Alchuang22-dev opened a new pull request, #18691:
URL: https://github.com/apache/iotdb/pull/18691
## Description
Add a reproducible NLQTSBench evaluation workflow for comparing the IoTDB SQL
and filesystem interfaces, together with a controlled DeepSeek Harness
adapter
that records agent latency, token usage, cache usage, tool calls, tool
errors,
and provider retries.
Extend the filesystem `stats` operation with typed time and TAG filters and
an
allowlisted set of aggregations. This lets an agent delegate deterministic
filtering and aggregation to the filesystem interface instead of retrieving
multiple pages and performing arithmetic in the model context.
### Content
- Add NLQTSBench task selection based on:
- whether the source database can be imported into IoTDB;
- whether both SQL and filesystem interfaces cover the required operations;
- whether IoTDB data-model, SQL-dialect, or schema adaptations are
required.
- Add data migration, TsFile preparation, answer-equivalence verification,
and
server environment manifests.
- Add a controlled DeepSeek Harness adapter that:
- exposes only `iotdb_sql` to the SQL arm or `iotdb_fs` to the filesystem
arm;
- fixes the accessible database and filesystem object path;
- validates typed tool arguments without invoking a shell;
- collects wall time, model time, TTFT, token/cache usage, tool calls,
errors,
and provider retries;
- audits the tool inventory recorded in every canonical session.
- Add structured filesystem pagination with `next_offset`, compact native
JSON
scalar encoding, bounded model-visible output, and complete adapter-side
help.
- Extend `stats` with controlled `start`, `end`, TAG-filter, and aggregate
parameters.
- Support the aggregate allowlist `count`, `min`, `max`, `sum`, `avg`, and
`median` for table, tree, and virtual-directory paths.
- Add NLQTSBench 1–10 diagnostic and ablation reports.
The `typed-filtered-stats` ablation improved accuracy from 4/10 to 10/10
while
reducing tool calls from 167 to 11, total wall time by 90.1%, and reported
token
usage by 97.6%. The current implementation computes the statistics in the CLI
process, so this result measures agent task-decomposition efficiency rather
than storage-engine query acceleration.
<!--
In each section, please describe design decisions made, including:
- Choice of algorithms
- Behavioral aspects. What configuration values are acceptable? How are
corner cases and error
conditions handled, such as when there are insufficient resources?
- Class organization and design (how the logic is split between classes,
inheritance, composition,
design patterns)
- Method organization and design (how the logic is split between methods,
parameters and return types)
- Naming (class, method, API, configuration, HTTP endpoint, names of
emitted metrics)
-->
<!-- It's good to describe an alternative design (or mention an alternative
name) for every design
(or naming) decision point and compare the alternatives with the designs
that you've implemented
(or the names you've chosen) to highlight the advantages of the chosen
designs and names. -->
<!-- If there was a discussion of the design of the feature implemented in
this PR elsewhere
(e. g. a "Proposal" issue, any other issue, or a thread in the development
mailing list),
link to that discussion from this PR description and explain what have
changed in your final design
compared to your original proposal or the consensus version in the end of
the discussion.
If something hasn't changed since the original discussion, you can omit a
detailed discussion of
those aspects of the design here, perhaps apart from brief mentioning for
the sake of readability
of this PR description. -->
<!-- Some of the aspects mentioned above may be omitted for simple and small
changes. -->
<hr>
This PR has:
- [x] been self-reviewed.
- [x] concurrent read
- [x] concurrent write
- [x] concurrent read and write
- [x] added documentation for new or modified features or behaviors.
- [ ] added Javadocs for most classes and all non-trivial methods.
- [ ] added or updated version, __license__, or notice information
- [ ] added comments explaining the "why" and the intent of the code
wherever would not be obvious
for an unfamiliar reader.
- [x] added unit tests or modified existing tests to cover new code paths,
ensuring the threshold
for code coverage.
- [x] added integration tests.
- [ ] been tested in a test IoTDB cluster.
<!-- Check the items by putting "x" in the brackets for the done things. Not
all of these items
apply to every PR. Remove the items which are not done or not relevant to
the PR. None of the items
from the checklist above are strictly necessary, but it would be very
helpful if you at least
self-review the PR. -->
<hr>
##### Key changed/added classes (or packages if there are too many classes)
in this PR
--
This is an automated message from the Apache Git Service.
To respond to the message, please log on to GitHub and use the
URL above to go to the specific comment.
To unsubscribe, e-mail: [email protected]
For queries about this service, please contact Infrastructure at:
[email protected]