ProCKSI

= System Description =

Cluster Design
The cluster is assumed to have one (or more) ''master nodes'' and several ''slave nodes'' (''compute nodes''). * The master node runs the web server, the file server, the central database server, the email server, the server for the queuing system, and the ProCKSI server itself. It can be used as a compute node, too. * The compute nodes run the calculations.

Image(ClusterDesign.png)

Hardware Specifications

P1. DELL !PowerEdge SC1420 ||'''CPU''' || dual-processor Intel Xeon (3.2GHz / 2MB) ||'''FSB''' || 800MHz ||'''RAM''' || 4GB (4x1GB, DDR2) ||'''HD''' || 1000GB SATA Raid5 (4x250GB, 7200rpm) ||'''ETH0''' || !SysKonnect SK-9871 V2.0 Gigabit Ethernet 1000Base-ZX Adapter, PCI64, Fiber ZX/SC ||'''ETH1''' || Intel Corporation 82545GM Gigabit Ethernet Controller

P2. DELL !PowerEdge SC1425 ||'''CPU''' || dual-processor Intel Xeon (3.2GHz / 2MB) ||'''FSB''' || 800MHz ||'''RAM''' || 4GB (4x1GB, DDR2) ||'''HD''' || 40GB SATA (7200rpm) ||'''ETH0''' || Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05) ||'''ETH1''' || Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05)

P3. DNUK Terraserver R420 ||'''CPU''' || dual-processor, dual-core AMD Opteron 2216 (2.4GHz / 2MB) ||'''FSB''' || 533MHz ||'''RAM''' || 4GB (4x1GB, PC2-4200 DDR2) ||'''HD''' || 80GB SATA (7200rpm) ||'''ETH0''' || Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05) ||'''ETH1''' || Intel Corporation 82541GI Gigabit Ethernet Controller (rev 05)

Computer Names and IP Addresses

=== Cluster ===

==== Hardware ====

|| '''ID''' || '''Description''' || '''Hardware''' 
|| M01      || Master Node       || P1             
|| S01      || Slave Node        || P2             
|| S02      || Slave Node        || P2            
|| S03      || Slave Node        || P3            
|| S04      || Slave Node        || P3

==== External Network ====

|| '''ID''' || '''Interface''' || '''MAC Address''' || '''DNS Name'''         || '''IP''' 
|| M01      || eth1            || 00:00:5A:9D:44:54 || procksi0.cs.nott.ac.uk || 128.243.21.180    
|| S01      || eth1            || 00:15:C5:5D:89:5A || procksi1.cs.nott.ac.uk || 128.243.21.181    
|| S02      || eth1            || 00:15:C5:5D:B8:AC || procksi2.cs.nott.ac.uk || 128.243.21.182    
|| S03      || eth1            || 00:E0:81:4A:6A:E7 || procksi3.cs.nott.ac.uk || 128.243.21.183    
|| S04      || eth1            || 00:E0:81:48:19:81 || procksi4.cs.nott.ac.uk || 128.243.21.184

==== Internal Network ====

|| '''ID''' || '''Interface''' || '''MAC Address''' || '''DNS Name'''         || '''IP''' 
|| M01      || eth0            || 00:13:72:A6:5B:1F || master01.procksi.local || 192.168.199.10 
|| S01      || eth0            || 00:15:C5:5D:89:59 || slave01.procksi.local  || 192.168.199.11 
|| S02      || eth0            || 00:15:C5:5D:B8:AB || slave02.procksi.local  || 192.168.199.12 
|| S03      || eth0            || 00:E0:81:4A:6A:E6 || slave03.procksi.local  || 192.168.199.13 
|| S04      || eth0            || 00:E0:81:48:19:80 || slave04.procksi.local  || 192.168.199.14

=== Others ===

	'''Description'''		'''External Name'''		'''External IP'''
	trac		psiren.cs.nott.ac.uk		128.243.24.84

Backup Strategy * '''Cluster master node''':br
Results are stored on a RAID5 storage element. br
No external backup. * '''Cluster slave nodes''':br
Do not contain sensible data except operating system.br
No external backup. * '''Trac''':br
Hosts Wiki pages, SVN repository, Ticketing system, etc.br
Backup to tape system (scripts on ''much.cs.nott.ac.uk''), incremental from Monday to Friday, full dump once a week: ||'''Partition'''||'''Day''' || ||/ ||every Tuesday || ||/home/ ||every Wednesday || ||/usr/local/ ||every Monday || Contact Person: [http://www.nottingham.ac.uk/cs/lookup/lookup_role.php?id=MDE4Mzgy&page_var=personal William Armitage]