TL;DR
Data integrity encompasses the policies and procedures that keep stored information accurate, accessible, and unaltered throughout its lifecycle. Threats come from human error, environmental hazards, and aging infrastructure. Protecting data integrity requires both physical safeguards (like off-site backups) and logical database rules (entity, referential, domain, and user-defined integrity). Routine audits, access controls, and activity logging are essential complementary measures.
Understanding data integrity: Definition and impact
Data integrity is a broad term, and if you put 100 programmers in a room, you'd emerge with 100 definitions. But at a basic level, data integrity refers to procedures and policies that protect data and maintain its accuracy from the moment you get it until you get rid of it.
Omit data integrity, and the consequences can vary. For example, if you limit data integrity on your home computer, you could lose all the work-related files you've stored there. But if your company omits data integrity on company servers, the organization could suffer catastrophic losses. (And some might have hefty fines to pay too.)
What is data integrity?
Solid data integrity policies ensure that you have a safe and secure place to store information. When you need to retrieve something, you can do so without harming it. And the data you pull out is the same as it was when you put it in.
How does data integrity differ from data security?
Data integrity is sometimes confused with data security, but the two are very different things.
| Program Type | Purpose |
|---|---|
| Data security programs | Keep information away from hackers and unauthorized users. |
| Data integrity programs | Ensure that you store data properly and do not damage or alter it while you have it. |
Your data integrity program should protect you from common challenges that stem from:
- Humans. Operator error can cause deletion, corruption, or loss of information.
- Environment. Storms, surges, and other acts of God can harm the physical pieces that store your information.
- Time. Older equipment and obsolete rules can put your data at risk.
Data integrity can seem like a trifling matter compared to all the other challenges IT teams face. But depending on your legislative environment, you may need to prove that you've built a system with data integrity in mind.
For example, Europe's General Data Protection Regulation (GDPR) includes a specific clause about data integrity. If you're subject to this legislation, you must protect your information against loss, destruction, or unauthorized use. Fines can be significant.
How do you protect physical data integrity?
Your planning should begin with the environment that houses your data. Physical data integrity challenges can seem daunting, but with planning, you can overcome them.
Common elements your plan must address include:
- Utilities. The United States has more power outages than any other developed country. Without a plan, the next one could zap your critical data.
- Weather. Floods could short out your backup servers, and extreme heat can tax your fans. Significant events could wipe out buildings altogether.
- Materials. Devices can corrode, fail due to design, or break with age.
Backups form a reasonable solution, but about a third of companies don't use them at all. If you're one of those companies, it's time to revise this policy.
Backing up data allows you to protect your data, and if your backup servers are in another location, you have built-in protections from episode-related issues (like storms).
How do you protect local data integrity?
We collectively generate about 7.5 septillion gigabytes of data each day. Companies don't use every byte they save. But when they want the information they've socked away, they need immediate access. Databases make this possible.
Local data integrity refers to database rules. The regulations can seem complex and unnecessarily arcane. But apply them properly, and they can ensure that your data is both protected and easy to find.
What are the four types of logical data integrity?
There are four types of logical data integrity.
- Entity integrity: Most databases rely on tables to organize data. Entity integrity rules refer to the pieces that make up the tables, including the columns and rows. You shouldn't have more than you need, none should be the same, and none should be blank.
- Referential integrity: These rules refer to data changes, additions, and deletions. You might constrain duplicate data or disallow bits that don't apply. You might also specify data that users can share.
- Domain integrity: Rules regarding how data is input into the database and how it's read belong here. You might limit the type, format, or amount of data you'll accept.
- User-defined integrity: Rules people create for their own purposes belong here.
Local integrity challenges are a bit easier to understand than the rules you'll use for safeguarding. For example, your data could face harm from:
- User error. People could tap in the wrong data in your fields and make something impossible to find.
- Software bugs. A coding problem could allow or disallow something that harms your data.
- Design flaws. The program you use to store your data may not do what you expect it to.
To ensure that you protect your logical data integrity, you'll need to perform routine checks. Pull up data and check it. Search for information and find it.
What else can you try?
Securing and safeguarding your data is critical, and we've given you several ideas you can put to use right away. But there are a few other steps to consider.
Protect your data integrity through:
- Limit. Restrict what people can see within the database. Ensure that you're not allowing everyone who can see data to also change or delete it.
- Check. Validate your data periodically. And walk through your storage facilities regularly.
- Track. Log when items enter your hands, when someone changes the information, and when anyone deletes data.
While all of these steps can protect the integrity of your data, they can't keep the information safe from hackers. We can. Find out more about how Okta can help to improve customer data security.
Frequently asked questions
What is the difference between data integrity and data security?
Data security focuses on keeping information away from unauthorized users and hackers, while data integrity ensures that stored data remains accurate, unaltered, and accessible throughout its lifecycle.
What are the main threats to physical data integrity?
Physical threats include power outages, extreme weather events like floods and heat, and hardware degradation from corrosion or age. Off-site backup servers are a key mitigation strategy.
What are the four types of logical data integrity?
The four types are entity integrity (table structure rules), referential integrity (rules for data changes and deletions), domain integrity (rules for data input format and type), and user-defined integrity (custom rules created for specific purposes).
Why do about a third of companies lack adequate data backup?
According to cited research, roughly one-third of businesses do not use data backups at all, leaving them vulnerable to data loss from physical or environmental incidents.
What regulations require organizations to maintain data integrity?
Europe's General Data Protection Regulation (GDPR) includes a specific clause requiring organizations to protect data against loss, destruction, or unauthorized use, with significant fines for non-compliance.
What practical steps can organizations take to protect data integrity?
Organizations should restrict database access permissions, perform periodic data validation checks, conduct regular walkthroughs of storage facilities, and maintain logs of when data is added, changed, or deleted.
References
The GDPR: Understanding the Six Data Protection Principles. (July 2020). IT Governance.
The US Has More Power Outages Than Any Other Developed Country. Here's Why. (August 2020). Popular Science.
A Third of Businesses Don't Back Up Crucial Data. (April 2016) IT ProPortal.
Companies Collect a Lot of Data, But How Much Do They Actually Use? (August 2019). Priceonomics.
