Fortanix Confidential Computing Manager - Simple Use Case - Python

1.0 Introduction

This article describes a simple example of how you can use Fortanix Confidential Computing Manager with Python to streamline data analysis within the airline industry.

As a dynamic player in the airline industry, you are constantly looking for innovative methods to improve the customer experience and drive revenue growth. Utilizing the capabilities of Fortanix Confidential Computing Manager, you are ready to revolutionize your data analysis practices and unlock valuable insights hidden within your datasets.

2.0 Contextualizing Computing Manager

Given the complexities of airline operations and customer interactions, you have decided to explore Fortanix Confidential Computing Manager. After releasing the potential of this application, you envisioned a future where data-driven decisions drive strategic initiatives, resulting in improved operational efficiency and customer satisfaction.

image (8).png

Figure 1: Architecture Diagram

To achieve optimal outcomes in airline data analysis, it is imperative to follow a structured approach. The following procedure outlines key steps to leverage Fortanix Confidential Computing Manager effectively, ensuring secure and insightful data exploration:

  1. Step 1: Accessing Data Sources - In this step, you will begin by accessing data from various sources securely and efficiently, laying the foundation for your analysis.
  2. Step 2: Data Analysis: Unveiling Insights - This step involves delving into your data to uncover valuable insights that can inform decision-making and improve operational efficiency.
  3. Step 3: Exploring Data with Python Scripts - Here, you will utilize Python scripts to explore your datasets in detail, extracting meaningful information to gain a deeper understanding of your data.
  4. Step 4: Aggregating Data for Comprehensive Analysis - This step focuses on merging datasets to create a comprehensive view, enabling more thorough analysis and interpretation of the data.
  5. Step 5: Exporting Analyzed Results - After your analysis is complete, you will securely export the analyzed results to share them with relevant stakeholders, ensuring data privacy and compliance.
  6. Step 6: Workflow Configuration and Execution - Finally, you will configure and execute workflows to streamline data processing and analysis, optimizing your use of the Fortanix Confidential Computing Manager platform.

3.0 Accessing Data Sources

To gather your real-world data effectively, you will utilize Inbound Data Connectors to access information from diverse sources, including CSV files and Google BigQuery. These connectors serve as gateways to collaborate data into the Fortanix Confidential Computing Manager, ensuring seamless integration and accessibility of your datasets for further analysis and processing.

3.1 Inbound Connector 1: HTTPS-CSV

To start, fetch the customer data from a CSV file securely hosted online. It provides access to essential details such as customer contact information and sales data pertinent to the airline industry.

3.2 Inbound Connector 2: HTTPS-CSV

Next, retrieve the customer data from a CSV file securely hosted online. It grants access to vital information regarding media customers, including contact details and relevant sales data.

  • Data Connector Type: CSV
  • Connector Name: media_customers
  • Description: This connector seamlessly accesses and integrates the data from an online CSV file, specifically tailored to media customers. You can gain insights into customer behavior, preferences, and engagement metrics to optimize marketing strategies and enhance customer satisfaction within the media industry.
  • Labels: (optional)
  • URL: 
    Ensure to obtain a valid and pre-signed URL for accessing the data if you want to fetch the data from a URL.  This URL can be hosted on Amazon Web Services (AWS), Microsoft Azure, or the Google Cloud Platform (GCP).

4.0 Data Analysis: Unveiling Insights

After securely importing your data into Fortanix Confidential Computing Manager, it is time to delve into the analysis. This involves merging various datasets such as airline customer data and media customer data. This is achieved by using Python scripts, which allow you to handle and process the data effectively.

4.1 Media Customer Data Table

Email Phone Number Status Age Band Sales Date Campaign Sec View Cost 000-000-0005 MEMBER 50 02-04-2021 campaign_1 25 2.31 001-001-0011 MEMBER 60 05-05-2021 campaign_1 35 0.06 002-002-0023 MEMBER 65 21-04-2021 campaign_1 39 0.03 003-003-0033 SILVER 30 18-04-2021 campaign_1 15 2.5 004-004-0042 GOLD 60 16-04-2021 campaign_3 34 2.75

4.2 Airline Customer Data Table

Email Phone Number Status Age Band Sales Date Campaign Sec View 000-000-0007 1A 51500 product_5 02-04-2021 43.57 001-001-00110 2A 20500 product_3 05-05-2021 97.3 010-010-0108 1A 77300 product_5 21-04-2021 94.33 100-100-1005 1A 77800 product_2 18-04-2021 75.39 101-101-1011 1A 55300 product_4 16-04-2021 36.27

5.0 Aggregating Data for Comprehensive Analysis

To streamline your analysis, merge the airline customer data and media customer data into a single dataset named Airline_Media_Customers_Python. This dataset will facilitate a comprehensive analysis of customer demographics, sales performance, and campaign effectiveness.

This Python script combines data from multiple tables, airline_customers and media_customers, using JOIN operations. It selects distinct customer IDs along with their contact information, sales data, and campaign details.

