Enterprise Computing Systems: Concepts, Traps and a Worked Rule
What you must understand
Enterprise computing is not simply "a big network". It is the integrated set of hardware, software, data, people and processes that supports an organisation's core operations across many sites and users. Be able to explain the layers: client devices, the network and transmission media, servers, storage, and the application and data layers. Know the difference between a LAN, WAN and the internet, and between centralised, distributed and cloud-hosted architectures.
You must also be able to discuss:
- Scalability and availability – redundancy, failover, load balancing and disaster recovery.
- Virtualisation and cloud models – IaaS, PaaS and SaaS, and public, private and hybrid deployment.
- Data management – relational databases, data warehouses, backup types (full, incremental, differential) and retention.
- Security – authentication, authorisation, encryption, firewalls, access control and audit logs.
- Integration – APIs, middleware and enterprise resource planning systems.
- Social and ethical issues – privacy, data sovereignty, accessibility, and the digital divide.
Traps in questions
- Confusing backup with archiving, or fault tolerance with disaster recovery.
- Treating cloud as automatically cheaper or more secure; answers must weigh trade-offs.
- Confusing authentication (who you are) with authorisation (what you may do).
- Ignoring non-technical factors such as training, policy, cost and change management.
- Describing a technology without linking it to the organisation's stated need.
Rule of thumb
For every scenario, answer in three moves: identify the requirement (e.g. 24/7 uptime), name the mechanism (e.g. redundant servers with automatic failover), and justify the fit (minimal downtime, though higher cost). Always add one limitation.
Worked example
A retailer's point-of-sale system fails during peak trading. A strong answer identifies availability as the requirement, proposes redundant servers plus automated failover and regular tested backups, and notes the trade-off of cost and complexity.