Search API
This guide provides an in-depth overview of the Search API dashboard, your central hub for monitoring the performance and health of your search queries. Here, you can analyse key metrics like throughput and response time, and review detailed logs of individual search requests to diagnose issues and optimise performance.
Accessing the Search API Dashboard
You can navigate to the Search API dashboard directly from the Home page.
From the Home dashboard, locate the APIs section.
Click on the Search API button to open the dashboard.
Monitoring Search Performance
The Search API dashboard is designed to give you immediate insights into how your search functionality is performing. The page is divided into two main sections: the Search Requests graph and the Request Logs table.
Analysing the Search Requests Graph
The Search Requests graph provides a visual representation of your search query performance and throughput over time.
At the top right of the graph, you can find key real-time metrics:
Throughput: Measured in Queries Per Second (QPS), this metric shows the current volume of search requests being processed.
Response Time: This shows the average time, in milliseconds (ms), that it takes for the API to respond to a search query. Lower values indicate better performance.
To analyse historical performance in more detail:
Hover your cursor over any point on the graph.
A tooltip will appear, showing a detailed breakdown of performance at that specific moment, including the total QPS and the response time percentiles.
Reviewing Individual Request Logs
The Request Logs table, located below the graph, provides a real-time, paginated feed of individual search queries. This is an invaluable tool for debugging issues and understanding user search behaviour.
The table displays the following information for each query:
Time: The exact timestamp when the search request was made.
Search Term: The query string that the user entered.
Strategy: The search methodology used to retrieve the results.
Duration (ms): The total time, in milliseconds, that the query took to execute.
Understanding Search Strategies The
Strategycolumn tells you how the search was processed:
FULL_TEXT: A standard keyword-based search.
SEMANTIC: An advanced search that understands the contextual meaning and intent behind a query, rather than just matching keywords.
HYBRID: A combination of both FULL_TEXT and SEMANTIC strategies to provide the most relevant results.
To navigate through the history of search requests, use the pagination controls located at the bottom of the table. Click the arrow icons to move between pages.
Last updated