In addition to using the Cloud Function Compute (CFC), you can also realize it by adding codes you need in the custom storage volume and calling one of the methods.

Before the operations described herein are started, create a core according to Getting Started - Build a Core.

If you have already operated according to How to configure multiple functions in the Cloud Function Compute (CFC), you only need to create the two storage volumes of bie-person-compute-code and bie-person-compute-conf in [Custom Storage Volume](#Create a custom storage volume), and the previous storage volumes can be reused for others.

In this article, it takes Python 3.6 as an example for explanation.

Create a Storage Volume for Computing Services

Create a Custom Storage Volume

Enter the Baidu IntelliEdge storage volume page, and two custom storage volumes are required - bie-person-compute-code and bie-person-compute-conf, which are used to save code and configuration respectively.

bie-person-compute-codeStorage Volume

First, click “Create a storage volume” button on the storage volume page, select the template custom storage volume, name it bie-person-compute-code, and click “OK”.

Enter the custom storage volume page, and select “Create an editable file”. Here, a total of 2 files have to be created.

Create the first file add2.py, and the code content is as follows:

# -*- coding: utf-8 -*- 

def add2(event, context): 
    result = {} 
    result['result'] = event['x'] + event['y'] + event['z'] 
    return result 
    

Save this file, and create anotheradd.py file. The code content is as follows:

# -*- coding: utf-8 -*- 
import add2 

def add(event, context): 
    result = {} 
    result['result'] = event['x'] + event['y'] + event['z'] 
    result['result2'] = add2.add2(event, context)['result'] 
    return result 

Two files -add.py and add2.py are created. Check that they are correct, and click “Release a new version”.

bie-person-compute-confStorage Volume

On the storage volume page, click “Create a storage volume” button, name it bie-person-compute-conf, and select baetyl template configuration for the template.

The storage volume generates the service.yml file for us by default. Click “Edit”, enter the following configuration and save it.

functions: 
  - name: 'add' 
    handler: 'add.add' 
    codedir: 'var/db/baetyl/compute' 
  - name: 'add2' 
    handler: 'add2.add2' 
    codedir: 'var/db/baetyl/compute' 

Lastly, you have to release the new version.

baetyl-hubCreate a Storage Volume for Services

There are three storage volumes, namely:

• bie-demo-hub-conf
• bie-demo-hub-log
• bie-demo-hub-data

For details, see Release and Distribute a Configuration. You only need to create these three storage volumes.

baetyl-function-managerCreate a Storage Volume for Services

On the storage volume page, click “Create a storage volume” button, and select baetyl template configuration for the template.

Enter the bie-cfc-func-conf storage volume, modify the service.yml, enter the following configuration and save it.

hub: 
  address: tcp://local-hub:1883 
  username: test 
  password: hahaha 
rules: 
  - clientid: localfunc-1 
    subscribe: 
      topic: t 
    function: 
      name: add 
    publish: 
      topic: t/add 
  - clientid: localfunc-2 
    subscribe: 
      topic: t 
    function: 
      name: add2 
    publish: 
      topic: t/add2 
functions: 
  - name: add 
    service: function-compute 
  - name: add2 
    service: function-compute 
logger: 
  path: var/log/baetyl/service.log 
  level: "debug" 

Similarly, you have to release the new version.

Add/Create a Service

All the 6 storage volumes to be used are created.

Then, enter the Baidu IntelliEdge core page, select the core created in our first step, and configure it.

baetyl-hubService Configuration

Click “Add a service” button.

Name it local-hub, select baetyl-hub as the module, enter 1883:1883 for port mapping, and then click storage volume mounting to mount the three corresponding storage volumes.

It is not mandatory for the mapping port to be 1883, but the specific mapping port should be consistent with the configuration file of each service. For example, for tcp connection, the local-hub service monitors the port 8883, and when other services establish connection with the local-hub service, set the connection port to be 8883.

Mount three storage volumes - bie-demo-hub-conf, bie-demo-hub-log and bie-demo-hub-data.

Finally, you can see that a total of three storage volumes are mounted.

baetyl-function-manager Service Configuration

Click “Add a service”, name it local-function-manager, and select baetyl-function-manager for the module.

Mount the bie-cfc-func-conf storage volume.

baetyl-function-python36 Service Configuration

Click “Add a service”, name it function-compute, and select baetyl-function-python36 for the module.

Mount the bie-person-compute-conf and bie-person-compute-code for the storage volumes.

View the Final Configuration Result and Distribute it

The service configuration is completed, and our core services are as follows:

Check that the service configuration is correct, and click “Release” button to release this version.

After the version is released, you can see that the latest version of the core is now the “v2” version.

If the core has been downloaded and started, click “Distribute” button directly, and wait for restart to upgrade.

If the core has not been downloaded, you can click “Download” button. Note that the master program should also be selected to download.

Lastly, use mqttbox to send messages for testing. For the use of mqttbox, please see [Use MQTTBox to Subscribe Data Check Results](BIE/Operation Guide/Core management/Typical Configuration.md#Use MQTTBox to Subscribe Data Check Results). Our input example is {"x":1,"y":2,"z":3}, and the result is shown in the figure. The output result is exactly the result computed in the functions add and add2 in the custom storage volume.