Merge branch 'development' into feat/support-otlp-telemetry

.github/workflows/deploy_dev.yml

@@ -6,7 +6,7 @@ on:
 jobs:
   gitlab-dev-deploy:
     if: ${{ github.event.registry_package.package_version.container_metadata.tag.name == 'development' }}
-    uses: epam/ai-dial-ci/.github/workflows/deploy-development.yml@1.10.0
+    uses: epam/ai-dial-ci/.github/workflows/deploy-development.yml@1.11.1
     with:
       gitlab-project-id: '1830'
       gitlab-project-ref: 'master'

.github/workflows/pr-title-check.yml

@@ -9,6 +9,6 @@ on:
 
 jobs:
   pr-title-check:
-    uses: epam/ai-dial-ci/.github/workflows/pr-title-check.yml@1.10.0
+    uses: epam/ai-dial-ci/.github/workflows/pr-title-check.yml@1.11.1
     secrets:
       ACTIONS_BOT_TOKEN: ${{ secrets.ACTIONS_BOT_TOKEN }}

.github/workflows/pr.yml

@@ -6,7 +6,7 @@ on:
 
 jobs:
   run_tests:
-    uses: epam/ai-dial-ci/.github/workflows/python_docker_pr.yml@1.10.0
+    uses: epam/ai-dial-ci/.github/workflows/python_docker_pr.yml@1.11.1
     secrets: inherit
     with:
       python_version: "3.10"

.github/workflows/release.yml

@@ -6,7 +6,7 @@ on:
 
 jobs:
   release:
-    uses: epam/ai-dial-ci/.github/workflows/python_docker_release.yml@1.10.0
+    uses: epam/ai-dial-ci/.github/workflows/python_docker_release.yml@1.11.1
     secrets: inherit
     with:
       python_version: "3.10"

Dockerfile

@@ -6,9 +6,9 @@ ENV PYTHONUNBUFFERED 1
 
 RUN apt-get update && \
     apt-get install -y python3 \
-                       python3-venv \
-                       python3-dev \
-                       python3-pip && \
+                    python3-venv \
+                    python3-dev \
+                    python3-pip && \
     pip install "poetry==1.6.1" && \
     python3 -m venv /opt/venv
 
@@ -49,6 +49,9 @@ ENV PATH="/opt/venv/bin:$PATH"
 
 USER appuser
 
+HEALTHCHECK  --interval=10s --timeout=5s --start-period=30s --retries=6 \
+    CMD wget --no-verbose --tries=1 --spider http://localhost:5000/health || exit 1
+
 # During debugging, this entry point will be overridden. For more information, please refer to https://aka.ms/vscode-docker-python-debug
 EXPOSE 5000
 CMD ["uvicorn", "aidial_analytics_realtime.app:app", "--host", "0.0.0.0", "--port", "5000"]

Makefile

@@ -1,21 +1,29 @@
-PORT = 5001
-IMAGE_NAME = ai-dial-analytics-realtime
+PORT ?= 5001
+IMAGE_NAME ?= ai-dial-analytics-realtime
+VENV_DIR ?= .venv
+POETRY ?= $(VENV_DIR)/bin/poetry
+POETRY_VERSION ?= 1.6.1
 ARGS =
 
 
-.PHONY: all build serve docker_build docker_serve lint format test test_all docs clean help
+.PHONY: all init_env build serve docker_build docker_serve lint format test test_all docs clean help
 
 
 all: build
 
 
-build:
-	poetry build
+init_env:
+	python -m venv $(VENV_DIR)
+	$(VENV_DIR)/bin/pip install poetry==$(POETRY_VERSION) --quiet
 
 
-serve:
-	poetry install --only main
-	poetry run uvicorn aidial_analytics_realtime.app:app --port=$(PORT) --env-file .env
+build: init_env
+	$(POETRY) build
+
+
+serve: init_env
+	$(POETRY) install --only main
+	$(POETRY) run uvicorn aidial_analytics_realtime.app:app --reload --port=$(PORT) --env-file .env
 
 
 docker_build:
@@ -26,24 +34,24 @@ docker_serve: docker_build
 	docker run --platform linux/amd64 --env-file ./.env --rm -p $(PORT):5000 $(IMAGE_NAME):dev
 
 
-lint:
-	poetry install --only nox
-	poetry run nox -s lint
+lint: init_env
+	$(POETRY) install --only nox
+	$(POETRY) run nox -s lint
 
 
-format:
-	poetry install --only nox
-	poetry run nox -s format
+format: init_env
+	$(POETRY) install --only nox
+	$(POETRY) run nox -s format
 
 
-test:
-	poetry install --only nox
-	poetry run -- nox -s tests -- -m "not with_external" $(ARGS)
+test: init_env
+	$(POETRY) install --only nox
+	$(POETRY) run -- nox -s tests -- -m "not with_external" $(ARGS)
 
 
-test_all:
-	poetry install --only nox
-	poetry run -- nox -s tests -- $(ARGS)
+test_all: init_env
+	$(POETRY) install --only nox
+	$(POETRY) run -- nox -s tests -- $(ARGS)
 
 
 docs:

README.md

@@ -4,38 +4,64 @@ Realtime analytics server for [AI DIAL](https://epam-rail.com). The service cons
 
 Refer to [Documentation](https://github.com/epam/ai-dial/blob/main/docs/tutorials/realtime-analytics.md) to learn how to configure AI DAL Core and other necessary components.
 
-# Usage
+## Usage
 
 Check the [AI DIAL Core](https://github.com/epam/ai-dial-core) documentation to configure the way to send the logs to the instance of the realtime analytics server.
 
