How to Filter a Multi-Cloud Database Inventory That Scales
At ten database instances, a spreadsheet works fine. You can see everything on one screen, and nobody needs to search for anything.
At fifty instances spread across three AWS accounts, two GCP projects, and a MongoDB Atlas organization, the spreadsheet stops working. Not because the data is wrong — because finding the one instance you actually need now means scanning a wall of rows, or asking in Slack whether anyone remembers which account the staging replica lives in.
This is the point where a database inventory needs to be filterable, not just visible.
Why Visibility Alone Stops Being Enough
A unified inventory that shows every instance across every provider is a real improvement over checking three separate consoles. But once the list gets long, an unfiltered list has the same problem the three consoles had — you still have to look everywhere to find one thing.
The failure shows up in small, recurring moments:
- An engineer needs the production MongoDB Atlas cluster in the EU region and scrolls past forty other instances to find it.
- A team lead wants to see every database in a specific AWS account before an access review, and has no way to isolate just that account.
- Someone needs to know which instances are currently stopped versus running, across every provider, before a cost review.
None of these are edge cases. They're the normal, weekly shape of working with infrastructure that spans more than one cloud.
What Filtering a Multi-Cloud Inventory Actually Means
1DataCloud's Unified Dashboard inventory can be filtered by four dimensions that map directly to how teams actually think about their infrastructure:
| Filter | What It Narrows Down |
|---|---|
| Account | Which AWS account, GCP project, or MongoDB Atlas organization the instance belongs to |
| Region | Where the instance is physically running (us-east-1, europe-west1, and so on) |
| Provider | AWS, GCP, or MongoDB Atlas (on-premise instances are included in the same inventory view) |
| Status | Whether the instance is currently running, stopped, or in another reported state |
These four filters combine. You can narrow the inventory to "every running instance in a specific GCP project," or "every MongoDB Atlas cluster in a specific region," in a couple of clicks — without writing a query or opening a provider console.
This is deliberately a small, well-defined set of filters. It's not a per-field search across every attribute a cloud provider happens to report — it's the four dimensions that actually correspond to how infrastructure is organized: by who owns it, where it runs, which provider hosts it, and whether it's currently active.
What the Inventory Shows You Once You've Found It
Filtering gets you to the right instance. Once you're there, the inventory table shows the fields the Unified Dashboard already discovered when it synced with each provider — displayed as plain columns, side by side, regardless of which cloud the instance lives in:
- Engine and engine version
- Region and availability zone
- Instance type or tier
- Backup retention setting
- Whether the instance is publicly accessible
- Current status
These are static, refresh-on-sync values — a snapshot of what each provider reported the last time the dashboard synced, not a live monitor watching for changes. Nothing here is flagged, scored, or alerted on. The value is that an AWS RDS instance, a GCP Cloud SQL instance, and a MongoDB Atlas cluster all show these same categories of information in one table, instead of three consoles with three different layouts and three different sets of field names.
A Practical Example
Say your team is preparing for a quarterly access review and needs to look at every instance in one specific AWS account before deciding what access to revoke.
Without a unified, filterable inventory, this means logging into that specific AWS account's console, navigating to RDS, and manually cross-referencing against whatever list of "who should have access" exists elsewhere.
With the filtered inventory: select the account, see every instance that belongs to it, and read off engine, region, and public-accessibility for each one directly from the table. The same approach works for "every stopped instance across all providers, this quarter" or "every production-region instance in GCP" — the filters narrow the list, and the columns answer the question once you're looking at the right rows.
Why This Matters More as You Grow
The value of inventory filtering scales with the size of the infrastructure, not the other way around. A team with five databases barely notices it exists. A team with two hundred instances across four providers and a dozen accounts depends on it every single day — because at that scale, the alternative isn't "check the console instead," it's "nobody actually knows what's running."
What This Solves in Practice
"What's running in eu-west-1?"
Before: open each provider's console, navigate to the relevant service, filter manually, repeat per account.
After: filter the 1DataCloud inventory by region = eu-west-1. Done.
"Everything in this one AWS account, for an access review."
Before: log into that specific account's console, cross-reference against an access list kept elsewhere.
After: filter by account, read engine/region/public-access straight off the table.
"Is this a per-field search across every attribute?"
No, intentionally not. It's four filters — account, region, provider, status — not a general-purpose query builder over every field a cloud API happens to return.
Frequently Asked Questions
Can I filter by engine version, like MySQL 5.7 versus 8.0?
Not as a separate filter today. Engine version is shown as a column in the inventory table for every instance, so you can read it directly, but the four filter dimensions are account, region, provider, and status.
Does filtering change what data is synced, or just what's displayed?
Just what's displayed. The Unified Dashboard syncs the full inventory from each connected provider; filters narrow the view, they don't change what's been discovered.
Is the inventory updated in real time?
No — it reflects the state at the last sync, not a continuously monitored live feed. For most infrastructure visibility use cases, a refresh-on-sync snapshot is sufficient; it is not intended as an uptime or alerting system.
Does this work for on-premise databases too?
Yes. On-premise MySQL and PostgreSQL instances appear in the same inventory alongside AWS, GCP, and MongoDB Atlas instances, and are included in the same filtering.
Try It
If you're managing database instances across more than one provider and the inventory has grown past the point where scrolling works, 1DataCloud is worth trying.
Free plan: connect up to 3 instances across any cloud. No credit card.
How many instances does your team track before a spreadsheet stops being viable? Curious where that line actually falls for other people — drop it in the comments.
