Skip to main content

Using actors as a throttle

I've been working on a web shop integration project from time to time for last six months. The use case is that there are several different instances of a webshop (slave) and one master shop. All product information should be integrated from the master shop to the slaves on regular intervals. There are also some business rules applied to the products upon integration.

The integration is set to happen on a certain time of day. It is a relatively long process because it checks through all the products in all shops. That's fine though, there is no requirements on how quick the integration should be. It should rather be a resource constrained process so it would not affect the users using the web shops.

Originally I thought this would be a perfect use case for serverless application running in AWS Lambda for example. I also tried out OpenWhisk from IBM which can be run as a self-hosted serverless platform. While it would have been interesting to try out those technologies I ended up using my DigitalOcean virtual server purely to save costs.

I wrote the app with Scala. I wanted to know the language better and also wanted to see how Akka works in the language it is built in (if all you have is a hammer...). The application is really simple, one just triggers the integration task, it will connect to the webshops to fetch all their product information, cache that to RocksDB, run the transformation rules and the updates the slave webshops.

The application has a configurable amount of workers which do the operations (fetch product data, store it...). The workers are essentially actors which do a lot of blocking operations. This is usually a problem especially if the thread pool for Akka is the default one. The operations using REST APIs of the shops are not exactly fast which can result in quick drainage of Akka threads.

The Akka default pool can be configured to have a fixed number of threads on startup. The amount of workers is derived from the maximum threads available. Each shop interface has a set of workers which is roughly calculated from the amount of threads Akka has divided by the number of shops minus a constant to give some processing time to the aggregate actors. This ends up working pretty well especially when I want to limit the load to the webshops and also keeps the integration application itself from hogging all my 1 core virtual machine resources.

Comments

Popular posts from this blog

I'm not a passionate developer

A family friend of mine is an airlane pilot. A dream job for most, right? As a child, I certainly thought so. Now that I can have grown-up talks with him, I have discovered a more accurate description of his profession. He says that the truth about the job is that it is boring. To me, that is not that surprising. Airplanes are cool and all, but when you are in the middle of the Atlantic sitting next to the colleague you have been talking to past five years, how stimulating can that be? When he says the job is boring, it is not a bad kind of boring. It is a very specific boring. The "boring" you would want as a passenger. Uneventful.  Yet, he loves his job. According to him, an experienced pilot is most pleased when each and every tiny thing in the flight plan - goes according to plan. Passengers in the cabin of an expert pilot sit in the comfort of not even noticing who is flying. As someone employed in a field where being boring is not exactly in high demand, this sounds pro...

PydanticAI + evals + LiteLLM pipeline

I gave a tech talk at a Python meetup titled "Overengineering an LLM pipeline". It's based on my experiences of building production-grade stuff with LLMs I'm not sure how overengineered it actually turned out. Experimental would be a better term as it is using PydanticAI graphs library, which is in its very early stages as of writing this, although arguably already better than some of the pipeline libraries. Anyway, here is a link to it. It is a CLI poker app where you play one hand against an LLM. The LLM (theoretically) gets better with a self-correcting mechanism based on the evaluation score from another LLM. It uses the annotated past games as an additional context to potentially improve its decision-making. https://github.com/juho-y/archipylago-poker

Careful with externalTrafficPolicy

On a project I am working on is hosted in an EKS cluster with the NGINX ingress controller (the one maintained by Kubernetes). It is deployed using it's official official Helm chart, which I realized, after a lengthy debugging session, was a mistake. The initial setup I aimed to improve had several flaws. Firstly, we were using the AWS Classic Load Balancer in front of the nginx ingress in the cluster, which has been deprecated for some time (years?). Continuing to use it makes little sense to us. The second issue was that we were only running one(!) nginx pod, which is quite sketchy since the exposed web services had essentially no high availability.  I switched to the Network Load Balancer (NLB), which was straightforward - I just needed to change the ingress-nginx service annotation to specify the load balancer type as NLB: service.beta.kubernetes.io/aws-load-balancer-type: nlb However, increasing the replica count turned out to be tricky. When I bumped it up to two, I began to ...