Info 3 - Introducing large scale systems

Scale:

• Large ICT systems can be used in not only single organisations, but also across a multiple of small organisations.

Reliability and testing

• Testing large ICT systems is a complex task, but it is essential to ensure that the systems operate reliably.

Ensuring that large ICT systems always operate as expected: (What to look for)

- Functional defects : for example the system does not produce the quality of information expected.

- Performance defects : for example if a search produces the required result but does so at a speed that is not acceptable.

- Usability defects : for example if the users find the interface is ineffcient and slows down data entry.

- Security defects : for example if a user can accidentally damage the system.

Reliabilty and testing

Requirements - based testing

This sets benchmarks and provides the basis for testing the software.

• That all requirments are clearly written and not ambiguous
• That all requirements are capable

Program testing

checking the code as it can't be completly error free, applying testing and inspecting the code from small teams that read the code and debug it.

system testing

ensures the system works according to the specification and is done be following the test plan

Reliability and testing

change control

making sure the changes happen in a controlled and coordinated way and are properly documented

Alpha testing

carried out during the development of the software, based on the test plan and data established in the design phase.

Beta testing

software is tested by a selected group of real users who use the software under their normal operating conditions, reporting back any difficulties and comments

Installation

methods of introducing systems

(direct changeover, phased changeover, pilot running, parallel running)

Direct changeover- old system stops and new one is started

• ads- quick, less risk of confusion between old system
• dis- difficult to train staff on the new system before, stressful, risk if doesn't work

Phased changeover- new system bought in stages, replaing a part of the exisitng system

• ads- strucutered, less risky, can train staff
• dis- slow, complex to integrate each phase

Installation

Methods of introucing systems

Pilot running-new stystem repalce the old one but on a small scale such as a single branch of a comapny or location.

• ads- easy control, easy to monitor and evaluate, train staff, low risk
• dis- slow to roll out in future, not show all problems when fully scaled, two different systems

Parallel running- involes the old system nad new system running alongside each other but working independetly.

• ads- esay to compare, can still use old system,can train staff
• dis- expensive to run both systems, confusing staff

Installation

Hardware installation and testing (planning)

• list of responsibilities
• schedule
• diagrams for the system
• hardware specification sheet
• tests

Cabling

• designed and planned by a specialist installer to installer or replace cables

Installation

software installation and testing (tasks in software installation:)

• identifying items of hardware, providing support
• ensuring manuals applicable are available, dry running procedures
• installing the software, conversion from current system
• check accessibility for users, initializing databases and other software with appropriate data

documentation

• document the development throughout, including flowcharts, results fo testing, technical manuals and user guides

resources

be required for the installation, be utitled by the new system, no longer required, need to be improved.

Backup and recovery

disaster recovery strategy- makes plans to avoid data loss and reinstate a working system within an acceptable period of time.

Data in large scale systems:

• inappropriate usage of devices- PDA's and Ipods can introduce viruses and spyware, waste corporate resource and cause data breeches.
• Network data breaches- e-mail and IM disclose personal data, patient information and customer account details.
• Laptop loss/theft- travelling or commuting employees risk the loss of large amounts of proprietary company data.
• Lack of education- employees can be persuaded to disclose sensitive information attackers using social engineering techniques

risk analysis:

• their value to the organisation
• the speed with which they need to be recovered

Backup and recovery

Recovery time objectives- the length of time that it will take to get a system fully operational after a crisis.

Data recovery point- the last point at which data can be recovered, i.e the time lapse between backups.

• Mission critical (e-commerce,email)- failover systems that pick up operation instantly and automatically
• near critical (databases)- replicated systems for quick recovery, regular data snapshots
• important (file systems)- periodic back up to disk
• need protecting (records)- archive, back up to tape

(the higher the bullet point the longer the recovery time and higher cost of protection)

Backup and recovery

Larger scale disasters, accommodation needs to be considered:

• existing accommodation- replacing the damaged equipment in same locaiton
• alternative accommodation- if the whole location is destroyed along with the system, then alternative accommodation for the whole business will be in the disaster recovery strategy.
• reciprocal site- arrangement with another organisation to have a similar system with some spare capacity
• standby site- having a site with a complete and up-to-date second system installed at a different location ready to be used if a disaster happens that is expensive, there are two types (cold standby- duplicate system not used unless needed + Hot standby- duplicate system is up and running as as mirror providing a failover system.
• contract with specialist disaster recovery company- a separate organisation hosts your back up and has the equipment and software for when the back up is needed.

back up and recovery

backup options available for large systems

• failover system- provides continues service, that is immediately switched.
• replicated system provided quick recovery of databases by creating a replica.
• regular snapshots- frozen image of a file system at a given instant time.
• Periodic back up to disk, backup copy to another hard disk.
• tape backup- storing backup portable away from server
• archiving-not needed immediately but must be retained.

backup and recovery

procedures of recovery of large scale systems

• identifying the facilities for providing service in the event of a failure
• process for recovery (types of failures)
• training technical staff who will carry out the recovery
• availability of hardware and software
• back-up electrical power systems
• project time and testing along with appropriate documentation

steps:

• 1- estimate situation
• 2-evaluate available resources
• 3-set-up
• 4-recovery run
• 5-evalue the results
• 6-clean up

backup and recovery

types of maintenance on large scale systems

• corrective maintenance- (traditional maintenance) the deals with errors.
• Adaptive maintenance- required on a regular basis over the life of any development to keep pace with changes in its environment.
• Perfective maintenance- functional enhancements to the system and activities to increase the systems's performance or to enhance its user interface.