bastion

Business Continuity and Disaster Recovery

Build this skill

Distinguish business continuity from disaster recovery, set RTO and RPO targets and decide the order in which services are restored.

Keeping medicines moving while the laptops are rebuilt

On Sunday morning the Nashik Road counter laptop is in an evidence bag, its replacement is being restored, and the shared drive is still being checked. Customers do not stop needing their blood-pressure tablets because of any of this. Mrs. Lata Gokhale's monthly order is due, and the store opens at nine. Business continuity is keeping an organisation's most important work going, in some form, while something is broken. Disaster recovery is the narrower job of bringing the broken IT systems and data back. Tulsi Health needs both at once: the medicines must keep moving while the computers are rebuilt.

Picture a wedding caterer whose main kitchen floods two days before a big wedding. Disaster recovery is drying, repairing and re-equipping that kitchen. Business continuity is the plan that still feeds 400 guests on time: cook in a partner's kitchen across town, use a simpler menu, hire extra vans. The guests may never know anything went wrong. A good caterer wrote that plan years before the flood, and a good business does the same.

Billing by hand today · medicines as usual Restoring billing from the offline backup 64% · about 3 hours left Farida Bill 0031 Mrs. Gokhale + Disaster recovery IT systems being rebuilt Business continuity the work carries on meanwhile The result medicines still go home

The two jobs happen side by side. The laptop is being recovered while the paper bill book keeps the counter serving customers.

Business continuityQuestion: how do we keep working?Covers people, places, suppliers, ITPaper bills, phone orders, another storeOwned by the business: VikramDisaster recoveryQuestion: how do we get IT back?Covers systems, data, networksBackups, rebuilt laptops, restored appOwned by IT: Neha

Disaster recovery is one part of business continuity. A store that can bill on paper has more time for IT to recover carefully.

Two numbers that shape every plan

Recovery planning starts by asking which activities matter most and what happens to the business for each hour they stop. This exercise is called a business impact analysis. For Tulsi Health the answer is clear: dispensing medicines and billing come first, the customer app next, and old sales reports can wait for days. From that, two numbers are set for each important system.

The recovery time objective (RTO) is the longest a system can be down before the harm becomes unacceptable. Vikram decides that store billing must be back within 12 hours. The recovery point objective (RPO) is how much recent data you can afford to lose, measured as time. If the last good backup was taken at 2 am and disaster strikes at 4 pm, everything from those 14 hours is gone unless it can be re-entered from paper. The RPO therefore depends directly on how often backups are taken.

Billing dataNot yet backed up: lostRestored from 2 am copyBilling systemWorkingDown: being restoredBackStore countersNormalPaper bill bookNormal0 h6 h12 h18 h24 h30 h36 hLast backup, 2 amDisaster, 4 pmRestored

Looking back from the disaster, the gap to the last backup is the data you lose (set by the RPO). Looking forward, the gap until systems return is the downtime (set by the RTO). The paper book covers that gap at the counter.

The small calculation below shows how the backup interval sets the worst-case loss. It works out how many app orders could be lost if disaster strikes just before the next backup. You can press Run without knowing any Python, and you can change the list of intervals on the line marked in a comment and run it again.

python · runs in your browser

Expected output:

text
Hours between backups: 24 | up to 840 orders lost
Hours between backups: 4 | up to 140 orders lost
Hours between backups: 1 | up to 35 orders lost
Line What it does
orders_per_hour = 35 Stores the number of app orders in an average hour
for backup_every in [24, 4, 1]: Repeats the next two lines three times, once for each backup interval in the list
lost_orders = backup_every * orders_per_hour Multiplies the hours since the last backup by orders per hour, the worst case
print(...) Shows the interval and the orders that could be lost

A nightly backup can lose a whole day of orders, while an hourly one loses at most a few dozen, so a tighter RPO simply means backing up more often, at a higher cost.

Recovery planning: what comes back first

A recovery plan is the written, step-by-step order for bringing systems back. Order matters because systems depend on each other. The billing software is useless without the store network and staff logins, and the customer app is useless without its database. So the plan restores the foundations first, then the most critical services, then everything else. It also says where the clean backups are, who does each step, and how long each step should take, so that the RTO is a realistic promise rather than a hope.

Networkstores, headofficeLoginsstaff accountsresetBillingcounter laptopsthenApp and dataorders,databaseThe restemail, archives

Foundations first, then the services customers feel most. Restoring the app before the logins are reset would let an attacker with a stolen password straight back in.

Decide the restoration order yourself. Ask of each system whether customers or other systems need it immediately, or whether the business can manage without it for a few days.

Service restoration and keeping disruption small

Service restoration is the careful return of each system to use. Neha restores from a backup known to be clean, checks the restored system before connecting it, brings services back in stages so that a problem in one does not hide in the rush, and confirms with real users, such as Farida billing a test sale, that it actually works. A backup nobody has ever restored is only a hope, so a restore test every month, timed with a stopwatch, tells Tulsi Health whether its 12-hour RTO is real.

Minimising operational disruption means designing the business so that one failure does not stop everything. Practical measures include a paper bill book and a printed price list at every counter, phone ordering when the app is down, a nearby store that can fill urgent prescriptions, backup copies kept in more than one place with at least one offline, and staff who have practised the manual routine before they need it. Each of these buys time, and time is what lets IT recover carefully instead of in a panic.

A plan that has never been practised usually fails in small, avoidable ways. The bill book is locked in a cupboard whose key is with the manager on leave, the phone-order number is saved only on the laptop that is down, or nobody remembers that the app needs the database before it can start. Twice a year Tulsi Health runs a short tabletop exercise: Neha describes an imaginary disaster, such as the head office losing power for two days, and each person says what they would do, minute by minute. Gaps found around a table cost an afternoon; the same gaps found during a real outage cost days of sales and customers' trust.

The model below lets you set Tulsi Health's backup interval, recovery method and counter fallback, and see the data lost, the downtime and the monthly cost. Try the mission: Vikram's targets are fixed, so you must find a combination that meets all of them within budget.

Follow your curiosity.

↑ ↓ to explore · Enter to open · Escape to close