This is the best time of the year, giving me the opportunity to brag about what we have accomplished. While looking back to a year in which *a lot* happened with Thundra, I believe we truly deserve to take the pride in. 2018 was the year that Thundra was transformed into a company from an intrapreneurship project in Opsgenie. As a young company fresh out of our safe zone, 2019 was the year for us to blossom into Application Observability and Security Platform running on containers, VMs and of course serverless. As a company, we have attended many serverless and DevOps events and organized the ServerlessDays in Istanbul. We’ve achieved important milestones with AWS partnership and many more.
Let’s take a closer look at all the milestones in Thundra’s life in 2019!
Time is relative. It really is. When I think about what we had at the beginning of 2019, it feels like it is 2015 with no exaggeration. We’ve come up with so many cool features leveraging the input coming from our customers and serverless community that it feels like a 5 years before. In 2019, we came up with lots of important features such as distributed tracing, topological overview, intelligent and actionable alerts, seamless monitoring with no code change, serverless security and finally container and VM support. We’ll browse all the updates below but before that, I should heartfully thank each and every teammate of mine for their tireless efforts to make this possible. Thundra wouldn’t be what it is today without this unique team.
Before creating this blog post, I requested some images from my friends about how Thundra is looking like in January 2019. Here it’s:
We were listing all of your functions and their invocations, and we were showing some charts about important metrics for all the functions connected to the Thundra account. This was the application that had been being used in Opsgenie for more than a year as a tool working inside for the company. Starting re:Invent 2018, we started listening to serverless community well and put their inputs and feedbacks in action with new features in our product. By January 2020, we are proud that Thundra becomes an enterprise platform that additionally provides debugging, monitoring, troubleshooting, and securing any apps on the cloud(serverless, containers, and VMs). The only thing that remains is our lively color!
Let’s look at the milestones that we achieved throughout the year.
- January: We’ve been asked by many developers from many companies at every scale about .NET support especially after they saw what we achieved for serverless Java apps. In the first month of 2019, we announced our .NET support and developed it throughout the year and now our .NET libraries can do whatever Java counterpart can do from distributed tracing to security and compliance control.
- February: We’ve introduced our new UI and said bye to our most ancient design coded for an internal app. It was a tragic moment for us but our new UI brought a fresh and more effective user experience and increased the usability of our application positively according to a survey with our customers. We’ve also announced our chaos engineering capabilities in AWS Lambda functions in this month. This made us clearly visible in the eyes of serverless communities.
- March: We introduced our topological view to serverless architecture in March. Until then, we were motivated by some customers and serverless community with these words: “You can’t understand serverless architectures by simply listing functions. You need to understand the relationship between functions and other serverless resources”. They were right and we came up with our architecture view as our response. We made several updates on it throughout the year like highlighting the resources with the biggest throughput and poorest health, interactions on nodes and edges, and some more. Below is how it looks like today.
- April: One of the most exciting updates of the year arrived: Distributed tracing. When Thundra was first invented in Opsgenie, the need was not distributed tracing but understanding what happens *in* a Lambda function. For this reason, we had local tracing (which is still unique to us) but not distributed when we first introduced Thundra. After talking with many customers and analyzing the competition, we decided to implement distributed tracing. Combined with local tracing, we have the best tracing experience that lets developers manage the aggregate set of distributed services an application consumes down to the line level of the runtime code for every local service or container-based application. Our unmatched capabilities with tracing are the biggest differentiator for almost all of our customers now.
- May: Now, we’ll go over some features which were supposed to be on our product from day 1. You can blame me about prioritization but we thought monitoring capabilities are more important than hygiene factor features. That’s why it took us that much to introduce multi-user support for a Thundra account. Late is better than never, right?
- July: After working on paying our technical debt in June, we came up with another feature that made us ashamed about shipping it that late. We brought alerting later than expected because we wanted it to be the most useful one in the market. Thanks to Thundra’s unique query support, you can set up alerts on almost everything. For example; you can set up an alert like “if the operation time exceeds 200 ms in a minute 5 times for my SQS queue with name users, alert me”. There’s no other tool in the serverless observability market that you can achieve the same.
- August: We came up with our performance reports this month. Our users now receive a report about their most costly and most unhealthy functions every day and every week and understand if there is something that requires attention.
