesystem offering maximum performance, scalability, high flexibility, and robustness.

BeeGFS is available free of charge for end-users. For enterprise systems, professional commercial support is also available through our BeeGFS Parallel Storage Solutions.

Key Benefits

  • Distributed File Contents and Metadata - One of the most fundamental concepts of BeeGFS is the strict avoidance of architectural bottlenecks. Striping file contents across multiple storage servers is only one part of this concept. Another important aspect is the distribution of file system metadata (e.g., directory information) across multiple metadata servers. Large systems and metadata intensive applications, in general, can greatly profit from the latter feature.
  • HPC Technologies - BeeGFS is built on highly efficient and scalable multithreaded core components with native RDMA support. File system nodes can serve RDMA (InfiniBand, Omni-Path, RoCE) and TCP/IP network connections at the same time and automatically switch to a redundant connection path in case any of them fails.
  • Easy to Use - BeeGFS requires no kernel patches (the client is a patchless kernel module, the server components are userspace daemons), comes with graphical cluster installation tools and allows you to add more clients and servers to the running system whenever you want it.
  • Optimized for Highly Concurrent Access - Simple remote file systems like NFS do not only have serious performance problems in case of highly concurrent access, they can even corrupt data when multiple clients write to the same shared file, which is a typical use-case for cluster applications. BeeGFS was specifically designed with such use-cases in mind to deliver optimal robustness and performance in situations of high I/O load.
  • Client and Server on any Machine - No specific enterprise Linux distribution or other special environment is required to run BeeGFS. BeeGFS client and servers can even run on the same machine to enable performance increases for small clusters or networks. BeeGFS requires no dedicated file system partition on the servers - It uses existing partitions, formatted with any of the standard Linux file systems, e.g., XFS, ext4 or ZFS. For larger networks, it is also possible to create several distinct BeeGFS file system partitions with different configurations.

Need a turnkey storage appliance built on validated hardware for demanding HPC workloads?
Learn more about Exxact's BeeGFS solutions


BeeGFS v7.2.6

The BeeGFS 7.2.6 release includes small bugfixes and QoL improvements.

General Changes/Improvements

  • Added CMake configuration files to enable builds with CMake.
  • Updated client module code to work with RHEL 8.5 and OFED 5.5.
  • Introduced an experimental client side option to disable session checks for read, write and fsync operations on files that were changed in that session. This can be useful to continue reading or writing after server crashes, at the expense of not being able to detect crashes and therefore potential loss of caches on the servers any more. This option should only be used after careful consideration and must not be used unless measures have been taken to prevent cache loss on the servers. Please see the documentation in the client configuration file for more information.
  • Some smaller changes to the documentation (quota, fsck).

Fixes

  • Fixed an integer overflow in the inode parsing code that could lead to miscalculations of stripe set sizes for very large chunksizes and numbers of targets.
  • Fixed a compile time issue with newer versions of GCC. The code will now build with GCC versions > 9.
  • Removed informational messages about RDMA connection initiation and teardown from the client log in standard log levels.
  • The fsck will now properly ignore the tuneENOENTCacheValidityMS introduced in 7.2.5.

Known Issues and Limitations

  • The Metadata daemon does not work reliably on RHEL/CentOS 8 and SLES 15.1 and 15.2 due to a problem in the versions of glibc. The problem was fixed in RHEL/CentOS 8.1 and SLES 15.3.
  • The client module might hang if the IBV device(s) used are taken offline during operation.
  • The client module does not compile on SLES 15.2 with Mellanox OFED 5.2
  • Attempting to mount the filesystem with RDMA over RoCE will crash the kernel on RHEL 8.5 (and Rocky, Alma 8.5). RDMA over IB with and without Mellanox OFED works fine.

Supported Linux Distributions and Kernels

Packages are provided for the x86_64 architecture and the following distributions:

  • RHEL 7 and 8
  • SLES 15
  • Debian 9 and 10
  • Ubuntu 18
  • Ubuntu 20
  • Rocky Linux and AlmaLinux should now be configured to use the RHEL 8 repository (note the known issue with RoCE and RHEL 8.5 described above)

The following Mellanox OFED driver versions are supported: 5.5, 5.4, 5.3, 5.2, 5.1, 5.0, 4.9

The full integration test suite was run on Rocky 8.5, RHEL 8.4, Rocky 8.4, Alma 8.4, CentOS 8.3, CentOS 7.9, OpenSUSE 15 SP3 and Debian 9 and 10.

Client build testing:

  • RHEL 7.9: no OFED, OFED 4.9, 5.0, 5.2, 5.3, 5.4
  • RHEL 8.3: no OFED, OFED 4.9, 5.0, 5.1, 5.2, 5.3, 5.4
  • AlmaLinux 8.4: no OFED, OFED 5.3, 5.4
  • AlmaLinux 8.5: no OFED, OFED 5.3, 5.4, 5.5
  • Rocky Linux 8.4: no OFED, OFED 5.3, 5.4
  • Rocky Linux 8.5: no OFED, OFED 5.5
  • SLES 15.1: no OFED, OFED 5.0
  • SLES 15.2: no OFED, OFED 5.1, 5.4
  • SLES 15.3: no OFED, OFED 5.4, 5,5
  • Debian 9: no OFED, 5.4
  • Debian 10: no OFED, OFED 5.2, 5.3, 5.4
  • Ubuntu 18: no OFED
  • Ubuntu 20: no OFED, OFED 5.4

Version Interoperability

BeeGFS v7.2.6 is compatible with BeeGFS v7.2.x.

Upgrading from Older Versions

To upgrade from an older version, please refer to the Upgrade Guide.


Have any questions about BeeGFS or other storage solutions? Contact Exxact Today


Free Resources

Browse our whitepapers, e-books, case studies, and reference architecture.

Explore