I created a sophisticated refi calculator to understand what would overall savings looks like with different loan options, additional principal payments considering the amortization and progression in the current loan.
I don't agree with "Kubernetes is an orchestration system for containers originally designed by Google" statement. White it is not false, it creates a wrong impression of what K8s is.
Kubernetes is a cloud operating system which lets you run modern apps in any environment. One important component of course container orchestration, but it went far beyond just orchestrator. Kubernetes has a very powerful ecosystem, and it managed to unite almost all infrastructure vendors and projects. Its amazing to see how so many competing companies could agree on something. And that become K8s.
Nomad is great when you're working with VMs, but I don't see it is much relevant in the modern era of K8s and cloud-native.
Not directly related to S3 traffic bill, but overall cloud cost management.
Maybe some are unintentional, but still very painful. My experience with AWS & GCP.
- AWS CloudWatch: expensive service, virtually unusable, hard to turn it off.
- AWS overall: finding and cleaning up resources is messy. The order of creating & cleanup is not same. Closing an account is a painful process. GCP Project structure is way easier.
- AWS EKS: You create a cluster, then a node group. Deleting a cluster fails if there is a node group. You go ahead to delete a node group, it complains because of "dependencies". While you're randomly looking for a "dependency" the $ clock is still ticking. You should delete the network interface before you could delete the node group, and only then the cluster. This does not sense because if the network interface was created implicitly by the node group, i should not be responsible for deleting the network interface. There should be a symmetry in create/delete operations.
- GCP GKE: You create a cluster, then delete it. Cluster gets deleted - kudos, usability much better then with AWS EKS. But it turns out lots of LoadBalancers and Firewall rules are left over and still appear on the cloud bill. Those are implicitly created and should be cleaned up implicitly by GKE.
This is quite useful, but frankly I doubt many would be willing to put the ASCII style laptop on their web site. Literally a week ago we had to design a bootstrap carousel inside macbook pro laptop here: https://berlioz.cloud
If you could add more style options of devices like macbook, imac, few PCs, tables and mobiles it would be so great!
You know that TLDs are not owned by any single government. Even if some countries introduce domain reselling laws, its just enough to have one country to not to have the same law.
It is sometimes frustrating, but we can live with that. As I said before, you just have to improvise a little bit more with naming.
How much government control would you need to make you happy? Where is the line that determines who has right to the domain? I think if you add any level of enforcement it would lead to too much chaos.
In either case it would not help startups. Lets say you decide to open a company named "Banana". How do you envision taking ownership of banana.com? :)
New companies have a luxury of picking ANY name, and a name for which domain is not taken. There are lots of practices on name picking.
Systems are getting too complex for traditional methods of validation and certification to operate. Safety and correctness of intelligent machines should be validated by even more intelligent machines.
Another big problem is that there is very little competition. In reality there are only two players. Smaller manufacturers get absorbed by big ones. As a result grounding a single model causes economic catastrophe in many airlines - leading to significant ticket price increase!
SQS-FIFO should let you process task once & only once. Just make sure you configure timing parameters correctly.
DynamoDB has triggers that get fired upon changes. That would strongly help with eventual consistency implementation (which i strongly recommend). But with this you should write Lambdas. Check how well is Lambda Java supported.
Are you sure you want to use EC2 directly? Why not to use ECS? This would let you focus more on the business and less on infrastructure
He doesn't seem to need a forever queue. Processing might take a while, but he just needs to get it processed once (and exactly once) and move on. Yes, if he wants to do event sourcing, then yes SQS would not suffice. But I don't think he needs that. But SQS seems to be a great choice.
You should provide more info regarding your environment. If you're running this in public cloud tell us which one. Every provider has native several queue services for different needs and makes things easier to work with and worry less about setting things up.
I could combine multiple creative projects with day job, but that all ended when I turned 30. Had to quit my main job to fully focus on the startup i work on.