Serverless Dask to make a mockery of your massive workloads
source link: https://www.producthunt.com/posts/hades
Go to the source link to view the article. You can view the picture content, updated content and better typesetting reading experience. If the link is broken, please click the button below to view the snapshot at that time.
Hades
Serverless Dask to make a mockery of your massive workloads
We built Hades as an internal tool to help scale our Dask workloads. We do a LOT of image processing on the geospatial side & have gotten some interest from customers of ours.
The problem we had was that iteration loops in data science projects can be months long to find high quality outputs and take a large team to get the project moving. End-to-end the process of standing up a team to build and evaluate new models and ideas takes months, and in some cases requiring budgets in the millions to be approved.
This is largely driven by the complexity and cost of curating analysis-ready data and building scalable compute infrastructure that requires teams of engineers to build.
The alternative is running analyses at slow speeds, wasting time and dramatically lowering the productivity of your team.
We built Hades.earth to save ludicrous amounts of time and cost by curating data and providing simple, supercomputer scale processing ability in only a few lines of code.
It's really helped us & we hope it can help you too!
Recommend
About Joyk
Aggregate valuable and interesting links.
Joyk means Joy of geeK