Data Security and Integrity

Privacy and Security Dangers of Personal Data

Computers are used to store and manage vast quantities of personal data — information that can identify an individual, such as names, addresses, financial details, medical records, and communication history. This creates significant privacy and security risks.

Dangers Arising from Computer-Managed Personal Data

Danger	Description
Unauthorised access	Hackers or malicious insiders may gain access to personal records, leading to identity theft or financial fraud
Data theft	Personal data may be stolen and sold, used for targeted advertising, blackmail, or criminal activity
Misuse by organisations	Companies may use personal data in ways individuals did not consent to (e.g. selling data to third parties)
Inaccurate data	Incorrect records (e.g. wrong credit history) can seriously harm individuals
Aggregation	Combining multiple harmless pieces of data can reveal sensitive information (e.g. combining location, time, and spending patterns)
Data breaches	System vulnerabilities or poor security can expose the personal data of thousands or millions of people at once
Surveillance	Governments or organisations could use computerised personal data to monitor and control individuals

---

Protecting the Security and Integrity of Data

Organisations use a combination of technical and procedural measures to protect data.

Standard Clerical Procedures

Administrative procedures that reduce the risk of data being compromised:

• Clear desk policy — staff should not leave sensitive documents or screens visible when away from their desk.
• Secure disposal — paper records and old storage media must be shredded or destroyed rather than simply discarded.
• Logging and auditing — records of who accessed what data and when, so that suspicious activity can be detected.
• Data minimisation — only collect and store personal data that is genuinely necessary.
• Staff training — employees must be trained on data handling procedures and the risks of social engineering (e.g. phishing).

Levels of Permitted Access

Not all users need access to all data. Access control limits what each user can see and do:

• Role-based access control — access is granted based on job role (e.g. a payroll clerk can access salary data, but a receptionist cannot).
• Need-to-know principle — users are only given access to the data required for their specific tasks.
• Read/write/execute permissions — users may be permitted to read data but not modify or delete it.
• Tiered access levels — different levels of clearance (e.g. standard, manager, administrator) each unlock progressively more sensitive data.

Passwords for Access

Passwords are the most common method of authenticating users before granting access:

• Strong passwords should be long, contain a mix of letters, numbers, and symbols, and avoid dictionary words.
• Password policies may enforce regular changes, minimum length, and complexity requirements.
• Multi-factor authentication (MFA) requires a second form of verification (e.g. a code sent to a phone) in addition to a password.
• Account lockout — accounts are temporarily locked after a set number of failed login attempts to prevent brute-force attacks.
• Password hashing — passwords should be stored as hashed values rather than plain text, so that a database breach does not immediately expose all passwords.

Write-Protect Mechanisms

Write protection prevents data from being modified or deleted:

• Read-only file attributes — files or directories can be marked as read-only so they cannot be overwritten.
• Write-protected storage media — some USB drives and SD cards have a physical write-protect switch.
• Database permissions — database roles can restrict certain users to SELECT (read) operations only, preventing INSERT, UPDATE, or DELETE.
• Backup copies — keeping an unmodifiable backup copy of critical data ensures recovery is possible if the working copy is changed or deleted.

---

Disaster Planning and Contingency Planning

A disaster in computing terms is any event that causes significant disruption to a computer system or data. Organisations must plan for disasters in advance so they can recover quickly.

Potential Threats to Computer Systems

Category	Examples
Natural disasters	Fire, flood, earthquake, lightning strike damaging hardware
Hardware failure	Hard drive failure, server crash, power supply failure
Human error	Accidental deletion of files, misconfiguration of systems
Malicious attacks	Ransomware, hacking, denial-of-service (DoS) attacks
Power failure	Sudden loss of power causing data loss or hardware damage
Theft	Physical theft of computers, servers, or storage media
Software failure	Bugs, corrupted software, failed updates

Contingency Planning

A contingency plan (also called a disaster recovery plan) sets out in advance what an organisation will do if a disaster occurs. The goal is to restore normal operations as quickly as possible and minimise data loss.