- September: In order to take advantage of our observability solution, the only way was to add our instrumentation libraries to your application. With the update that we rolled out September, Thundra became a plug-and-play solution that can be connected to your AWS account in less than 5 minutes.
- October: We took the update in September one step further in October. After enabling our customers to plug Thundra to their AWS account effortlessly, we reduced the hassle related to the instrumentation to a seamless one-click-process from Thundra console. With this update, we have made our onboarding a lot easier than it was before. A remarkable milestone in our history!
- November: Here we come to the month of the biggest updates. In November, we have announced the most important updates of the year for our product. First, we shipped our AI-driven dashboards with active anomaly detection and insight generation. In this way, our customers got the chance of catching any problem that may occur with their modern cloud architecture. Then, we came up with our security addition on top of our observability features. As an observability solution, we were already acting as a man-in-the-middle in order to measure the performance. We thought that it’s time to use our position to provide more service to our customers. With this new update, Thundra enabled our customers to configure the list of whitelisted and blacklisted resources to control the behavior of the applications. This unlocked the opportunity of helping our customers to maintain the security compliance standards with their application. In November, we also pre-announced two important additions in our skillset. The first was remote debugging. With this update, we made the debugging of AWS Lambda functions while they run on AWS on the local IDE. We are now working on it to deliver it as a fully-featured in the first quarter of 2020. In this way, you’ll be able to debug AWS Lambda functions with a specific request while other instances are working independently. The other feature also changed the way we speak about Thundra because it was not only a serverless tool anymore. With this new update, we have abstracted our observability and security features away from the platform. In this way, we transformed our platform into “Application Observability and Security Platform” rather than being a serverless observability solution.
2019 brought many achievements and happy customers for us. We had the chance of speaking thousands of serverless practitioners and onboarding hundreds of them to our platform. It’s good to see that our product’s achievement was recognized by independent researchers. In March, we have achieved the top 3 monitoring vendors prize by EMA which demonstrates our capability of helping serverless developers even before we rolled out many important features. In October, we have achieved the DevOps Competency level by AWS which demonstrates our ability to define and implement AWS customers’ business transformation journey by leveraging the best in class DevOps technology and services required based on their business and technical needs. Then, we became eligible to sell Thundra with enterprise contracts on AWS Marketplace. Thanks to this update, we managed to ease the procurement process of our customers by simply adding Thundra cost on top of their AWS bill.
Top Reads by Thundra
In 2019, we created tons of fresh content such as serverless tutorials, use cases, best practices, and feature updates. Here are the top 5 posts from our blog:
- AWS Step Functions - Doing Serverless is Easier Than You Think
- The Comforts of GraphQL with AWS AppSync
- Debug Your Python Functions Locally
- AWS Lambda: Real-World Use Cases for the DevOps Engineer
- 3 Tips to Control the Cost of AWS Lambda
We also wrote whitepapers which created thousands of downloads from the serverless community. To recall, those were:
- Introduction to Serverless Monitoring
- Observability in Serverless Ecosystem
- Building Serverless IoT Systems: From Node-Red to AWS Lambda
- How to test your AWS Lambda functions
In 2020, we’ll create even more and higher quality content to deliver the freshest and most applicable knowledge to the serverless folks.
In 2019, Thundra supported many companies improving their process for troubleshooting and debugging their serverless architecture. Arcelik, a heavy user of AWS Lambda on IoT devices achieved a 50% saving on their costs with the insights provided by Thundra. Daysmart, Winnow, Fotograf.de, Wecheer, and Soulpicks are the other businesses leveraging Thundra to improve the health of their application by completely understanding the behavior of the application with full tracing, anomaly detection, actionable alerts. This quote by Mihai Ciureanu is enough to demonstrate the importance of Thundra to our customers:
“I can’t imagine what we could do to filter the issues without Thundra as a solution“
In 2020, we will have a lot more stories to share with further additions to our feature set.
Leaving behind a year full of progress and success, we are completely confident about what 2020 will bring. We’ll sail our ship to bigger successes with much more happy customers. Stay tuned to our blog for more updates and top serverless reads, and let’s catch up over Twitter or any event all around the globe. We’ll be in the majority of the AWS Summits, ServerlessDays, and DevOpsDays in 2020.