Quickstart
Concept syntax for indexing one NTFS volume and opening an interactive query session.
hdt scan C: --mode mft --incremental hdt catalog status hdt query "ext:pdf size:>50MB"
Query language
Predicates are designed to remain readable while still mapping cleanly to indexes.
| Predicate | Example | Meaning |
|---|---|---|
| ext | ext:pdf | extension equals PDF |
| size | size:>1GB | files larger than 1 GB |
| path | path:*archive* | path contains archive |
| volume | volume:disk21 | specific catalog volume |
| offline | offline:true | currently detached sources |
| dupes | dupes:>1 | content has duplicates |
Dedup pipeline
hdt dedup --scope "path:V:\SharePoint" --group size --sample 1MB --hash blake3 --verify exact
Benchmark
hdt benchmark ingest --volume O: --threads 16 --json result.json hdt benchmark query --suite large-catalog --iterations 1000
WebFarm named pipes
The server can expose numbered listeners such as WebApiServer001 and WebApiServer002, with multiple async instances behind each logical name.
The website you are viewing is intended to travel through that same request/response path in the Windows deployment.
FAQ
Yes, if its catalog snapshot is retained. Content-dependent operations should report that the source volume is offline.
No. The included SQL only replaces harddisk.tools content rows and updates domain metadata.
No. They are illustrative UI data for this test site and should later be replaced by reproducible measurements.