The original Galileo cluster was built in 1997, and it has evolved
over the years through many changes in hardware and software. The
cluster currently consists of 65 computers running Linux. Each
of these computers has mutiple cores (most have four) and 4GB of
memory. The cluster contains a total of 230 3GHz cores.
File storage is provided by five departmental file servers and several
file servers owned by individual research groups. Two of the
departmental file servers are "condo" servers, which have disk crates
attached to them into which disks belonging to individual research
groups can be inserted. For groups that need only a few terabytes,
this provides a cheap way to add storage to Galileo. Groups with
larger storage needs can purchase their own file server to be
added to the cluster.
- Design Goals:
Even though Galileo is
cheap as supercomputers go, it still represents
a large monetary investment for our department. Because of this,
we've designed Galileo with the intent that almost everone in the department
will benefit from it in some way. Most supercomputing clusters
are useful to only a few talented programmers, who know how to write
parallel code that takes full advantage of the cluster. These users
are only a small fraction of our user base. In designing Galileo,
we've also kept in mind the average grad student or undergrad
(or faculty member) who doesn't want or need to spend time parallelizing code,
but needs more computing power than that provided by a desktop computer.
Our intent is that everyone will realize a benefit from the availability
Most of Galileo's computing power is available through its batch
queues. For information about using the batch queues, see:
For information about the software available to Galileo users, see:
The figure below shows Galileo's network architecture, and how it
fits in with the rest of the department's network infrastructure.
As you can see, Galileo and its associated file servers form the
core of our computing infrastructure.