Skip to content
Back to blog
HostProbe Domains Internet infrastructure Dataset

HostProbe has identified 668 million domain names

HostProbe's global corpus now contains 668,247,904 unique identified domain names, excluding subdomains.

HostProbe Team

Project update

2 min read

HostProbe has reached an important data milestone: our global domain corpus now contains 668,247,904 unique domain names.

At 668,247,904 unique domain names, we believe HostProbe now maintains the largest publicly known domain corpus of its kind, excluding subdomains.

These are identified domain names, not a claim that 668 million domains are currently registered or active.

Our corpus combines domain names discovered from multiple independent sources. It includes both current and historical names. A domain may have expired, stopped resolving in DNS or disappeared from the web, but evidence that it existed remains useful information.

That distinction is deliberate.

Domain discovery answers one question:

What domain names have existed?

HostProbe’s probing infrastructure answers another:

What is happening with them now?

We independently probe domains and the infrastructure connected to them. DNS observations tell us whether a name currently resolves and where it points. HTTP and HTTPS probing adds another view of the servers and services behind those addresses. As observations accumulate, HostProbe can build a historical picture rather than treating the Internet as a single snapshot.

No single list contains everything

One of the most interesting lessons from building the corpus has been how much independent domain datasets differ.

Large datasets can overlap substantially while each still contains millions of names missing from the other. A newer list is not automatically a superset of an older one, and a larger list is not necessarily complete.

For that reason, HostProbe does not treat any single discovery source as authoritative.

When we encounter another useful source of domain names, we can compare it against the corpus, identify names we have not seen before and expand our historical domain universe without throwing away previous discoveries.

The result is a corpus that can grow over time even as individual domains appear and disappear from the live Internet.

Discovery is only the beginning

A list of domain names is not the finished HostProbe dataset. It is raw material for measurement.

The much larger job is determining which names currently resolve, which addresses they use, which nameservers support them, what infrastructure answers requests and how those relationships change over time.

That work is ongoing.

668,247,904 is our current global baseline.

And it is not intended to stay there.