134 lines
6.9 KiB
Markdown
134 lines
6.9 KiB
Markdown
---
|
|
title_tag: Menta Network | Case Studies
|
|
title: "Menta Network: Reducing Operating Costs by 60%"
|
|
layout: case-studies
|
|
description: |
|
|
Menta Network used Pulumi and GitLab to build ephemeral environments in Python and
|
|
helped its customers to get ready for peak traffic.
|
|
meta_desc: Learn how Menta Network worked with Pulumi and GitLab to build ephemeral environments in Python and help its customers to get ready for peak traffic.
|
|
|
|
customer_name: Menta Network
|
|
customer_logo: /logos/customers/menta_logo.svg
|
|
customer_url:
|
|
|
|
exec_summary: |
|
|
Menta Network was looking for the right tools to build a
|
|
robust, scalable, and cost-effective load testing service for their clients in the
|
|
retail sector. They needed cloud infrastructure that could be quickly deployed,
|
|
scaled, and decommissioned while evaluating the unique traits of each e-commerce site
|
|
that they worked with. The team also wanted tools that were familiar to engineering
|
|
and fit within their existing tooling and CI/CD workflow. After evaluating multiple
|
|
options, Menta Network chose Pulumi as their Infrastructure as Code solution because it made
|
|
defining and provisioning infrastructure on Amazon Web Services far easier than other
|
|
tools.
|
|
|
|
sections:
|
|
- label: Exec Summary
|
|
anchor: executive-summary
|
|
- label: Customer Challenge
|
|
anchor: customer-challenge
|
|
- label: Solution Overview
|
|
anchor: solution-overview
|
|
- label: Results
|
|
anchor: results
|
|
---
|
|
|
|
## About Menta Network
|
|
|
|
Menta Network is a professional services company located in Mexico City that provides
|
|
quality and performance assurance and monitoring for their clients' applications and
|
|
infrastructure. They regularly need to spin up, scale, and destroy new cloud environments
|
|
as part of their managed load testing solution, which helps clients to evaluate their web
|
|
services and applications.
|
|
|
|
## Customer Challenge
|
|
|
|
In Menta Network's local market, there are multiple 'Black Friday'-style retail events
|
|
including El Buen Fin ("The Good Weekend") in November, and "Hot Sale" in May. Together,
|
|
these events capture a significant share of the $22.6B total e-commerce revenue in
|
|
Mexico and drive extremely high traffic to retailers' websites during sale events.
|
|
|
|
For Menta Network's retail clients, having a responsive and performant digital experience
|
|
is the key to success during these sales. This is especially critical as Mexico has one of
|
|
the fastest-growing e-retail markets --- growing 35% last year.
|
|
|
|
To ensure that these services are scalable and ready to survive the crush of sale-driven
|
|
traffic, clients turn to Menta Network's managed load testing service to identify and
|
|
correct issues before big sales weeks begin.
|
|
|
|
For Menta Network, the challenges came from managing these tests across their client base
|
|
throughout the year and were both technical and financial. They needed infrastructure for
|
|
periodic testing that could scale quickly to mimic peak traffic. Then, automation was
|
|
needed to decommission test resources to minimize the cost of running regular tests at
|
|
scale. Finally, each client required their own unique configuration and load-tests
|
|
tailored to their specific web services.
|
|
|
|
The combination of Pulumi and Amazon Web Services presented the ideal solution to Menta
|
|
Network's challenge.
|
|
|
|
## Solution Overview
|
|
|
|
Menta Network's team had experience with a variety of legacy and modern configuration
|
|
management tools; however, previous tools would not allow Menta Network's engineering team
|
|
to define infrastructure using their language of choice: Python. Managing each client's
|
|
configuration using domain-specific languages (DSLs) proved to be too onerous of a task
|
|
for both initial implementation and ongoing maintenance.
|
|
|
|
> "At Menta Network we have a lot of experience with Ansible and Puppet but wanted to
|
|
shift to more modern tools to manage our infrastructure. We tried Terraform but found that
|
|
Pulumi was far easier to use given its Python support. Ultimately, Pulumi is just as
|
|
robust as Terraform and has the added benefit of reducing the complexity of managing many
|
|
configurations and environments."
|
|
|
|
> --- Ernesto Mendoza, CTO
|
|
|
|
Menta Network has also standardized on [GitLab](https://about.gitlab.com/) for managing their
|
|
entire development lifecycle, container registry and continuous integration / continuous
|
|
deployment (CI/CD) process. Pulumi provided a unique alternative that allowed engineers to
|
|
define both their load testing application and associated infrastructure in the same
|
|
language and tooling --- perfectly fitting infrastructure into their existing [GitLab CI/CD
|
|
pipeline](/docs/using-pulumi/continuous-delivery/gitlab-ci). [Support for
|
|
GitLab login](/blog/welcoming-gitlab-users-to-pulumi) further simplified
|
|
engineering workflows.
|
|
|
|
> "Using Pulumi and GitLab provides the Menta team with a simple and robust platform for
|
|
developing, managing and deploying our managed load testing solution from infrastructure
|
|
to application."
|
|
|
|
> --- Ernesto Mendoza, CTO
|
|
|
|
The Menta Network team chose AWS because they wanted the most mature and robust public
|
|
cloud, trusted by private and public companies in the Mexican e-commerce market and used
|
|
by the majority of Menta Network's customers. On top of that, their strategic partners for
|
|
monitoring and SIEM solutions, [New Relic](https://newrelic.com/) and [Sumo
|
|
Logic](https://www.sumologic.com/) provided integrations that were easy to deploy with
|
|
AWS.
|
|
|
|
Menta Network's managed load testing solution employed Amazon EC2 for creating scalable
|
|
load testing clusters across multiple availability zones, Amazon CloudWatch for telemetry,
|
|
and Amazon Virtual Private Cloud, Internet Gateways and VPN features for secure networking
|
|
for their service. The team also chose to use [AWS Lambda](https://aws.amazon.com/lambda/)
|
|
to automate delivery of mission-critical KPIs to customers' executive dashboards. Pulumi
|
|
made configuring and deploying all of these resources across each customer's environment a
|
|
snap.
|
|
|
|
<img class="block mx-auto md:max-w-4xl my-8"
|
|
src="/images/case-studies/menta-architecture.png" alt="Menta architecture">
|
|
|
|
## Results
|
|
|
|
Adopting Pulumi for one project took an engineer just a couple of days, given that the
|
|
codebase and architecture were already planned for AWS. In the process the team learned
|
|
that it's safe to bet on Infrastructure as Code with Pulumi from a very early stage in the
|
|
engineering process because its abstractions are clear and simple but do not omit
|
|
important architectural aspects of the underlying cloud services such as AWS Lambda, EC2
|
|
and Networking features.
|
|
|
|
Compared to their previous load testing platform, Menta Network lowered their operating
|
|
costs by 60% by adopting Pulumi and AWS. They achieved this while leveraging new
|
|
efficiency afforded by Pulumi to reduce delivery times and increasing features for their
|
|
customers.
|
|
|
|
Overall they were pleasantly surprised by the low barrier to entry and ease of use of
|
|
Pulumi on AWS and plan to adopt Pulumi across all of their AWS projects.
|