The Leading Parallel

Cluster File Sy stem

www.thinkparq.com
www.beegfs.io
ABOUT BEEGFS
What is BeeGFS Sy stem Architecture

BeeGFS (formerly FhGFS) is the leading parallel cluster file

system, developed with a strong focus on performance and
designed for very easy installation and management. If I/O
intensive workloads are your problem, BeeGFS is the solution.

Why use BeeGFS

BeeGFS transparently spreads user data across multiple

servers. By increasing the number of servers and disks in the
system, you can simply scale performance and capacity of
the file system to the level that you need, seamlessly from small
clusters up to enterprise-class systems with thousands of nodes.

Get The Most Out Of Your Data

The flexibility, robustness, and outstanding performance of

BeeGFS help our customers around the globe to increase
productiv ity by delivering results faster and by enabling new
data analy sis methods that were not possible without the
advantages of BeeGFS.
KEY ASPECTS
Max imum Scalability Max imum Flex ibility

BeeGFS offers maximum performance and scalability on various BeeGFS supports a w ide range of Linux distributions such
levels. It supports distributed file contents with flexible as RHEL/Fedora, SLES/OpenSuse or Debian/Ubuntu as well as a
striping across storage servers on a per-file or per-directory basis wide range of Linux kernels from ancient 2.6.18 up to the
as well as distributed metadata. latest vanilla.

BeeGFS is optimized especially for use in environments w here The storage services run on top of an ex isting local
performance matters to provide: filesy stem (such as xfs, zfs or others) using the normal POSIX
interface and clients and servers can be added to an ex isting
� Best in class client throughput: 8 GB/s with only a single sy stem w ithout dow ntime.
process streaming on a 100GBit network, while a few
streams can fully saturate the network. BeeGFS supports multiple netw orks and dynamic failover in
� Best in class metadata performance: Linear scalability case one of the network connections is down.
through dynamic metadata namespace partitioning.
� Best in class storage throughput: BeeGFS servers allow
flexible choice of underlying file system to perfectly fit the
given storage hardware.

BeeGFS Storage Pools make different types of storage devices

available within the same namespace.
By having SSDs and HDDs in different pools, pinning of a user
project to the flash pool enables all-flash storage performance
for the current project while still providing the advantage of the
cost-effecient high capacity of spinning disks for other data.

BeeGFS client and server components can also run on the same
physical machines. Thus, BeeGFS can turn a compute rack
into a cost-efficient conv erged data processing and
shared storage unit, eliminating the need for external storage
resources and providing simplified management.
BEEGFS ON DEMAND
Max imum Usability BeeOND

The BeeGFS serv er components are userspace daemons, BeeOND (BeeGFS on demand) allows on the fly creation of a
while the client is a nativ e kernel module that does not complete parallel file system instance on a given set of hosts
require any patches to the kernel itself. All BeeGFS components with just one single command.
can be installed and updated without even rebooting the
machine.

For installation and updates there are rpm/deb package

repositories available; for the startup mechanism, easy-to-use
system service scripts are provided.

BeeGFS was designed with easy administration in mind. The

graphical administration and monitoring sy stem enables
dealing with typical management tasks in a simple and intuitive
way, while everything is of course also available from a
command line interface:

� Live load statistics, even for individual users

� Cluster installation
� Storage service management BeeOND was designed to integrate with cluster batch systems
� Health monitoring to create temporary parallel file sy stem instances on a per-
� And more... job basis on the internal SSDs of compute nodes, which are
Excellent documentation helps to have the whole system up part of a compute job. Such BeeOND instances do not only
and running in one hour. provide a very fast and easy to use temporary buffer, but also
can keep a lot of I/ O load for temporary or random access
files away from the global cluster storage.
BUDDY MIRRORING
Fault tolerance BeeGFSv7: Newest features

BeeGFS storage servers are typically used with an underlying

� Storage pools
RAID to transparently handle disk errors. Using BeeGFS with
� Free space balancing when adding new hardware
shared storage is also possible to handle server failures. The
� Metadata event logging enhancement
built-in BeeGFS Buddy Mirroring approach goes even one � fsck improvements
step further by tolerating the loss of complete servers including � Kernel 4.14, 4.15 support
all data on their RAID volumes - and that with commodity � NIC handling during startup
servers and shared-nothing hardware.
More Features

We already talked about the BeeGFS key aspects scalability,

flexibility and usability and what‘s behind them. But there are
way more features in BeeGFS:

