Globo Centralizes Service Connectivity and Security With Kong Enterprise

Latin Americas principal media group partners with Kong to create a unified API approach

The Largest Media Group in Latin America

Founded in 1925 and headquartered in Rio de Janeiro, Grupo Globo (Globo) is the largest media group in Latin America and covers a wide range of fields, including film and television, radio, publishing, music, and real estate. Its affiliate, Rede Globo, is the largest commercial TV network in Latin America and produces 90% of its own programming, including news, sports, TV dramas, and documentary programs every year.

Globo consolidates access to these many forms of media, acting as a portal for news, sports, gaming, TV, and more content.

IndustryMedia & Internet
RegionLatin America
Use CaseArchitectural Freedom
Customer Since2019

Reduced service deployment time


Established Digital Hub of Globo


Globo Leads Cloud Native Adoption for a Media Empire

Acting as the digital branch of the Globo media conglomerate, Globos Digital Hub has more than 3,000 microservices in production that provide a streaming platform for the portals. One of those portals, the news portal G1, receives on average 20 million page views per month and has reached peak loads of 29 million page views per month.

Behind the scenes of Globo, the Backstage team is responsible for empowering developers across Globo.

We are a platform of platforms, said Marcos Petry, engineering and product manager at the Digital Hub of Globo. We deliver software for a publishing platform, video platform, big data platform, sales platform, and more. What we support is driven by our end users products.

To accomplish this mission, Petrys team owns delivering API gateways, databases, and other API infrastructure to help end-user developers build their platforms and applications.

As a team, we are passionate about innovation and leading-edge technologies, said Lucca Mendona, software developer at Globo. We were early adopters of containers and even built our own open source Platform as a Service called Tsuru, which we donated to the Cloud Native Computing Foundation.

While the Backstage team was very technically skilled and clearly not afraid to develop entire platforms if that is what the business required, the team also understood that not every tool needed to be built.

For an API gateway, we knew that rather than building our own, it made sense to look at what was already available from vendors, said Petry. We have too many development priorities to spend months rebuilding functionality that is already available commercially.


Globo Upgraded From Kong Open Source to Kong Enterprise

Given the teams passion for open-source technologies, its first experience was with Kongs open-source gateway.

We had 60 instances running in production and over 100,000 requests per minute, said Petry. Kong fit in well with our stack and easily met the needs of a business at our scale.

However, as the system scaled up in support of multiple departments, the team chose to upgrade to Kong Enterprise for enterprise technical support SLAs, a management GUI more appropriate for scaling up adoption and access to many enterprise policies available out of the box.

We also had a corporate decision to standardize on OpenID Connect for authentication, shared Mendona. We knew that Kong Enterprise had an out-of-the-box policy for OIDC and that upgrading would save us months of development time.
For a team that cared about securing hundreds of services, this was an obvious benefit.


Centralizing All API Access Through Kong

The Backstage team now relies on Kong Enterprise as the single gateway in front of 150+ production services, a number that has increased significantly over their time using Kong.

The team also makes use of the Kong Admin API.

The API-enabled nature of Kong lets us manage hundreds of services programmatically, explained Mendona. This is key for the number of integrations our team is required to support.

These capabilities allow a relatively small team to manage Globos entire infrastructure efficiently, with the potential to scale even further in the future.

The team also uses the Kong Dev Portal, both to enable internal developers to publish and consume APIs, as well as to eventually offer APIs to partners and customers.

The Kong Dev Portal gives us a centralized view of all our APIs and services, said Petry. This helps with visibility and organization in managing our API knowledge base across many teams.

Thanks to the architectural flexibility of Kong, we have a single gateway to manage multiple platforms. We run Kong in Kubernetes clusters and are able to natively manage the gateway from Kubernetes.

Marcos Petry
Engineering & Product Manager

Globo Trusts Kong as its Service Connectivity Layer

We are very happy with how Kong has improved internal operations for our development teams, shared Petry. It used to take three to four business days for an end user to add a simple route to a service. With Kong, that has decreased to a one-hour maximum and often is simply instant.

The orders of magnitude gain in efficiency were the result of Petrys team providing a gateway-as-a-service to the rest of the organization.

We created a layer where our development teams can use the Kong platform within the parameters and permissions we have defined, explained Mendona. They are empowered to write and dispatch commands from the command line, with minimal need for intervention or support from our central team.

Thanks to the scalability of this model, Petrys team has already more than doubled the number of services being managed by Kongs API gateway.

With additional business units consolidating across Grupo Globo, the vision is bringing hundreds of more services under the Kong gateway too. The Globo team has emerged as an important partner for the business in defining the technology that will be crucial to achieving Grupo Globos ambitious business goals.

With millions of viewers consuming content daily, at Globo, all eyes are on Kong and the Globo team to lead the companys digital transformation across brands.

For a global media brand like ours, there are hundreds of internal APIs to consider, but we dont want to stop there. We also have our eyes on eventually making APIs available to partners and customers, and creating a real ecosystem around our APIs.

Marcos Petry
Engineering & Product Manager

Recommended Case Studies