Solved my scheduling issues for python data pipelines
What do you like best about the product?
The ability to pull code from github repo, run any bash code you need during machine load and schedule the code to be run when needed.
What do you dislike about the product?
It would be fantastic if they could run Knime workflow like they do Jupyther notebooks.
What problems is the product solving and how is that benefiting you?
Although lambda functions evolved a lot, it is not simple to run data pipelines directly from jupyther notebooks. Saturn cloud solves this issue. It is really easy to implement an scheduled jupyther run.
Try Saturn for better computing
What do you like best about the product?
more space , flexible , multiple node access , best platform to learn
What do you dislike about the product?
unfamiliar , poor branding even though with good features
What problems is the product solving and how is that benefiting you?
if i want to go with different modeules like pytorch , tensorflow i found all under one roof
Nizar's review
What do you like best about the product?
I wanted an easy way to run my jupyter notebook on the cloud. I was using google colab and I wasn't getting the performance I expected.
What do you dislike about the product?
Nothing for now. I have yet to try it but I hope I get the performance boost I need.
What problems is the product solving and how is that benefiting you?
I'm building a deep learning model that requires significant computation power
Saturn Cloud is one of the best AI computation platform I ever used
What do you like best about the product?
1. Stable services. I wouldn't need to worry about downtime or crashes when using it.
2. Rich features. It supports commonly used features like Jupyter notes and bare vms.
What do you dislike about the product?
It would be very appreciated if Saturn Cloud could provide more free computation time. Currently, it only provides 30 hours per month. But most of the other vendors could provide 30 hours per week.
What problems is the product solving and how is that benefiting you?
The free computation which Saturn Cloud provide is really helpful, especially for individual developers and small companies. For those kinds of users, they may not have enough resources and funding to start their projects.
Great support, good availability, and seamless integration capabilities
What is our primary use case?
I am primarily using Saturn Cloud as a student, primarily for training deep reinforcement learning (RL) agents. My projects usually involve Python 3.10, CUDA, and PyTorch and they typically require heavy computation power like GPUs and multi-core CPUs.
Saturn Cloud's environment has been able to sufficiently cater to these needs, providing a platform where I can run and manage these resource-intensive tasks easily. The support has been very helpful in clearing up questions, such as creating a custom image.
How has it helped my organization?
Before using Saturn Cloud, I was relying on my laptop for running my deep RL models. The training process was time-consuming and it was difficult to multitask. However, with Saturn Cloud, I can offload this heavy computing to the cloud. This not only accelerated my research process but also enabled me to multitask efficiently.
It didn't take long to see that Saturn Cloud could scale with my needs, providing more resources when required. This scalability is invaluable for a student researcher like myself where tasks can vary widely in computational requirements.
What is most valuable?
One of the features I appreciate the most about Saturn Cloud is its seamless integration with Jupyter notebooks. It provides an interface that I am familiar with and use extensively. It makes it easy for me to run, track, and debug my RL models.
The second feature I like is the ability to easily scale up and down the resources (like GPUs and CPUs), providing a flexible and cost-effective solution.
The third feature that is useful is the availability of pre-configured environments. It saves a lot of time and hassle, especially when working with complex setups involving packages like CUDA and PyTorch.
What needs improvement?
While Saturn Cloud offers a rich set of features, there are areas where I feel there could be improvements. For instance, the process of setting up custom environments could be more user-friendly. Currently, it is a bit technical and might be daunting for beginners.
Additionally, providing more detailed and beginner-friendly documentation, especially for advanced features, could greatly enhance the user experience. So far, I was relying on the support for setting up custom images and understanding why things don't work properly.
For how long have I used the solution?
I've been using the solution for one month.
What do I think about the stability of the solution?
What do I think about the scalability of the solution?
The solution is very scalable; machines can go up to extreme configurations.
How are customer service and support?
Support is very friendly, proficient, and helpful. They typically reply within hours.
How would you rate customer service and support?
Which solution did I use previously and why did I switch?
Before using Saturn Cloud, I was primarily running models on my laptop and occasionally using Colab. However, Colab was not feasible with the captchas and deleting the runtimes after a few minutes of being away from the keyboard.
As my models became more complex and required more computational power, I found that this was not scaling well with my needs. I switched to Saturn Cloud primarily for its superior computational power, scalability, and ease of use. In the beginning, the free 30 hours of computing convinced me to try it.
How was the initial setup?
The custom image was a bit more tricky and I required support. That said, it was solved within the same day.
What about the implementation team?
The initial setup was handled in-house.
What was our ROI?
I don't have an ROI; this is an academic project I am working on.
What's my experience with pricing, setup cost, and licensing?
In terms of pricing, they should make it as transparent as it is here.
Which other solutions did I evaluate?
I did not evaluate other options.
Which deployment model are you using for this solution?
Public Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Other
its easy to work with kind of platform
What do you like best about the product?
easy to learn the working steps, I find every required tool in one place which makes me like this platform even more. this platform is user-friendly.
What do you dislike about the product?
still to find one which i am uncomfortable with.
What problems is the product solving and how is that benefiting you?
its best platform to work on if you are working in machine learning field.
Great Resource for "upping" your research
What do you like best about the product?
User friendly-good information-lots of examples and visuals
What do you dislike about the product?
It can be a bit busy, but the information is helpful
What problems is the product solving and how is that benefiting you?
It is giving me another source of information that helps in my understanding.
Good JupyterLab experience
What do you like best about the product?
Good conda support with built-in mamba for faster package resolution. Setting up a new project is quick and easy, with various docker images available and support for SSH connections.
What do you dislike about the product?
The 30h per month isn't a lot considering other free alternatives, but none of them support stock JupyterLab so that is acceptable, although more hours would be welcome.
What problems is the product solving and how is that benefiting you?
Running computation locally is costly and requires a considerable amount of man-hours. Saturn Cloud lets us quickly iterate over our ML tasks without worrying about maintaining computing infrastructure on-premises.
Very helpful support, easy to use, transparent prices, and great compute resources
What do you like best about the product?
I love that you can build your own images, spin up pcs of your own configuration. Additionally, the platform is amazingly easy to use.
The support answers quickly - I had my problems solved within 1-2 hours always.
What do you dislike about the product?
I had to build my own image for python 3.10, but that was no problem at all, with the help of the support.
The default shut-down timer is at 1 hour, which I didnt see at first so watch out for that.
What problems is the product solving and how is that benefiting you?
Training Reinforcement Learning algorithms on GPUs for my Master's thesis. Solely using CPU is not feasible in my case and my own laptop is not able to do this computation.
Best service for fast scaling and training for teams!
What do you like best about the product?
The part I love the most - how easy it was to configure everything and start training. Intuitive UX and nice docs!
What do you dislike about the product?
Nothing yet. I have not met anything that I dislike
What problems is the product solving and how is that benefiting you?
Scaling of training and optimization processes
