Please enable JavaScript.
Coggle requires JavaScript to display documents.
INFO 3 C8 - Introduction Large ICT systems into Organisations (Reliability…
INFO 3 C8 - Introduction Large ICT systems into Organisations
Scale
A large system may be computerised and introduced into a single organisation. Can be introduced as from a package or customised from other programs, either by a software house or an organisation's own IT department
Can be introduced across a range of different organisations; the advantage of this being that organisations deal with other members of the same group or others. Therefore data can be moved to each of these units
Reliability and Testing
Many large IT projects fail for various different reasons, some technological, not match new tech (air traffic control) or go over budget (NHS system)
IEEE 829 Testing Standard - Wishes to standardise testing documents that is ;produced a standard set of documents called IEEE 829 Standard for software and test documentation for any type of software testing, including user acceptance testing
it is 8 separate documents that are split into 3 different sections
1) Preparation of Tests
Test plan: how testing will proceed
Test design spec: what needs to be tested
Test cause spec: Create the tests to be run
Test procedure: how tests are run
Test Item Transmittal Report: Specify items released for testing
2) Running the Tests
Test Log: Record the details of tests in time order
Test Incident Report: Record details of events that need investigating
3) Completion of testing
Test Summary Report: summarise and evaluate tests
Program testing - it is difficult to test programming, esp if more that one team has used it. Test Cases tend to be used to check programming; code can be walked through to test
System Testing, Alpha, Beta and User Testing - explained previously
Specialist Testing Groups - British Computer Society has a specialist group on software testing which has 2,500 members. Set up to produce standards of software testing
Regression testing - this goes back to old tests and re-runs them with the changed coding in case the changes cause problems
Performance Testing - tries to emulate many people working on the software in order to check performance
Benchmarking - tests are sometimes carried out using targets: called benchmarks - example database must search for entries in under 30 seconds
Volume Testing - has previously been explained
Environment Testing - testing must be carried out in the environment in which it is going to be used
Installation
Before any system can be installed there has to be a clear plan worked out by a Project Manager. List of Responsibilities and a schedule should all be made clear
Direct Changeover: the old system stops and is immediately replaced, quickest,... no backup to fall on, training is difficult
Parallel Running: Old and new run together and old system is replaced when new is fully working, any issues and old can still be used, low risk method, can be trained on new system... Can be expensive as two data entries, smaller systems may be able to cope
Pilot Running: A single branch or department has it implemented and then it is rolled out to other areas, much easier to stop and control process, easier to monitor the changeover, if the pilot fails then there is low risk of data loss... rollout is very slow, complex as 2 systems will be running, some problems may occur in roll out difficult to find
Phased Changeover: The new system is brought in in stages until the whole system is implemented, structured method allowing each stage to be evaluated before operation... complete system cannot be evaluated until it is fully implemented, complex integrating systems
.
Hardware Installation
- Setting up of hardware and cabling is an important process. May work from existing infrastructure or from new - a schematic must be drawn up to show the position of workstations and cabling: also show the layout of the rooms, power cables and spec for each unit. Explains what each thing does - positioning of cables is important so that power cables and network is not mixed up, also define the routes that cables take this helps to find errors - Power sockets, uninterruptable power units and backup generators are all essential.
Software Installation
- will be the last element in the installation process but may not be easy. Work together to ensure that the correct software is being installed on the right workstations, working with the hardware team. Documentation is essential, for any problem in the future. Test data can be used. Support staff and engineers are on hand to assist with any software problems. They need training first, prior to installation.
The effects of installation
- Some of the jobs will be automated by the new system, so will result in the loss of jobs
May move the office so staff have to move with, other jobs like technicians will be created - a move from operational staff to blue collar workers
People have to learn the new skills of the system and de-skill in what they already knew
Computers work 24hrs a day. So new shifts will be needed
It leads to the flattening of the structure of an organisation
Server Room - must be environmentally controlled with Air conditioning and preferably dust free - note the plans and room layout. It is important as hardware is installed it is tested to make sure it works. Results are also recorded to see what happened during testing
#
Backup and Recovery
Backup is an important part, but in a large scale system it is paramount: losing data can be costly
Essential to preserve data, also important to chose the correct storage devices so they can handle the backup data
Different types of Backup systems
A system that is designed to come into operation as soon as there is a fault. Basically two system that are replicated so everything is backed up and prevents data loss
Disaster recovery data sent via fibre optic - 9/11
More expensive depending on the size of the company, big more expensive, small not as - only a few tapes needed for small big needs a generator and Uninterupred power supply(UPS)
Types of Backup for large systems
Immediate response - Failover systems
Mirror Systems
Snapshots (frozen images of file systems as given time)
Delayed Response - Periodic Backup
Risk Assessment
Setting up a backup system means that the organisation will have to carry out a risk assessment:
A risk grid above gives the level of impact on an organisation and the probability of the risk occurring, elements are calculated to give an overall risk factor
How can data become compromised?
Users tend to unknowingly damage data - bring data from home or to home via a removable drive
Use of emails - they are monitored now, can send information through email, laptops carry business data on them
Risk analysis can be carried out to ID the value of data and the speed in which it would need to be recovered in
Risk analysis model = Assets, Threats, Vulnerabilities -> risks -> counter measures
It is up to the organisation to:
Deter - access rights
Prevent - using a firewall to stop hackers
Correct - using an audit trail to stop unauthorised access to files
Detect - using anti-virus software
Need to create a disaster recovery plan in the hope that it never gets used, but it should be tested on a regular basis to make sure it works for all eventualities
Factors that need considering when drawing up a recovery plan:
The scale of the organisation and how much data is being held
Nature of its operation
Importance of the data it holds
Timescales to recover from disaster
Costs to recover from a disaster
Likelihood of a disaster
It contains:
Who to contact in event of a disaster
a step-by-step list of how system should be recovered
Jobs people need to carry out
All the information needed to get the organisation back up and running
Recovering Large Systems
Following the full or partial failure of a system, recovery takes place. Tested during the backup
ID the facilities available following a disaster
Document what should happen in the event of different disasters
Staff need to be trained in order to know what to do
Also, upload software and OS
Availability of hardware, UPS system etc.
Availability of documentation
estimated time to restore failed system
Maintenance
Corrective Maintenance - correction of errors resulting from system functions not working according to agreed specification
Adaptive Maintenance - adapt the system to changes within an organisation eg changes in VAT
Perfection Maintenance - end user requires additional functionality, not in original specification, or cosmetic changes to improve usability and appearance of UI
Maintenance Releases - a completely new version of the software may be needed