The Claudia.JS team has a new release of their serverless NodeJS management tool. The new release also has it filling in a gap if you use AWS SAM for managing your project. Once deficiency of SAM is it doesn't handle tasks like dependency management. Now you can use Claudia.JS for that.

We admittedly don't do a lot of Node.JS but we're sure to checkout the tool in more depth when a project comes up. By the way, the folks behind the tool are also the authors of Serverless Applications with Node.JS.

Dashbird has a piece on the use of serverless in data science and big data. This was very relevant because I have a background in healthcare and life sciences IT, plus we're developing a training scenario for a biomed company. Having spent the past few days looking at the bioinformatics ecosystem, there's much room for improvement in making tooling and processing pipelines easier to manage and support.

If you have a life sciences background, feel free to shoot me some ideas around serverless processing of variant call format, sequence/binary alignment map, and FASTA files. Also, where can I find datasets where VCF and BAM files aren't several gigs in size?

Part 3 in Yan Cui's series on serverless adoption. All the posts have been solid. See the considerations for what to migrate in the beginning of this post. They'll help you prioritize migration.

If you've missed any part of the series, catch up here:

• Part 1
• Part 2

As one series ends, another begins. This time from Stackery!

A bit jealous that Seattle Serverless was able to see a talk based on this blog post PLUS a talk from Pulumi. There's so much serverless in Seattle.

Move over "infra as code", it's time for "business as code". Where this nascent idea goes I don't know. But incorporating budget concerns into our applications is a first step in getting engineers to start caring about business impact.

I'll also add I think there's an immediate practical use for tracking budgets in CloudFormation. If you're in a startup or follow any sort of iterative approach to development and growth, it's worth tracking budget impact as serverless billing scales more tightly with serverless usage. As your feature's usage grows, so does your bill.

If you're interested in the idea of "moving" from DevOps to FinDev then this piece is worth a read.

How are you doing service discovery? Raise your hand if you're using CloudFormation outputs... It's quick and dirty but it's debt to be replaced with a better idea like SSM Parameter Store.

On a side note, would like to know of ways people are doing service discovery for web sites (HTML + JS) hosted on S3. Ended up with a CloudFormation custom resource that wrote a config file which other JavaScript files could import.