Operations | Monitoring | ITSM | DevOps | Cloud

At re:Invent 2019, Farrah and I gave a talk on our paths into the tech industry, how serverless helped us both build some of our first products, and how serverless represents a new mindset. I’d wanted to share a version of that talk on our blog for those who couldn’t make it, though we hope to give it again in the near future.

API Gateway is a serverless service by AWS to expose cloud services through private or public HTTPs endpoints. It is used by many serverless teams to connect frontend applications to backend systems in a secure, scalable and seamless way. API Gateway integrates with Lambda, DynamoDB, S3 and a variety of other AWS services. The main issue with API Gateway, so far, was its cost. At $3.50 per million requests, it can be more expensive than Lambda itself.

We did a compilation of all announcements from the AWS re:Invent 2019 that are relevant for Serverless teams, broken down by services.

Up until now, DynamoDB has been the only option of a truly serverless database battle-tested for production environments. Especially after launching the on-demand throughput capacity optimization, is a perfect-fit database engine for serverless projects.

As serverless application architectures have gained popularity, AWS Lambda has become the best-known service for running code on demand without having to manage the underlying compute instances. From an ops perspective, running code in Lambda is fundamentally different than running a traditional application. Most significantly from an observability standpoint, you cannot inspect system-level metrics from your application servers.

Queues are a very common piece of infrastructure. It offers First In, First Out (FIFO) message delivery to one or more competing consumers. As the queue grows the last item added to the queue typically must wait longer to be processed. Just like someone waiting in line to order food, the first one in line gets to order first. Microsoft Azure Service Bus (ASB) is Microsoft’s premier enterprise-level messaging technology that uses this principle of FIFO.

Have you ever have that dream where you’re in a class on Classical Tibetan Algebra? And you haven’t done any homework all semester? AND it’s the final? That was me in this session: I was WAY in over my head with this one. Tod Golding’s material was so high-level that I got a nose bleed. Seriously. My nose started gushing 10 minutes in. This Portland dewdrop is NOT used to the dry desert and casino AC air.

As our application scales and many services are accessing a multitude of data points for each workload needed, it is difficult to ensure each part of the system has the right set of data access permissions. In today’s world, one of the worse nightmares of a software business is data leaking and data privacy issues. Not only it affects the brand reputation, but could also expose the company to heavy fines and other regulatory sanctions.

Cold Starts have been a massive issue with FaaS. In summary, it makes functions slower to startup in some cases. That’s in the opposite way of every effort to improve web applications performance. Many efforts have been made in the recent years to solve AWS Lambda cold starts or educate on handling them. Many have mitigated the issue, but none really solved it. AWS has just made a great progress on the area with the Provisioned Capacity feature announcement.