Tower | Senior/Principal Software Engineers | London, UK / Berlin, DE | https://jobs.tower.dev
Data teams waste countless hours wrestling with infrastructure instead of solving their businesses’ most critical problem: How to get more value from their data. The problem gets even worse with the AI revolution in flight. Tower changes that.
We're a Python-native serverless platform built on Apache Iceberg with powerful ETL and Lakehouse management tools. We integrate seamlessly with enterprise systems and make companies more nimble by letting them choose the best engine for each job while maintaining full data ownership.
We're focused on hiring developers that love the data ecosystem, are interested in really hard distributed systems problems, and working on dev tools.
Send a message to me, the CTO, at brad (at) tower.dev
Yeah I think there are a number of "hidden" dependencies on different regions, especially us-east-1. It's an artifact of it being AWS' largest region, etc.
us-east-2 does exist; it’s in Ohio. One major issue is a number of services have (had? Not sure if it’s still this way) a control plane in us-east-1, so if it goes down, so does a number of other services, regardless of their location.
Seems the underlying issue is with DynamoDB, according to the status page, which will have a big blast radius in other services. AWS' services form a really complicated graph and there's likely some dependency, potentially hidden, on us-east-1 in there.
us-east-1 was, probably still is, AWS' most massive deployment. Huge percentage of traffic goes through that region. Also, lots of services backhaul to that region, especially S3 and CloudFront. So even if your compute is in a different region (at Tower.dev we use eu-central-1 mostly), outages in us-east-1 can have some halo effect.
This outage seems really to be DynamoDB related, so the blast radius in services affected is going to be big. Seems they're still triaging.
Data teams waste countless hours wrestling with infrastructure instead of solving their businesses’ most critical problem: How to get more value from their data. The problem gets even worse with the AI revolution in flight. Tower changes that.
We're a Python-native serverless platform built on Apache Iceberg with powerful ETL and Lakehouse management tools. We integrate seamlessly with enterprise systems and make companies more nimble by letting them choose the best engine for each job while maintaining full data ownership.
We're focused on hiring developers that love the data ecosystem, are interested in really hard distributed systems problems, and working on dev tools.
Send a message to me, the CTO, at brad (at) tower.dev
reply