Microsoft Tech Talks EMEA | Azure Cost Optimization Practices & Tools
February 26, 2021
The Good, the Bad and the Ugly in Cybersecurity – Week 9
February 26, 2021
Making it faster and easier to connect
Although many livestream options have popped up in response to COVID-19, I founded Switchboard Live in 2016 with a goal of making it faster and easier for content creators and brands to reach a bigger audience. Before creating my company, I worked for 15 years in the streaming industry. I always noticed how difficult it was to manually stream events across different platforms and workflows with little to no interruptions. The valuable experience I gained from manually working through the technical aspects and processes of streaming early in my career, illuminated the highly-scalable potential of a technology like Switchboard Live.
While initially, Switchboard Live focused on providing houses of worship a way to connect remote congregations together, we quickly realized there was a massive need for this technology across industries. We quickly expanded into industries such as sports and media, education, government, nonprofits, media and entertainment, technology, and digital creators.
Switchboard Live – Moving from Azure to Google Cloud
Switchboard started by using Google Cloud for small workloads, with Azure as our primary provider. In particular, we used Google App Engine to spin up additional resources when website traffic increased, to ensure our customers have a seamless experience with our platform. However, as we grew, we wanted a fully managed Kubernetes platform that was easy to manage and had an effortless developer experience.
It’s not surprising that Google Kubernetes Engine (GKE) was the best fit given these requirements. Particularly given it is the most mature container orchestration service available today, providing industry-first capabilities such as release channels, multi-cluster support, unique 4-way auto scaling, along with node auto repair to help us improve availability.
Likewise, Cloud Build offered a unique developer experience when it came to CI/CD. Since it was fully serverless, there was no need to pre-provision servers or pay in advance for additional capacity. Most importantly, we only have to pay for what we use. And finally given deep integration of Cloud Ops with GKE, actionable insights are just a click away. When we detect an anomaly in a dashboard or when an alert is fired, there is no need to log into another portal or console to start troubleshooting and remediating. Everything is available in-context with deep links to logging and monitoring
With the help of the Google for Startups Black Founders Fund we will soon be expanding to more Cloud tools such as Cloud DNS and Cloud SQL. These will help reduce our latency to almost zero.
