MMCafé, a leader in the internal communication sector since its foundation in 1993, offers an innovative SaaS platform that provides a complete solution for companies of all sizes, recognized as one of the best Intranets in Brazil. Widely adopted by large and small organizations, including renowned banks, MMCafé has solidified its relevance in the market.
With over 500 projects implemented in 23 countries, MMCafé is recognized as a leader in the Intranet sector, validated by awards such as IPP, which named it the Best Intranet in Brazil for two consecutive years.
Its global presence and acceptance by renowned companies highlight its position as one of the main Intranet solutions. Recognition from prestigious awards demonstrates the trust and satisfaction of customers worldwide, further solidifying its relevance and authority in the sector.
Build a Kubernetes cluster on Google Cloud infrastructure using high availability resources, following security best practices.
Ensure that the Kubernetes cluster has cost-saving features without compromising the environment’s performance.
Collaborate with the MMCafé development team to implement robust DevOps practices, including continuous integration and continuous delivery (CI/CD).
Implement a comprehensive monitoring solution to ensure infrastructure availability and performance 24x7.
Our work involved creating a Kubernetes cluster scaled to serve a large customer base, many of whom operate 24x7. We implemented deployment rules and security levels to ensure the stability and integrity of the environment.
We improved the performance of the CI/CD flow by reducing build times, standardizing pipelines, implementing automated deployment in environments, and improving integration with branches (trunk-based deployment).
We established a comprehensive monitoring process that operates 24x7, including infrastructure, services, and applications, with custom dashboards and alerts, allowing for quick incident identification and response, as well as understanding environment bottlenecks.
Continuous monitoring provided valuable data to the MMCafé technical team, enabling adjustments to codes that resulted in significant performance improvements and infrastructure consumption reduction. We adopted dashboards and alerts to optimize and enhance pipelines, reducing the build time for each version.
In terms of technologies, integrating Grafana and Zabbix, we continuously monitor the performance of our nodes and virtual machine clusters. We conduct daily performance reviews to obtain real-time insight into the health and efficiency of the infrastructure. Additionally, we develop comprehensive bi-monthly reports and participate in bi-weekly meetings providing detailed insights into trends, anomalies, and areas for improvement. These meetings empower us to take proactive measures to continuously optimize the infrastructure and ensure its optimal operation, aligned with the company’s strategic objectives.
The ability to automatically expand computing resources as needed not only ensures operational stability but also optimizes costs by avoiding resource underutilization or overprovisioning. In MMCafé’s case, Autoscaling was an essential pillar, allowing our Kubernetes cluster to adjust in real-time to users’ varying needs, simultaneously ensuring operational efficiency and agile responsiveness, crucial factors for the success of our platform and continuous customer satisfaction.
We implemented development (dev), quality (QA), and production (prod) environments for MMCafé, ensuring stability, security, and efficiency. We created a Kubernetes cluster with strict deployment rules and security levels to maintain environment integrity. This approach allowed the technical team to work in a secure and stable development environment, while QA and prod environments were configured to test and deploy new features effectively and securely. This implementation was essential to ensure reliable and high-quality services to MMCafé’s customers.
We implemented a 24x7 monitoring system using open-source tools, covering MMCafé’s infrastructure, services, and business flow. This system includes custom dashboards that identify bottlenecks and problems in all system layers in real-time.
The monitoring spans from Auto-Scaling to application queues, with specific alerts for each process. These alerts are sent to the communication group connecting MMCafé’s DevOps team with our DevOps Gole, facilitating joint problem-solving and failure notifications to the development team.
Custom alerts allow quick identification of the problematic module, facilitating incident response and understanding bottlenecks. We use dashboards and alerts to optimize pipelines, reducing build times for each version.
Our monitoring system measures application SLA and generates detailed incident reports, ensuring efficient management and continuous service improvement.
Since the environment’s implementation, we have adopted an agile methodology based on Scrum, with bi-weekly sprints to identify and execute all improvement activities. Our focus is on automating continuous environment updates, prioritizing security and stability. Whenever necessary, new services are implemented to ensure operational efficiency.
After each sprint meeting, we create minutes that include the date, participant names, and the status of each task (e.g., Open, In Progress, or Done).
The monitoring tools provided to MMCafé are continuously updated with both programs and plugins as well as custom dashboards. Continuous monitoring provided valuable data for code adjustments, resulting in performance improvements and reduced infrastructure consumption.
On weekdays, during peak access times, we manually review the dashboards to ensure the environment is in excellent condition. This manual monitoring is done at least three times a day.
Whenever we handle an incident, we create a detailed report. This includes a description of the symptom, start and end times, actions taken, temporary solution, and cause. When the cause is unknown, we suggest opening a problem ticket and later seek a definitive solution.
Every two months, we deliver a bi-monthly report containing the SLA of environments and sites, the status of manual reviews, and all actions taken over the two months.
The modern infrastructure of Google Cloud and the Kubernetes architecture implemented by Gole brought agility, efficiency, and cost reduction. Due to the implementation approach and the solution’s elasticity, MMCafé can continue acquiring new clients without worrying about infrastructure. Google Cloud’s security ensured data integrity and strengthened the company’s competitive position.
“We are very satisfied with how your services have positively impacted ours. The team’s availability to handle urgent demands has been crucial to our operation. On several occasions, when we faced unexpected challenges, the team was readily available to help us solve problems efficiently, which directly impacted our ability to serve our clients.” Adriano Silva - DevOps Analyst
“Gole created a monitoring process for the entire environment so that the MMCafé team could make several adjustments to improve performance and reduce infrastructure consumption. Together, both teams made optimizations and improvements to the pipelines, reducing the build time of each version. The platform is stable and healthy. The search for improvements is constant. Without the implementation done by Gole, the current infrastructure would cost at least 2.5 times more.” Eduardo Pires - Commercial Director