Key elements of a contingency plan:

• Risk assessment — identify what threats are most likely and what the impact of each would be.
• Backup strategy — define how often backups are taken, where they are stored, and how they are restored (see below).
• Failover systems — maintain duplicate systems that can take over automatically if the primary system fails (e.g. mirror servers).
• Off-site storage — store backups at a physically separate location so that a local disaster (fire, flood) cannot destroy both the primary system and its backup.
• Recovery time objective (RTO) — the maximum acceptable time for restoring the system.
• Recovery point objective (RPO) — the maximum acceptable amount of data loss, measured in time (e.g. “we can afford to lose up to 4 hours of data”).
• Testing — disaster recovery plans must be regularly tested to ensure they actually work.
• Staff roles — clearly defined responsibilities for who does what during and after a disaster.

---

Backup and Recovery Procedures

Regular backups are the most fundamental protection against data loss.

Types of Backup

Type	Description	Advantages	Disadvantages
Full backup	A complete copy of all data is taken	Simple to restore; everything is in one backup	Takes longest to create; requires most storage
Incremental backup	Only data that has changed since the last backup (full or incremental) is copied	Fast to create; uses less storage	Restoration is slow — requires the last full backup plus every incremental backup since
Differential backup	Only data that has changed since the last full backup is copied	Faster to restore than incremental (only need last full + last differential)	Grows larger over time; slower to create than incremental

Backup Rotation and Grandfather-Father-Son (GFS)

A common backup rotation scheme is Grandfather-Father-Son (GFS):

• Son — daily backups (kept for one week)
• Father — weekly backups (kept for one month)
• Grandfather — monthly backups (kept for one year)

This provides multiple recovery points without requiring infinite storage.

Off-Site and Cloud Backup

• Backups stored on-site are vulnerable to the same local disasters (fire, flood, theft) as the primary data.
• Off-site backups are physically transported to or stored in a separate location.
• Cloud backups are stored on remote servers over the internet, providing geographic separation and easy access.

Recovery

The process of restoring data from a backup must be planned and tested:

1. Identify which backup to restore from (most recent clean backup before the failure).
2. Restore the full backup first.
3. Apply any incremental or differential backups in the correct order.
4. Verify the restored data is complete and uncorrupted.

---

Malicious and Accidental Damage

Data can be damaged or destroyed through both deliberate attacks and unintentional mistakes.

Malicious Damage

Intentional damage caused by people or software designed to harm:

Type	Description	Example Situation
Virus	Self-replicating malicious code that attaches to legitimate files	A user opens an infected email attachment; the virus spreads to other files
Ransomware	Malware that encrypts files and demands payment for the decryption key	A hospital’s patient records are encrypted and become inaccessible
Hacking / unauthorised access	Deliberately gaining access to a system without permission	A hacker breaks into a bank’s database and steals customer account details
Denial-of-Service (DoS) attack	Flooding a server with requests to make it unavailable to legitimate users	A website is taken offline by a flood of fake traffic
Insider threat	A trusted employee deliberately deletes or leaks data	A disgruntled employee deletes critical customer records before leaving
Social engineering / phishing	Tricking users into revealing passwords or installing malware	A fake email impersonates IT support and convinces a user to reveal their password
Physical theft	Stealing hardware containing data	A laptop with unencrypted customer data is stolen

Accidental Damage

Unintentional damage caused by human error or system failure:

Type	Description	Example Situation
Accidental deletion	A user deletes files or records by mistake	An employee accidentally deletes a shared folder
Overwriting data	A user saves over a file, losing the previous version	A document is saved with the wrong filename, replacing an existing file
Hardware failure	A drive or device fails unexpectedly	A hard drive fails and the data it holds is lost
Power failure	An unexpected power cut causes data to be lost or corrupted	A power cut occurs during a database write, leaving the data in an inconsistent state
Software bugs	A bug in an application corrupts the data it processes	An accounting package calculates incorrect totals due to a rounding error in the code
User error during input	Incorrect data is entered	A medical record is saved with the wrong patient’s name

Identifying Situations Where Each Could Occur

• Malicious damage is more likely in environments where data has high value (financial, medical, government), where there are disgruntled employees, or where systems are internet-facing and exposed to external attackers.
• Accidental damage is more likely where staff are poorly trained, where there are no access controls limiting what users can delete or modify, or where hardware is old and unreliable.