The need for speed is a continuous quest in software development. Nowhere more so than in enterprise-ready platforms, where large amounts of users and data present unique performance challenges. Today, we’ll look at ways to maximize performance and responsiveness for platforms running at any scale.
November 2022 saw CHILI publish launch its new CHILI GraFx Creative Automation platform. It was an opportunity to bring exciting new technology like animated media to CHILI publish’s software offering. But for CHILI publish’s Chief Architect, Pieter Van Parys and his team, it was also a chance to kick off some behind-the-scenes changes in how CHILI publish’s technology performs. As Pieter notes:
No matter what new features we’re developing, there is also always a goal to guarantee and increase performance. This is especially critical for enterprise-ready software because of the large numbers of users and big datasets.
Pieter Van Parys
Chief Architect
@ CHILI publish
CHILI GraFx is a SaaS platform designed to help brands and agencies boost their marketing creative output through automation. Users log in from across the globe to create visual assets that can be output at scale as different variants for different channels.
CHILI GraFx shares a lot in common with how other SaaS platforms work, so how do you define performance for such a cloud-based creative automation platform? Pieter Van Parys thinks there are two keys areas to focus on:
The performance goal then is to have a highly responsive, browser-based tool able to trigger output that can scale to meet any enterprise needs.
Sounds simple? Well, if you’re thinking great coding coupled with good IT infrastructure must be essential in achieving this, you’re right. But, at a deeper level, there is an often overlooked aspect of software development that can also bring huge performance gains: communication.
CHILI publish is a remote-first company that develops its solution using an Agile methodology. But, more fundamentally, we are a company made up of people. With the pandemic thankfully behind us, we see the value in getting together not just to share, but also spark and nurture creative ideas.
In a recent case-in-point, the CHILI DevOps came together at a “CHILI Kick Off” (read: “internal event”) to focus on performance, discuss problems and possible improvements, explore measurability and reporting, etc. – basically everything that would get the CHILI GraFx platform running even faster and able to scale output without diminishing performance. Pieter notes:
Getting people together with a precise focus like this creates an energy and enthusiasm that can have really impressive repercussions. At our recent performance-focused Kick-Off gathering, we got a deeper understanding of some data indexing issues we were encountering. As a result, we implemented new queries used by the API… and boom! In some instances, the platform can now deliver a x100 speed boost, and—something of huge benefit to enterprise customers—there is no longer any performance difference between a 10-document environment and a 1 million-document environment.
Pieter Van Parys
Chief Architect
@ CHILI publish
So maybe one of the best ways to get a performance boost for your platform is to get your team together to talk in person round a table. Or several tables… like in the office, the restaurant, and for those really enthusiastic team members, the ones with beer glasses on them.
Cloud-based SaaS platforms also bring some big performance and operational benefits over traditional on-prem solutions. Cloud-native services are well documented, can scale rapidly, and provide far better control of the environment and software running in the environment than can be achieved with on-prem deployments.
As a result, Users/Customer and Developers experience multi advantages to cloud solutions:
With SaaS, you see rapid, limitless scalability. Imagine, for example, the holiday season is approaching and your team needs to create and output x100 more digital and print marketing content than during a normal period. With an on-prem solution, your creative automation platform server is likely to grind to a halt under the hugely increased workload. As a result, so will your marketing campaign! With a cloud-based infrastructure, however, rules can be in place to spin up automatically a second (or as many as are needed) server as soon as the platform detects it is getting close to running at capacity.
All this requires zero human intervention, and ensures your creatives aren’t left twiddling their thumbs waiting for progress bars to cross their screens. The result: server resources always match output requirements. What’s more, when the workload diminishes, the server instance is no longer needed and is automatically terminated.
Creating this type of architecture on-premises is obviously incredibly hard, if not impossible – especially for an external software provider. It’s also wildly expensive, and wasteful, as hardware resources are often left idling.
In the case of CHILI GraFx, moving over a million graphical assets to the cloud for enterprise customers was a challenge worth embracing. The reengineering opportunities it provided have lead to tangible performance and savings boosts from every aspect of our platform. We’ve even given one of these benefits a name: hyper scaling.
Teams building software also see performance advantages to working with cloud platforms, and this has certainly been the case at CHILI publish. We’ve been able to increase transparency and reduce time spent compensating for the individual quirks found in on-prem installations. Cloud infrastructure has also provided realtime tools that make it far easier to test and monitor the success of platform modifications in sandbox, beta, and live environments.
This ability to monitor and measure API and infrastructure efficiency in detail – even in realtime – is of great benefit to developers and integrators. We get a bigger, clearer picture of how our platform is running in a single location, and this, in turn, makes it far easier to identify and resolve bugs, as well as performance bottlenecks. Cloud tools also provide information on platform usage, and this gives actionable insights on how to improve user experience for customers working with the platform.
Ready to learn more about enhancing enterprise platform performance and experience Bram Verniest’s high-performing hosting skills? Then watch CHILIpulse Episode 4 to see Bram’s full conversation with Pieter Van Parys, Chief Architect at CHILI publish, where they cover in detail the topics discussed above and more.
And make sure you stay in the loop of what’s going on at CHILIpulse, by joining our newsletter.