Explorer

Explorer is an intuitive web-based interface that displays API usage data gathered by Konnect Analytics from your data plane nodes. You can use this tool to promptly diagnose performance issues, monitor LLM token consumption and costs, or capture essential usage metrics. Explorer also provides the option to save the output as custom reports. You can access the Explorer from any Advanced Analytics chart by clicking through the context (kebab) menu.

Explorer Dashboard

To begin using Explorer, go to the Analytics section and select Explorer. You can select from API Usage or LLM Usage from the dashboard options.

API usage reporting

Grouping and Filtering

Metrics

Time intervals

API Product
API Product Version
Application
Consumer
Control Plane
Control Plane Group
Data Plane Node
Data Plane Node Version
Gateway Services
None
Response Source
Route
Status Code
Status Code Group
Upstream Status Code
Upstream Status Code Group

Metric	Category	Description
Request Count	Count	Total number of API calls within the selected time frame. This includes requests that were rejected due to rate limiting, failed authentication, and so on.
Requests per Minute	Rate	Number of API calls per minute within the selected time frame.
Response Latency	Latency	The amount of time, in milliseconds, that it takes to process an API request. Users can select between average (avg) or different percentiles (p99, p95, and p50). For example, a 99th percentile response latency of 10 milliseconds means that every 1 in 100 requests took at least 10 milliseconds from request received until response returned.
Upstream Latency	Latency	The amount of time, in milliseconds, that Kong Gateway was waiting for the first byte of the upstream service response. Users can select between different percentiles (p99, p95, and p50). For example, a 99th percentile response latency of 10 milliseconds means that every 1 in 100 requests took at least 10 milliseconds from sending the request to the upstream service until the response returned.
Kong latency	Latency	The amount of time, in milliseconds, that Kong Gateway was waiting for the first byte of the upstream service response. Users can select between different percentiles (p99, p95, and p50). For example, a 99th percentile response latency of 10 milliseconds means that every 1 in 100 requests took at least 10 milliseconds from the time the Kong Gateway received the request up to when it sends it back to the upstream service.
Request Size	Size	The size of the request payload received from the client, in bytes. Users can select between the total sum or different percentiles (p99, p95, and p50). For example, a 99th percentile request size of 100 bytes means that the payload size for every 1 in 100 requests was at least 100 bytes.
Response Size	Size	The size of the response payload returned to the client, in bytes. Users can select between the total sum or different percentiles (p99, p95, and p50). For example, a 99th percentile response size of 100 bytes means that the payload size for every 1 in 100 response back to the original caller was at least 100 bytes.

Interval	Description
Last 15 minutes	Data is aggregated in one minute increments.
Last hour	Data is aggregated in one minute increments.
Last six hours	Data is aggregated in one minute increments.
Last 12 hours	Data is aggregated in one hour increments.
Last 24 hours	Data is aggregated in one hour increments.
Last seven days	Data is aggregated in one hour increments.
Last 30 days	Data is aggregated in daily increments.
Current week	Data is aggregated in one hour increments. Logs any traffic in the current calendar week.
Current month	Data is aggregated in one hour increments. Logs any traffic in the current calendar month.
Previous week	Data is aggregated in one hour increments. Logs any traffic in the previous calendar week.
Previous month	Data is aggregated in daily increments. Logs any traffic in the previous calendar month.

System defined groups

Empty - Empty is a system-defined group that indicates that API calls do not have an entity like consumers or routes, selected for grouping. Empty allows you to group API calls that don’t match specific groupings so you can gain more comprehensive insights.

Empty can be used in cases like this:

Identify the number of API calls that don’t match a route.
Identify API calls without an associated consumer to keep track of any security holes.

Empty is displayed in italics in Konnect, and it is not mandatory, using Is Empty or Is Not Empty, you can filter results in Explorer.

LLM usage reporting

Advanced Analytics allows you to monitor and optimize your LLM usage by providing detailed insights into objects such as token consumption, costs, and latency.

With LLM usage reporting, you can:

Track token consumption: Monitor the number of tokens processed by the different LLM models you have configured.
Understand costs: Gain visibility into the costs associated with your LLM providers.
Measure Latency: Analyze the latency involved in processing LLM requests.

