RHUL (10 Questions)

From GridPP Wiki
Revision as of 11:13, 25 November 2005 by Steve traylen (Talk | contribs)

(diff) ← Older revision | Latest revision (diff) | Newer revision → (diff)
Jump to: navigation, search

'Status: Complete'

Question 1

Provide the name and contact details of your local (Departmental) and Institutional network support staff.

  • My Departmental network support contact is: Barry Green, RHUL Centre for Particle Physics
  • My Institutional network support contact is: Nigel Rata, RHUL Computer Centre

Question 2

Provide details of the responsibilities, together with the demarcation of those responsibilities, of your local and Institutional network support staff.

  • The departmental contact is responsible for: Particle Physics network - all equipment on subnet 134.219.108.0/22 except for gateway address, and physically all equipment up to but not including fibre pair used as uplink to Campus backbone routers.
  • The institutional contact is responsible for: the above fibres up to edge of campus (connection to London MAN)

Question 3

What is a Regional Network Operator (RNO), and why does this matter to you?

  • An RNO is: London MAN (LMN)
  • I care because: they provide our connection to the rest of the internet.

Question 4

What is SuperJANET4? And more importantly what is SuperJANET5?

  • SuperJANET4 is: UK academic network backbone
  • SuperJANET5 is: proposed upgrade

Questions 5, 6, 7 and 9 (part)

5: Draw a simple diagram showing your local (Departmental) network and sufficient of your Institutional network such that you can trace a line from your end-system to the connection from your Institutes network into the RNO infrastructure.

6: On the diagram produced in answer to Question 5, show the capacity of each link in the network and provide a note against each link of its contention ratio.

7: On the diagram produced in answer to Question 5, colour and distinguish the switches and routers and for each device provide a note of its backplane capability.

9.x: On the diagram produced in answer to Question 5 colour in the firewall(s) (or other security devices).

(upload an image via http://wiki.gridpp.ac.uk/wiki/Special:Upload)

File:NetworkAndContention.png

Question 8

What is the average and peak traffic flow between your local (Departmental) network and the Institutional network?

  • Average traffic: average measured over a few days was ~ 1.5 Mbit/s
  • Peak traffic: during last few days of monitoring, have seen max 70 Mbit/s

What is the average and peak traffic flow between your Institutional network and the RNO?

  • Average traffic: For October, the 100 Mbit/s link from RHUL to LMN at ULCC was on average 34% utilised.
  • Peak traffic: During the same period, the same link utilisation peaked at 54%.

What is the total capacity of your Institutional connection to the RNO?

  • Our total capacity is: 100 Mbit/s

What are the upgrade plans for your local (Departmental) network; your Institutional network and the network run by the RNO?

  • Departmental plans: None
  • Institutional plans: aspiration to upgrade to a 1 Gbit connection to LMN

but no solid plans or timeline at present.

  • RNO plans: Not known

Question 9

Do you believe in IS Security? Does your Institute believe in IS Security?

  • I'm a believer: YES
  • We're collective believers: YES

Do you believe in firewalls? Does your Institute believe in firewalls?

  • I'm a believer: YES
  • We're collective believers: YES

Provide information of how changes are made to the rule set of the firewall.

  • Firewall rules are changed by: particle physics network/system admins

Provide a note of the capacity of this device and what happens when that capacity is exceeded.

  • The capacity is: 1Gb/s
  • When it goes over-capacity, the following happens:

Don't know. There is a 100 Mb/s bottleneck that generally prevents this.

Question 10

What is the best performance you can achieve from your end-system to an equivalent system located in some geographically remote (and friendly!) Institute?

  • Best performance is: ~5 MByte/s with bbftp (parallel streams) or 1.2 MByte/s with sftp (single stream).

For your end-system:

Do you understand the kernel, the bus structure; the NIC; and the disk system?
  • I understand: qualified YES

(i.e. yes, to some extent, but one could always understand more)

Do you understand TCP tuning and what it can do for you?
  • I understand: NO, at least not across a WAN.
Do you understand your application and what it can do to your performance?
  • I understand: qualified NO

What is meant by "my application"? I don't consider the grid middleware or VO jobs to be mine, I am just letting LCG and VOs run them on my systems. It is their business to tune these applications.