My gitlab is here: https://gitlab.com/the_turned_table/simply-distribute
Itβs currently bare-bones and itβs basically a POC. But my thoughts were that the application would be containerized and uses RabbitMQ as a message broker for data and requests.
I have got plans to better document my code and create architecture documents and all the rest.
Very happy with the current POC and itβs been a personal goal of mine to learn more about distributed file systems and distributed computing in general.
Very happy to hear any feedback on this.
Just out of curiosity, would there be any use case for implementing distributed computing to help train a neural net? I know George has said that "data" is not the problem at the moment, but I am thinking that this could be a useful way of allowing interested people to help improve the comma if they don't yet have a car that is supported. Disclaimer: I don't know much about ML, and put this here as a sounding board.
The idea comes from using things like "Folding@Home" or "Rosetta@Home," where they were able to get exaflops of compute to help fold proteins during the height of the COVID pandemic. Obviosuly, that task might be more ammenable to being parsed out into smaller segments, because the calculations can be run in paralell fairly easily. But if there is a way to use a distributed system like BOINC to help train the net, I'd be all for it. I think that fans of comma who have extra compute lying around (like my old computer that I don't use) could contribute in a small way to helping with training, and help improve the system more rapidly. Just a thought.
Apologies is this makes no sense, as for comparison, I am a bit of a tech. luddite.
The company that I work for recently started to migrate our monolith to microservices and this inspired me to try to build my own distributed system to learn the fundamentals.
After about a month of development in my free time, this is my result: https://github.com/evgenirusev/Distributed-eStore
I am sharing this mainly to ask for feedback from anyone who is experienced in the field about my approach and to have meaningful and insightful discussions with anyone interested. :)
Thanks in advance!
https://preview.redd.it/uq4b1ibu9jx61.jpg?width=1280&format=pjpg&auto=webp&s=955cd75c98baf6be80383ad4e2a27bce66a6000e
πhttps://www.kucoin.com/news/en-skey-rewards-has-been-distributed
Blog π’: No one likes to write dockerfiles and Kubernetes manifests in their development workflow. In the second (& final) installment of the series, we use Project Tye to make the development and deployment process of our microservices drop-dead simple.
Distributed Tracing in ASP.NET Core with Jaeger and Tye Part 2 - Project Tye
I am on Twitter π€ : https://twitter.com/rahulrai_in
Did the Pi Network have to register their project with the Securities and Exchange Comity in order to issue these βI owe you coinsβ representative of the real tokens being stored by them?
Can someone please clear this up for me?
https://github.com/jamespratt/configuration-service
Configuration service is a remote configuration host and client for .NET Core. With Configuration Service you can host configuration in a git repo or file system and automatically publish changes to containerized applications without having to redeploy. Any feedback is appreciated!
Does anyone here work at or know of distributed companies working on .NET and hire globally?
In case, you are not familiar with the term distributed company, it refers to companies with employees working from different timezones and remotely. They work remotely even during non-pandemic times. For example, GitLab, Buffer etc.
I have searched in some usual job boards such as StackOverflow, but most job posts out there hire only in US timezones. I am from South-East Asia and that's a no-go for most of these jobs. Sometimes they even try to hire me as a freelancer with comparatively low pay (even lower than what I can already earn working in local companies) than mentioned in the description.
Blog π’: No one likes to write dockerfiles and Kubernetes manifests in their development workflow. In the second (& final) installment of the series, we use Project Tye to make the development and deployment process of our microservices drop-dead simple.
Distributed Tracing in ASP.NET Core with Jaeger and Tye Part 2 - Project Tye
I am on Twitter π€ : https://twitter.com/rahulrai_in
Blog π’: Microservices are everywhere. What happens when one of the services of your application fails? Do you start inspecting issues from the head to the tail of your application?
Open Tracing and Jaeger can be your allies on the doomsday. Here's how:
π«ππππππππππ π»ππππππ ππ π¨πΊπ·.π΅π¬π» πͺπππ ππππ π±πππππ πππ π»ππ π·πππ 1 - π«ππππππππππ π»ππππππ
Follow me on Twitter π€ : https://twitter.com/rahulrai_in