To use this feature, navigate to the Explorer dashboard and switch between API usage and LLM usage using the new dataset dropdown. Metrics and groupings will dynamically adjust based on the selected dataset.

Note: LLM usage reporting requires at least Kong Gateway version 3.8.

Grouping and Filtering

Metrics

Time intervals

Application
Cache Status
Consumer
Control Plane
Control Plane Group
Embeddings Model
Embeddings Provider
Provider
Request Model
Response Model
Route

Attribute	Unit	Description
Completion Tokens	Count	Completion tokens are any tokens that the model generates in response to an input.
Prompt Tokens	Count	Prompt tokens are the tokens input into the model.
Total Tokens	Count	Sum of all tokens used in a single request to the model. It includes both the tokens in the input (prompt) and the tokens generated by the model (completion).
Time per Tokens	Number	Average time in milliseconds to generate a token. Calculated as LLM latency divided by the number of tokens.
Costs	Cost	Represents the resulting costs for a request. Final costs = (total number of prompt tokens × input cost per token) + (total number of completion tokens × output cost per token) + (total number of prompt tokens × embedding cost per token).
Response Model	String	Represents which AI model was used to process the prompt by the AI provider.
Request Model	String	Represents which AI model was used to process the prompt.
Provider Name	String	Represents which AI provider was used to process the prompt.
Plugin ID	String	Represents the UUID of the plugin configuration.
LLM Latency	Latency	Total time taken to receive a full response after a request sent from Kong (LLM latency + connection time).
Embeddings Latency	Latency	Time taken to generate the vector for the prompt string.
Fetch Latency	Latency	Total time taken to return a cache.
Cache Status	String	Shows if the response comes directly from the upstream or not. Possible values: `hit` or `Miss`.
Embeddings Model	String	AI providers may have multiple embedding models. This represents the model used for the embeddings.
Embeddings Provider	String	Provider used for generating embeddings.
Embeddings Token	Count	Tokens input into the model for embeddings.
Embeddings Cost	Cost	Cost of caching.
Cost Savings	Cost	Cost savings from cache.

Interval	Description
Last 15 minutes	Data is aggregated in one minute increments.
Last hour	Data is aggregated in one minute increments.
Last six hours	Data is aggregated in one minute increments.
Last 12 hours	Data is aggregated in one hour increments.
Last 24 hours	Data is aggregated in one hour increments.
Last seven days	Data is aggregated in one hour increments.
Last 30 days	Data is aggregated in daily increments.
Current week	Data is aggregated in one hour increments. Logs any traffic in the current calendar week.
Current month	Data is aggregated in one hour increments. Logs any traffic in the current calendar month.
Previous week	Data is aggregated in one hour increments. Logs any traffic in the previous calendar week.
Previous month	Data is aggregated in daily increments. Logs any traffic in the previous calendar month.

Actions

After customizing a view using Explorer’s metrics and filters, you can perform several actions:

Save as a Report: This function creates a new custom report based on your current view, allowing you to revisit these specific insights at a later time.
Export as CSV: If you prefer to analyze your data using other tools, you can download the current view as a CSV file, making it portable and ready for further analysis elsewhere.

Diagnosing latency issues example

Explorer in Kong Konnect gives you the power to monitor your API data in detail and export that data to a CSV file. Let’s go through an example situation where you could leverage custom reports created in Explorer. With this type of report, you can start exploring which upstream service might cause the latency spike.

One way you can build custom reports is by navigating to Analytics in the Konnect menu, then Explorer. This brings you to a page where you can control which analytics data you want to visualize. From here, configure the filter settings as follows:

Show: Line
Date/Time: Last 15 minutes
With: Kong Latency (p95), Upstream Latency (p95)
Per: Minute

Then, you can add a filter for the control plane:

Filter By: Control Plane
Choose Operator: In
Filter Value: prod

Production - Kong vs Upstream Latency (last hour)

Figure 1: Line chart showing average upstream and Kong latency over the last hour.

More information

Custom Reports Use Case - Analyze API Usage and Performance: Learn how to analyze API usage and performance using custom reports.