Cloud cost and recovery for a small operations team

What should a small team actually manage in the cloud?

A small operations team does not need a cloud program. It needs to know what is running, what it costs, and how to get the business back if the main system is unavailable. Everything else can wait.

Which costs are worth looking at first?

Start with the bill lines that move, not with a full tagging project.

• Environments that were left on after a test.
• Storage that grows because nothing is archived.
• A database or app tier sized for a peak that never arrives.
• Data transfer you cannot explain to the person who pays the invoice.

Turn off what is unused. Resize what is oversized. Put a monthly number next to the systems the business actually depends on, so a spike is visible before it is a surprise.

What does recovery mean if you are not a large IT department?

Recovery means a named person can restore the system the business runs on, to a known point, in a time the business has agreed. That requires three facts:

1. Where the backup lives, and that someone has restored from it.
2. How long you can afford to be down, and how much recent work you can afford to redo.
3. Who is called, and which system comes back first.

A backup that has never been restored is a hope. A restore drill, even once a quarter on a non-production copy, is the evidence.

What can you postpone?

Multi-region designs, detailed chargeback, and a large set of accounts can wait until the simple bill and the simple restore are boring. Boring is the goal.

Checklist

• Monthly cost of the systems the business depends on is known
• Unused environments are off
• Backup location is named
• A restore has been done, not only scheduled
• First system to bring back, and the person who does it, are written down

---