Table Name: JOIN

Query Language: Python

import pandas as pd
from typing import Dict

def process(data_frames: Dict[str, pd.DataFrame]) - pd.DataFrame:
    airline_df = data_frames["airline_customers"]
    media_df = data_frames["media_customers"]

    # Join DataFrames on EMAIL
    merged_df = airline_df.merge(media_df, on='EMAIL', how='inner')

    # Create desired column names with prefixes
    merged_df.rename(columns={'PHONE_x': 'Airline_Phone', 'PHONE_y': 'Media_Phone',
                               'ZIP': 'Airline_ZIP', 'PRODUCT': 'Airline_Product',
                               'SLS_DATE_x': 'Airline_Sales_Date', 'SLS_DATE_y': 'Media_Sales_Date'},

    # Select desired columns
    result_df = merged_df[['EMAIL', 'Airline_Phone', 'Zone', 'Airline_ZIP',
                           'Airline_Product', 'Airline_Sales_Date', 'SALES_DLR',
                           'Media_Phone', 'STATUS', 'AGE_BAND', 'Media_Sales_Date',
                           'CAMPAIGN', 'SEC_VIEW', 'COST']]

    # Display the result
    return result_df

6.0 Exploring Data with SQL Scripts

This section delves into Python scripts to explore your data more deeply. These scripts help you find specific information you may be interested in, like customer demographics and purchasing patterns.

6.1 SQL Script 1: Campaign Cost

This script calculates the total cost spent on campaigns.

  • Name: Campaign Cost
  • Description: Calculates total expenditure on campaign.
  • Group: CCM_DEFAULT
  • Query Language: SQL Aggregate
    SELECT SUM(COST) AS total_cost_spent
    FROM python_script_output;

6.2 SQL Script 2: Common Age Group

This script helps you understand the common age groups among your customers.

  • Name: Common Age Group
  • Description: Identify the most common age group among customers.
  • Group: CCM_DEFAULT
  • Query Language: SQL Aggregate
    SELECT COUNT(AGE_BAND) AS common_age_band_count 
    FROM python_script_output 
    ORDER BY common_age_band_count DESC 
    LIMIT 1;

6.3 SQL Script 3: Sales Revenue

This script determines the average sales revenue per campaign.

  • Name: Sales Revenue
  • Description: Calculate the revenue earned through sales.
  • Group: CCM_DEFAULT
  • Query Language: SQL Aggregate
    SELECT SUM(SALES_DLR) AS total_sales_revenue
    FROM python_script_output;

7.0 Exporting Analyzed Results

After completing your analysis, it is essential to securely share your findings with the relevant stakeholders. To accomplish this, you will need to set up Outbound Connectors within the Fortanix Confidential Computing Manager. These connectors enable you to export your processed data securely to external systems or stakeholders. You can configure the Outbound Connectors with specific details such as the type of connector, description, and destination URL to ensure the safe transfer of data.

7.1 Outbound Connector 1: HTTPS-CSV

This connector securely transmits data to the designated Customers_Output CSV file. It enables seamless integration for exporting data to the specified destinations while ensuring data security and integrity.

The following is the output of Python queries, each offering insightful metrics regarding email counts within the dataset:

"SELECT SUM(SALES_DLR) AS total_sales_revenue
FROM python_script_output;",38158.58
"SELECT COUNT(AGE_BAND) AS common_age_band_count 
FROM python_script_output 
ORDER BY common_age_band_count DESC 
LIMIT 1;",435
"SELECT SUM(COST) AS total_cost_spent 
FROM python_script_output;",1414.81

8.0 Workflow Configuration and Execution

Now that all your data connectors and scripts are set up, it is time to create and fine-tune your workflow to ensure smooth data processing and analysis.

8.1 Creating a Workflow

Create a workflow in the Fortanix Confidential Computing Manager UI. For detailed information, refer to the Creating a Workflow documentation.

Screenshot (220).png

Figure 2: Drafting the Workflow

8.2 Configuring the Workflow

After you have added the inbound connector, scripts, and outbound connectors in the Fortanix Confidential Computing Manager UI, now you must place them in the workflow working area and connect them. For detailed information, refer to the Configuring a Workflow documentation.

Screenshot (221).png

Figure 3: Connect the Nodes

8.3 Requesting and Approving the Workflow

With your workflow configured, you need to send a request for approval and wait for them to review the workflow approval request and either approve or decline as necessary. For detailed information, refer to the Requesting the Workflow Approval documentation.

Screenshot (222).png

Figure 4: Approve the Worklfow

8.4 Running the Application Workflow

After your workflow is approved and finalized, you can proceed with running the ACI application workflow. For detailed information, refer to the Running the ACI Application Workflow documentation.

Screenshot (223).png

Figure 5: Run the Workflow

Congratulations! Your efficient use of Fortanix Confidential Computing Manager has enhanced your healthcare data analysis workflows. Through secure data collection, thorough analysis, and smooth export processes, you have acquired invaluable insights. These insights will support better decision-making and improve patient outcomes across your organization.


Please sign in to leave a comment.

Was this article helpful?
0 out of 0 found this helpful