This week our serverless superhero is Pubudu Jayawardana, a senior backend developer at Starred and AWS Community Builder. Pubudu regularly posts articles answering the hard serverless questions like what happens when you update a state machine when it’s running. Be sure to catch his upcoming talk at Serverless Saturdays this weekend on how to build a leave management system in AWS. Thank you, Pubudu, for all your hard work and dedication to the community.
You’ve heard it from me time and time again, security is the most important component of any application. Chirag Rathod puts an emphasis on this by showing us how to whitelist IP addresses in an HTTP API. This is particularly useful if you need to do whitelisting on multi-tenant applications where a WAF doesn’t give you enough fine-grained control.
One of the beautiful things about GraphQL is its ability to fetch exactly what you want (it’s kinda the whole point). But what about when your fetch relies on a series of operations to get the data requested? Rosius Ndimofor gives us an in-depth look at how to build pipeline resolvers with CDKv2 in his post. He walks through every step of the process, how they link together, and what they look like in the CDK.
If you follow my blog you know I post about the complexities of designing serverless applications. This week, I (Allen Helton) wrote about designing around serverless service limits. This talks about digging into the services in your design and taking their service quotas into consideration when developing a plan. Not doing this has bitten me on more than one occasion, resulting in unnecessary rework. This post is meant to share my lessons learned so you can build high quality software the first time.
Distributing computing is hard. But distributed databases are even harder. Alex Debrie helps clarify some of the complexity by explaining the various forms of database consistency. Funnily enough, the word consistent is not consistent when talking about different database models in distributed systems.
When you come from a relational background, NoSQL can feel a number of things: silly, complex, unstructured, intimidating. Max Rohde helps us build some confidence with it with his beginner’s guide to DynamoDB with Node.js. He covers everything you’d need to know to get started. He even provides you with a working template to get you going faster.
Viljami Kuosmanen shares his story about his first year as head of engineering at epilot in this week’s spotlight. He walks through the steps he took moving his company to being serverless and API-first in an incredibly interesting post. He walks through the various components, his struggles, and where they are now after a year. Very inspiring and helpful for those of us early on in our cloud migration!
Practice makes perfect. Keep up your exposure to various serverless projects and reference architectures to build your confidence and understanding in this space. Serverless is much more involved than our first hello world projects.
Serverless jumps up in complexity quickly and keeping track of the components can get overwhelming if you aren’t careful. Side projects, blog posts, and sharing with others all help keep up your familiarity and comfort when building production-level apps.