PHPMem v2.0.1
Version
1.6.45
Uptime
7 days 9 hours 20 minutes 46 seconds
Memory
Total
512MB
Used
13,32MB (2.6%)
Free
498,68MB
Keys
Current
10 162
Total (since start)
11 092
Evictions
0
Reclaimed
157
Expired Unfetched
0
Evicted Unfetched
0
Connections
Current
3 / 1 024 max
Total
64 950
Rejected
0
llm:fd93f07c0aa9c25a949a7d4269ce8a2854ff060037f4a29221e38943cd983903
Edit
Based on the dataset's overview and topology, here is a summary of what this dataset is about and what its records represent:
### What is this dataset about?
This dataset contains the **internal telemetry and metadata catalog of a data intelligence platform**. The platform's primary function is to profile external datasets, classify their structures, and track the workflows used to analyze them.
The architecture is split into two main areas:
1. **Catalog Metadata:** Information about what was profiled (e.g., external datasets, tables, and columns).
2. **Execution Telemetry:** Operational data about how and when the profiling occurred, including orchestration (batches and analysis runs) and AI-assisted tasks (LLM calls for classification and tagging).
*Note on Data Quality:* The dataset is currently highly fragmented. While there are 74 detected relationships across its tables, none are currently join-viable. This means cross-functional analysis (like tying LLM costs directly to specific dataset characteristics) is currently difficult without manual key matching.
### What does a single row represent?
Because this is a multi-table relational database (containing 51 tables and over 2 million rows), a single row represents a different entity depending on the table you are looking at.
Here is what a row represents in the five largest tables, which together make up 96% of the dataset:
* **`columns_catalog` (74.6% of total rows):** A single row represents **metadata for one specific column** found within an external dataset that the platform has profiled. This is the central repository of the dataset.
* **`job_logs` (14.8% of total rows):** A single row represents **one execution event or log entry** generated by the platform's processing pipeline.
* **`llm_calls` (2.4% of total rows):** A single row represents **one request/response cycle to a Large Language Model**, capturing the AI-powered classification and tagging activity.
* **`column_tags` (2.1% of total rows):** A single row represents **one semantic category or tag** applied to a specific column to give it business meaning.
* **`tag_lineage_edges` (2.0% of total rows):** A single row represents **one dependency link** tracing the operational history between a profiling run and its output.