Sun Fire Cluster
|
The SunFire cluster is a Symmetric Multiprocessor (SMP) system based on the UltraSPARC
line of processors and the Solaris Operating Environment.
All cluster nodes are currently running Solaris 10, Sun HPC ClusterTools 6.0 and 7.0,
Sun GridEngine 6.0 Enterprise Edition, and Sun Studio 12 and are connected
using Gigabit ethernet. The main cluster is comprised of seven Sun Fire 25000 servers (hpcvl0 to hpcvl6)
each of which have 72 x (2MB on-chip L2 cache and 32MB L3 cache) dual-core (CPU) UltraSPARC-IV+ processors.
These nodes are each configured with 576 GB of RAM and are also connected using Gigabit Ethernet.
An additional three Sun Fire 15000 servers (hpcvl7 to hpcvl9) are configured with 72 x UltraSPARC-III
processors and 288 GB of memory.
Current Configurations:
- Six Sun Fire 25000 Nodes (hpcvl0 to hpcvl5) with 72 X dual-core UltraSPARC-IV+ 1.5 GHz processors with 576 GB of RAM.
- One Sun Fire 25000 Node (hpcvl6) with 72 X dual-core UltraSPARC-IV+ 1.8 GHz processors with 576 GB of RAM.
- Three Sun Fire 15K Nodes (hpcvl7 to hpcvl9) with 72 x UltraSPARC-III processors and 288 GB of RAM.
- Two Sun Fire 6900 Nodes (1 at U of O, and 1 at Carleton) with 24 x UltraSPARC-IV+ processors with 192 GB of RAM. Both are to be mainly used as workup nodes.
- 1 Sun Fire 4800 with 12 x UltraSPARC-III processors with 48 GB of RAM at Ryerson University. Currently used as a workup node.
- A total of 160 TB of Sun StorEdge 3510 disk.
Interactive logins to the Sun Fire Cluster are done via the login node sfnode0.hpcvl.queensu.ca.
The standard login procedure is through the HPCVL Secure Portal. We also support the Secure Shell (SSH, v2) suite of utilities - ssh, scp, and sftp.
Click here for additional information on user accounts,
security requirements, cluster access, and usage policies.
|
|
Back to top
Workup Facilities
Please see Workup Facilites for more information.
File Storage: Disk
Disk storage for software applications and user home directories, temporary
space, scratch, and long term is provided by 24 TB, 12 TB, 12 TB, and 28 TB respectively of Sun StorEdge 3510 disk technology.
We are now using Sun SAM-QFS (v4.5) High Performance HPC system to manage the arrays.
Two Sun Fire V890s act as the SAM-QFS servers for the Sun Fire cluster and serve the storage.
The global scratch space is available to all nodes in the cluster via SAM-QFS. For more information on the File System, please read the File System FAQ.
Back to top
File Storage: Tape
Tape storage is attached as part of our SAM-QFS set up. A Sun StorEdge L1400 tape library containing 2 x 650 x 400 GB tapes (native capacity of 1.04 PB) with an effective capacity of 520 TB.
Back to top
File Storage: Backup
We are currently backing up User's home directories for a period of ONE Month
after which, the disk space will be recycled.
We have deployed a large DLT (digital linear tape) tape backup system using
Sun StorEdge L1400 tape library. We have implemented a backup and retrieval strategy
of user files located in each users home directory (i.e., all files/directories
referenced by the $HOME environment variable).
See SAM-QFS FAQ for details.
Back to top
Software Environment
Currently all Sun servers run the 64-bit Solaris 10 Operating Environment. Additional installed software includes:
- Sun
Studio 12
Sun Studio 12 is Sun's software development environment,
which includes a complete set of graphical and command line tools
to help you build, debug, run, and tune your C, C++, Fortran, and high performance
FORTRAN applications.
[more»]
- Sun HPC
ClusterTools 6.0 & 7.0
Sun HPC ClusterTools 6.0 & 7.0 are suites of applications and libraries for high
performance software development and workload management of serial and parallel
applications. [more»]
- Sun
Grid Engine 6.0 Enterprise Edition
Sun Grid Engine software is distributed workload management software that
optimizes utilization of software and hardware resources in heterogeneous
networked environments. [more»]
Back to top
Specialized Resources
HPCVL also operates a Gridrack consisting of 20 - Sun X4100 servers. These resources are reserved for
specialty projects such as SNO and SNOLabs at the Sudbury Neutrino Observatory.
Gridrack Configuration:
- 20 - Dual CPU/Dual Core 2.4 GHz Opteron Sun Fire X4100 Servers
- 1 - Dual Core Sun Fire X2100 Server acting as the Login node
- All nodes have 4 Gigabytes of RAM available
- Gigabit ethernet interconnect
Back to top
|
