How Bizzabo Avoided the AWS Outage To Ensure Seamless Event Performance
The past two years have been filled with virtual transformation and experimentation. Like everyone, we’ve had our share of ups and downs. We’ve learned that stability and dependability are key to ensuring event success.
Can you imagine what would happen if you were hosting a major event and — minutes into the keynote — the venue’s power crashed or the pipes burst and flooded the event space? These are chaotic, panic-inducing scenarios for event professionals and attendees.
As the pandemic persists and the events industry evolves, we’ve realized our tech has become the venue for countless brands and organizations. Our tech has to operate flawlessly. Without it, there is no event.
Learning Our Lesson: How We Prioritize Our Customers
In August 2020, we learned these lessons the hard way. We experienced two instances in which Bizzabo’s systems went down. As a customer-centric business, we recognized we failed our customers and knew we had to take action.
In many ways, these failures drove us to build the Event Experience Operating System. Like so many companies, we met the challenges of the COVID-19 shutdown in March 2020 by repurposing our technology, integrating new partnerships, and building new solutions. We also accrued a healthy amount of tech debt in the process.
Constant uptime and seamless event performance are non-negotiables. We are committed to our customers and to ensuring they can count on us — regardless of event format, size, or location. At a time when new competitors cropped up with flashy features, we made the important decision to invest meaningful resources in building the strongest foundation possible for our stable, scalable enterprise technology solution for the events industry.
Avoiding Being a Headline in the AWS Outage
Earlier this week, while countless websites and services — including a few event management software companies — were down because of the Amazon Web Services (AWS) outage, Bizzabo customer events ran flawlessly around the world.
Was there a bit of luck? Yes. Was it also plenty of foresight and planning? Absolutely.
Over the past year, we made a number of technology and people investments to help our customers avoid these types of outages. Here are just a few:
- We hired more than 60 engineers and are hiring dozens more engineers (use the R&D filter to see open roles) across our global locations.
- We have created a culture of experimentation to build confidence in our platform’s capability to withstand turbulent conditions in production. Bizzabo’s R&D team is encouraged to take advantage of the flexibility of the cloud to create well-architected systems that perform at scale — even in a “non-steady” state.
- We use a high-entropy approach to EC2 instance (server) allocation by creating a highly fragmented ecosystem of multiple instance types, featuring different families and business models.
- We increased resiliency by deploying our platform on top of multiple availability zones and in multiple regions within AWS. Additionally, we maximized resiliency through proper application design and implementation, as well as through thorough stress endurance testing.
- All data adheres to the highest levels of security standards and industry best practices, including SOC 2, ISO27001, General Data Protection Regulation (GDPR), and California Consumer Privacy Act (CCPA).
Lastly, in addition to the infrastructure AWS provides, Bizzabo leverages kubernetes (also known as K8s) to manage, deploy, scale, and orchestrate our application. According to ContainerJournal.com, “K8s is like the ultimate game of Tetris — it chooses the placement of containers to optimize computing resources, ‘orchestrating’ where these workloads end up.”
This approach creates greater reliability by leveraging spread constraints for pods (computing resources), as well as taints/tolerations and affinity/anti-affinity to create a clear and highly available architecture for our production clusters. This powerful combination of infrastructure and container orchestration is what, we believe, enables advanced, mature platforms to endure sub-optimal states like the one observed earlier this week throughout the world.
Securing the Future of the Events Industry
The events industry and the technology that supports it are continuously evolving. But one thing is stable: our strong commitment to event experience and solving issues before they happen.
We are humble and always share full root cause analyses with our customers when we fail. This transparency is what helps us build trust over time. You can always get real-time information on Bizzabo’s system status and security at https://status.bizzabo.com.
Technology dependencies will always exist. Tech failures will always exist. It’s what you do when outages and failures happen that matters most.
If you want to be a part of a customer-centric organization and work to change the events industry by delivering meaningful event experiences, then we want you to join Bizzabo. We’re hiring!