39 lines
1.7 KiB
ReStructuredText
39 lines
1.7 KiB
ReStructuredText
==========
|
|
Background
|
|
==========
|
|
|
|
Warewulf is based on the design of the original Beowulf Cluster design
|
|
(and thus the name, soft\ **WARE** implementation of the beo\
|
|
**WULF**)
|
|
|
|
The `Beowulf Cluster <https://en.wikipedia.org/wiki/Beowulf_cluster>`_
|
|
design was developed in 1996 by Dr. Thomas Sterling and Dr. Donald
|
|
Becker at NASA. The architecture is defined as a group of similar
|
|
compute worker nodes all connected together using standard commodity
|
|
equipment on a private network segment. The control node (historically
|
|
referred to as the "master" or "head" node) is dual homed (has two
|
|
network interface cards) with one of these network interface cards
|
|
attached to the upstream public network and the other connected to a
|
|
private network which connects to all of the compute worker nodes (as
|
|
seen in the figure below).
|
|
|
|
.. image:: beowulf_architecture.png
|
|
:alt: Beowulf architecture
|
|
|
|
This simple topology is the foundation for creating every scalable HPC
|
|
cluster resource. Even today, almost 30 years after the inception of
|
|
this architecture, this is the baseline architecture that traditional
|
|
HPC systems are built to.
|
|
|
|
Other considerations for a working HPC-type cluster are storage,
|
|
scheduling and resource management, monitoring, interactive systems,
|
|
etc. For smaller systems, much of these requirements can be managed
|
|
all from a single control node, but as the system scales, it may need
|
|
to have groups of nodes dedicated to these different services.
|
|
|
|
Warewulf is easily capable of building simple and turnkey HPC
|
|
clusters, to giant massive complex multi-purpose computing clusters,
|
|
through next generation computing platforms.
|
|
|
|
Anytime a cluster of systems is needed, Warewulf is your tool!
|