-The realtime analytics server analyzes the logs stream in the realtime and writes the metric `analytics` to the InfluxDB with the following data:
+The realtime analytics server analyzes the logs stream provided by [Vector](https://vector.dev/docs/reference/configuration/sinks/http/) in the realtime and writes metrics to the InfluxDB.
+
+The logs for `/chat/completions` and `/embeddings` endpoints are saved to the `analytics` measurement with the following tags and fields:
 
 |Tag|Description|
 |---|---|
-|model| The model name for the completion request. |
-|deployment| The deployment name of the model or application for the completion request. |
-|project_id| The project ID for the completion request |
-|language| The language of the conversation detected by the messages content. |
-|topic| The topic of the conversation detected by the messages content. |
+|model| The model name for the request. |
+|deployment| The deployment name of the model or application for the request. |
+|parent_deployment| The deployment name of the model or application that called the current deployment. |
+|execution_path| A list of deployment calls representing the call stack of the request. E.g. `['app1', 'app2', 'model1']` means `app1` called `app2` and `app2` called `model1`. The last element of the list equals to the `deployment` tag. The penultimate element of the list *(when present)* equals to the `parent_deployment` tag. |
+|trace_id| OpenTelemetry trace ID. |
+|core_span_id| OpenTelemetry span ID generated by DIAL Core. |
+|core_parent_span_id| OpenTelemetry span ID generated by DIAL Core that called the span `core_span_id`. |
+|project_id| The project ID for the request. |
+|language| The language detected for the content of the request. |
+|upstream| The upstream endpoint used by the DIAL model. |
+|topic| The topic detected for the content of the request. |
 |title| The title of the person making the request. |
-|response_id| Unique ID of this response. |
+|response_id| Unique ID of the response. |
 
 |Field|Description|
 |---|---|
-|user_hash| The unique hash for the user. |
-|price| The calculated price of the request. |
-|number_request_messages| The total number of messages in history for this request. |
-|chat_id| The unique ID of this convestation. |
-|prompt_tokens| The number of tokens in the prompt including conversation history and the current message |
-|completion_tokens| The number of completion tokens generated for this request |
+|user_hash| The unique hash identifying the user. |
+|deployment_price| The cost of this specific request, excluding the cost of any requests it directly or indirectly initiated. |
+|price| The total cost of the request, including the cost of this request and all related requests it directly or indirectly triggered. |
+|number_request_messages| The total number of messages in the request. For chat completion requests it's number of messages in the chat history. For embedding requests it's number of inputs. |
+|chat_id| The unique identifier for the conversation that this request is part of. |
+|prompt_tokens| The number of tokens in the request. |
+|completion_tokens| The number of tokens in the response. |
 
+The logs for the `/rate` endpoint are saved to the `rate_analytics` measurement:
 
-# Configuration
+|Tag|Description|
+|---|---|
+|deployment| The deployment name of the model or application for the request. |
+|project_id| The project ID for the request. |
+|title| The title of the person making the request. |
+|response_id| Unique ID of the response. |
+|user_hash| The unique hash identifying the user. |
+|chat_id| The unique identifier for the conversation that this request is part of. |
+
+|Field|Description|
+|---|---|
+|dislike_count| 1 for a thumbs up request, otherwise 0. |
+|like_count| 1 for a thumbs down request, otherwise 0. |
+
+## Configuration
 
 Copy `.env.example` to `.env` and customize it for your environment.
 
 ### Connection to the InfluxDB
+
 You need to specify the connection options to the InfluxDB instance using the environment variables:
+
 |Variable|Description|
 |---|---|
 |INFLUX_URL|Url to the InfluxDB to write the analytics data |
@@ -46,6 +72,7 @@ You need to specify the connection options to the InfluxDB instance using the en
 You can follow the [InfluxDB documentation](https://docs.influxdata.com/influxdb/v2/get-started/) to setup InfluxDB locally and acquire the required configuration parameters.
 
 ### Other configuration
+
 Also, following environment valuables can be used to configure the service behavior:
 
 |Variable|Default|Description|
@@ -56,6 +83,7 @@ Also, following environment valuables can be used to configure the service behav
 |LOG_LEVEL|INFO|Log level. Use DEBUG for dev purposes and INFO in prod|
 
 Example of the MODEL_RATES configuration:
+
 ```json
 {
     "gpt-4": {
@@ -85,29 +113,28 @@ Example of the MODEL_RATES configuration:
 }
 ```
 
-
-______
-# Developer environment
+## Developer environment
 
 This project uses [Python>=3.11](https://www.python.org/downloads/) and [Poetry>=1.6.1](https://python-poetry.org/) as a dependency manager.
 Check out Poetry's [documentation on how to install it](https://python-poetry.org/docs/#installation) on your system before proceeding.
 
 To install requirements:
 
-```
+```sh
 poetry install
 ```
 
 This will install all requirements for running the package, linting, formatting and tests.
 
-# Build
+## Build
 
 To build the wheel packages run:
+
 ```sh
 make build
 ```
 
-# Run
+## Run
 
 To run the development server locally run:
 
@@ -117,21 +144,23 @@ make serve
 
 The server will be running as http://localhost:5001
 
-# Docker
+## Docker
 
 To build the docker image run:
+
 ```sh
 make docker_build
 ```
 
 To run the server locally from the docker image run:
+
 ```sh
 make docker_serve
 ```
 
 The server will be running as http://localhost:5001
 
-# Lint
+## Lint
 
 Run the linting before committing:
 
@@ -145,15 +174,15 @@ To auto-fix formatting issues run:
 make format
 ```
 
-# Test
+## Test
 
 Run unit tests locally:
 
 ```sh
 make test
 ```
 
-# Clean
+## Clean
 
 To remove the virtual environment and build artifacts: