What’s Behind the Shiny Surface of Contentful? The Engine of the CMS

Yep, 2020 was tough for business, but those low and desperate moments sparked some serious breakthroughs both as an industry as well as at the company level. (Perhaps only once we emerge out of this together will we fully appreciate it; only with stillness can you reflect on the turbulent.)

One company that experienced breakthroughs is one of the leading content management system solutions out there, Contentful.

They’ve been helping organisations like GoodRx (a healthcare company), Loblaw (Canada’s largest food retailer), and the Covid Tracking Project – all of which have been experiencing massive upswings in traffic, while its tech partner ecosystem, in which ArganoUV is proudly part, has grown apps to strengthen a brand’s capabilities.

These breakthroughs were not spontaneous; they grew out of the solid foundations that Contentful built throughout the previous years, as well as what they’ve constructed within the last year.

And we can take a look at some of these infrastructure projects, as we lift up its hood and check out the CMS’ engine.

Using its cloud infrastructure, and carrying out its API-first strategy, Contentful is concentrating on nurturing a handful of key areas:

• As a PaaS, to strengthen its in-house engineering of cloud infrastructure.
  
• In terms of fleet management, to automate dull and repetitive processes, while making sure that the platform scales in line with its customers.

Underpinned by research to confront current issues, their teams construct prototypes before fulling launching solutions.

Let’s look at a handful of ways in which Contentful works behind the scenes to build CMS solutions of today.

Tracing queries

“We like to think of our customers in two categories: the 99% and the 1%,” confirmed Contentful. Their customer majority will rarely notice dips and boosts in their performance. While the very few, the 1%, are looking to bring out a new product or service in a blaze of glory, with firefighters struggling to manage the outbreak.

For example, one brand decides to bring together a host of services and weld them into one unified tech stack, but it slipped their mind to query their GraphQL API with a high-res timestamp, which results in a widespread freeze of their delivery cache.

To counteract this, Contentful drives down into the root causes of slow queries, takes samples of actual queries, and then traces them throughout all internal services.

Customer queries can be replayed and reanalyzed if they’re considered dubious, using open-source powertools like Jaeger.

Higher performance

“While most of the technical users think of Contentful as a collection of APIs for distributing content,” stated the CMS platform, “behind the scenes our platform is a composite of dozens and dozens of individual microservices. 

“Some of them authenticate users, others count consumed resources and others transform assets before delivering them to the visitor of a website powered by Contentful. To ensure the overall health of our platform, we spend a lot of time fine-tuning the performance of these services.”

It was clear that these microservices needed to be automated, as well as being finely separated to fix scaling mechanisms into them. In this way they can then track in real time their performance and debug as they go along.

Zero Rest 

“We run a lot of database server instances, also known as shards, to store customer data,” added Contentful. “Individual shards vary in purpose, size and utilization rate, requiring our infrastructure team to do a fair amount of data migrations.

Take for example, the necessity to shift customer spaces to a unique shard, for times when they sign a contract with a big player.

“Unfortunately,” added Contentful, “the process of copying data can take hours and calls for a temporary freeze on content changes. This is why SaaS businesses have to introduce regular maintenance windows.”

Contentful was not very… well, content with this situation. Like stated earlier, it experimented and built prototypes to reach alternative solutions. This is what they did to find a way to execute data migrations yet without locking the original database.

They managed to figure out how to do this, so that in a scenario where customers are trying to update their content during this period of updating, their changes are not going to be lost.

It’s time to close the hood of the engine for now, and let Contentful roar. Until it’s time for our next inspection. 

PS: ArganoUV is one of the world’s leading Contentful specialists. Contact us to see how we can work together.