Hi Bitecode, hope you are fine. I have two new questions :)
1 - what do you think of the recent Redis license changes?
2 - Have you use Django Huey? I ask a seasoned Django developer (who contributes to the project) what he thinks about it, and there are some concerns about synchronizing multiple instances (machines), he thinks it was designed for a single node. Have you encountered similar issues? How did you solve them if it is the case?
1 - I'm not a good weatherman, so I try to stay away from predicting the future too much. Given I have a huge respect for antirez, and that he chose the people at Redis Lab, I'll give them the benefit of the doubt and wait a year or two and see how it turns out.
2 - Yes. It's a good tool, and indeed it works better with a single node in mind. You can use multiple consumers (https://huey.readthedocs.io/en/latest/consumer.html#multiple-consumers), and load balance your redis instance, but I wouldn't call it battle field proof: not enough projects use it that way to find where the gremlins are hidden. And there are always gremlins in the network.
I'm hesitating to pick a task queue for a project. Celery seems to be the default choice for the researchs I made, but Huey seems interesting. The only disavantage I see at the moment, is that Celery has an additional project flower to see how tasks are working, but it seems RedisInsight can be a solution for that issue 🤔
We use celery extensively, but I'm interested in learning more about Django Q2 and Redis queue, in addition to Huey.
One thing I would like to better understand is monitoring resource usage. How much do my workers need? We use K8s pods, and their still somewhat of a mystery to me.
Hi Bitecode, hope you are fine. I have two new questions :)
1 - what do you think of the recent Redis license changes?
2 - Have you use Django Huey? I ask a seasoned Django developer (who contributes to the project) what he thinks about it, and there are some concerns about synchronizing multiple instances (machines), he thinks it was designed for a single node. Have you encountered similar issues? How did you solve them if it is the case?
I know, it is more than 2 questions XD
1 - I'm not a good weatherman, so I try to stay away from predicting the future too much. Given I have a huge respect for antirez, and that he chose the people at Redis Lab, I'll give them the benefit of the doubt and wait a year or two and see how it turns out.
2 - Yes. It's a good tool, and indeed it works better with a single node in mind. You can use multiple consumers (https://huey.readthedocs.io/en/latest/consumer.html#multiple-consumers), and load balance your redis instance, but I wouldn't call it battle field proof: not enough projects use it that way to find where the gremlins are hidden. And there are always gremlins in the network.
Thank you for your reply :)
Thanks for the great article (as always).
Redis is awesome, so I'm sure you'll appreciate a modern replacement for it called Dragonfly.
https://github.com/dragonflydb/dragonfly
https://redis.com/blog/redis-architecture-13-years-later/
Thanks for the article.
Quick question: LINDEX colors 1 # like colors[1]
Why green here?
Best!
The reason is very technical and requires a deep understanding of...
nah, I fucked up the copy/paste
I have to keep up my spree of nitpicking something in every new post so it's very considerate from you to hide something for me to find :-)
I wish I could hide them. But it's kinda of a grandiose gift.
Thank you for the article. Great as always.
I'm hesitating to pick a task queue for a project. Celery seems to be the default choice for the researchs I made, but Huey seems interesting. The only disavantage I see at the moment, is that Celery has an additional project flower to see how tasks are working, but it seems RedisInsight can be a solution for that issue 🤔
Then you will like the soon to be written article on huey.
We use celery extensively, but I'm interested in learning more about Django Q2 and Redis queue, in addition to Huey.
One thing I would like to better understand is monitoring resource usage. How much do my workers need? We use K8s pods, and their still somewhat of a mystery to me.
hehe! I can't wait for! 😝