How Promevo's Cloud Migration Services Improve Efficiency & Reduce Cloud Spend

Businesses today are constantly seeking ways to streamline their operations and maximize efficiency. One technology that has gained significant traction is serverless computing, which allows developers to build and run applications without provisioning or managing servers or backend infrastructure.

Last year, Promevo and Google Cloud held a virtual workshop exploring how to build scalable, secure applications in the Google Cloud Platform (GCP) using Cloud Run and other serverless technologies.

During that session, one of our certified Google Cloud Architects presented a case study of a Promevo client who successfully transformed their API deployment strategy with the help of serverless architecture. 

By moving their API to the cloud and leveraging the power of Google Cloud's innovative services such as Cloud Run, this customer achieved significant improvements in scalability, reliability, cost-efficiency, and developer satisfaction.

Watch the video below for the full demonstration and then join us as we delve into the details of this client's journey towards a more seamless and efficient API deployment process.

Understanding the Problem 

The customer, whose identity remains confidential, had initially been operating their API on a single Ubuntu VM in another cloud on the East Coast. However, they encountered numerous challenges related to updating, scalability, and overall management of the VM. 

Frustrated by the limitations and complexities of their existing architecture, they sought a more robust solution that would support their anticipated growth and handle increasing traffic from their frontend applications.

Architecture Overview

The customer's existing architecture was relatively simple: a RESTful API built with Express.js, multiple instances of Node.js running on a single Ubuntu VM, and NGINX acting as a load balancer. 

To manage their database needs, they utilized a medium-sized Mongo Atlas cluster deployed on Google Cloud Platform (GCP). One notable concern with their setup was the fact that their MongoDB database had public internet access, albeit with IP whitelisting.

Deployment Challenges

The deployment process for the customer's application was time-consuming and prone to human error. It involved manually zipping up their code, transferring it to the server, unzipping it, and restarting the instances one by one.

This process often led to significant deployment times, ranging from two to three hours, and increased the risk of mistakes and downtime. Moreover, the customer was struggling with vertical scaling, increasing the VM size to accommodate peak load and wasting resources during low traffic periods.

Developing the Solution

To address the customer's deployment challenges, Promevo’s Google Cloud Architects proposed a migration to a serverless architecture using Google Cloud's Cloud Run. Cloud Run enabled a shift towards a more agile deployment process, significantly reducing deployment times and eliminating the need for manual intervention. 

By containerizing their Node.js API using Docker and implementing a CI/CD pipeline, the customer achieved automated and efficient deployment based on branches, tags, and merges from their GitLab repository.

Optimizing Security & Efficiency

To enhance security, the customer's MongoDB database was migrated to a Google Cloud VPC through a peering connection with Google Cloud's MongoDB Atlas. This migration not only improved security by isolating the database within an internal network but also reduced egress costs from 12.5 cents per gigabyte to 1 cent per gigabyte. 

Additionally, the customer took advantage of Google Cloud services such as Cloud Armor, global load balancer, and Secret Manager to ensure comprehensive security measures and streamlined API access.

Achieving Scalability & Performance

With their API now running on Cloud Run, the customer experienced seamless auto-scaling to handle increased traffic. 

By carefully tuning the scaling parameters, min and max instances, and using traffic splitting, the customer could effortlessly scale from zero instances and benefit from dynamic scaling as per demand. This eliminated the need for maintaining oversized VMs and allowed the customer to fully focus on application development.

Results & Benefits

The impact of the migration to a serverless architecture was transformative for the customer. They saw their peak requests per second grow from 10 to 100, with zero intervention required. 

The adoption of a robust CI/CD pipeline and automated deployments reduced deployment time from hours to just a few minutes, resulting in improved developer productivity and significantly mitigating the risk of human error. 

Furthermore, the customer achieved a 75% reduction in egress costs through optimized data transfer and improved security measures. 

With Cloud Run's serverless capabilities, the client now enjoys faster, more consistent performance, and ample flexibility for scaling up or down, all while achieving an impressive 97% increase in overall efficiency.

Get Started With Promevo

This success story exemplifies how embracing serverless architecture and leveraging the power of Google Cloud's innovative services can revolutionize API deployment strategies.

By migrating their API to Cloud Run and implementing a containerized CI/CD pipeline, the customer achieved remarkable improvements in scalability, reliability, cost-efficiency, and developer satisfaction. The success of this transformation serves as an inspiration for businesses seeking to optimize their cloud infrastructure and streamline their application deployment processes.

If you’re interested in exploring the possibilities of a serverless architecture for your API deployment or require assistance in implementing a seamless CI/CD pipeline, contact us. Our team of experts is passionate about helping businesses harness the full potential of the cloud and enabling them to achieve accelerated growth and success.