� Runs on various platforms, such as x 86, OpenPOWER,

ARM, Xeon Phi and more
� Re-export through Samba or NFSv4 possible
� Support for user/group quota and ACLs
� Fair I/O option on user level to prevent a single user with
multiple requests from stalling requests of other users
The built-in BeeGFS Buddy Mirroring automatically replicates � Automatic network failover, e.g. if InfiniBand is down,
BeeGFS automatically switches to Ethernet and back later
data, handles storage server failures transparently for running
applications and provides automatic self-healing when a � Online file system sanity check that can analyze and repair
while the system is in use
server comes back online, efficiently resyncing only the files
that have changed while the machine was offline. � Built-in benchmarking tools to help with optimal tuning for
specific hardware and evaluate hardware capabilities
� Support for cloud deployment on e.g. Amazon EC2 or
Microsoft Azure

Not in the list? Just get in touch and w e‘re happy to

discuss all y our questions.
BENCHMARKS
Metadata Operations Throughput Scalability

BeeGFS was designed for extreme scalability. In a testbed1 In the same testbed system with 20 servers, each equipped
with 20 servers and up to 640 client processes (32x the with a single node local performance of 1332 MB/s (write) and
number of metadata servers), BeeGFS delivers a sustained file 1317 MB/s (read), and 160 client processes, BeeGFS
creation rate of more than 500,000 creates per second, demonstrates linear scaling to a sustained throughput of 25
making it possible to create one billion files in as little time as GB/s - which is 94.7% of the maximum theoretical local write
about 30 minutes. and 94.1% of the maximum theoretical local read throughput.

539.7 25.2
600 30
read 24.8
creates
write
25
500
Thousand Creates / second

20
400

GB/s 15
300

10
200

5
100
20
20 18
18 16
16 14
14 12
12 0 10
0 10
8 8
6 6
4 4
1
2 # Metadata Servers 2 # Storage Servers
1

1
Bechmark Sy stem:

20 servers with 2x Intel Xeon X5660 @ 2.8 GHz, 48GB RAM running a Scientific Linux 6.3, Kernel 2.6.32-
279. Each server is equipped with 4x Intel 510 Series SSD (RAID 0) running ext4 as well as QDR
Infiniband. Tests performed using BeeGFS version 2012.10.
LICENSING MODEL
Free to use & Open source User Comments

BeeGFS is free to use for self-supporting end users and can be

downloaded directly from www.beegfs.io
”After many unplanned downtimes with our
previous parallel filesystem, we moved to BeeGFS
more than two years ago. Since then we didn't
have any unplanned downtime for the filesystem
Professional Support anymore."
Michael Hengst, University of Halle, Germany
If you are already using BeeGFS but you don’t have professional
support yet, here are a few reasons to rethink this.

“We are extremely happy with our 3.1PB BeeGFS

Free With installation on 30 servers. It is rock-solid.“
Edition Support Rune Møllegaard Friborg,
University of Aarhus, Denmark
Community Mailing List

Free Updates
“Now under heavy load, our large BeeGFS system
is performing well - bioinfo users are seeing >2x
Ready-to-use binary packages & speedup in their apps, with no more hotspots
Source Code
when hitting common index files. Unlike the
previous system, BeeGFS does not lock up under
Enterprise Features (High-availability, heavy load and even under heavy load,
quota, ACLs) interactive use remains zippy. The only complaint
is how long it's taking us to move data off the
Next Business Day Service Level old system.“
Agreements for Support Harry Mangalam, UC Irvine, USA

Direct Contact to the File System

Developers ”The network file access performance of BeeGFS
is on the same level that we see when our
Early Updates and Hotfixes applications are running on servers with a direct-
attached RAID.“
Customer Section: HowTos and
more Documentation Genomatix, Germany
HAPPY USERS WORLD-WIDE
Scientific Computing

BeeGFS is widely popular among universities and the global

research community, powering some of the fastest
supercomputers in the world to help scientists analyze large
amounts of data efficiently every day.

Life Sciences

BeeGFS is the parallel file system of choice in life sciences. The

fast growing amount of genomics data to store and analyze
quickly in fields like Precision Medicine make BeeGFS the first
choice for our customers.

Finance, Oil & Gas, Media, Automotiv e, ...

BeeGFS is used in many different industries all around the globe

to provide fast access to storage systems of all kinds and sizes,
from small scale up to enterprise-class systems with thousands
of hosts.

ThinkParQ GmbH

Corporate Headquarters Global Sales & Consulting v1711

Trippstadter Str. 110 Phone: +49 631 277 576 300 www.thinkparq.com
67663 Kaiserslautern, Germany sales@thinkparq.com www.beegfs.io