Home

Dask + Coiled

Extend your existing Python code

You don't have to migrate to a new system.

Dask and Coiled fit into your current workflows, and are easy for your teammates to use today.

SEE USE CASES

>>> futures = []

>>> for filename in filenames
...    data = client.submit(load, filename)
...    result = client.submit(process, data)
...    futures.append(result)

>>> for future in as_completed(futures):
...    ...

See more of this example

>>> import dask.dataframe as dd    

>>> df = dd.read_parquet(
...     "s3://bucket/input.parquet",
... )


>>> df = df[df.balance < 0]
>>> df.to_parquet("s3://bucket/output.parquet")

See more of this example

>>> import xgboost.dask, optuna

>>> def objective(...):
...     ...
...     train = xgboost.dask.DaskDMatrix(...)
...     ...
...     return score

>>> study.optimize(objective, ...)

See more of this example

$ pip install coiled
$ coiled setup  # connect coiled to your cloud 
$ ipython

>>> import coiled, dask
>>> cluster = coiled.Cluster(n_workers=200)

Get started with Coiled for Free

Feature Highlights

No Docker Required

Coiled syncs all packages you have installed to the cluster — even local code. You don’t have to worry about building docker images, specifying dependencies, or mismatched versions.

Learn More

Dask + Coiled

Parallelize with Dask

Dask breaks your computation into thousands of small tasks and runs those tasks on distributed hardware.

Dask works great on a laptop, or on massive clusters.

REQUEST A DEMO

Dask + Coiled

Manage cloud infrastructure with ease

Coiled manages your cloud resources.

It's 100x easier than Kubernetes or native cloud APIs.

Observe and Optimize

Stuff breaks. Machines cost money.

Easily discover how to make your team more effective and more cost efficient. Coiled makes it easy to use spot, ARM, GPUs, and other cost-saving measures.

build vs. buy comparison

"Changed how we work"

"The speed is nice, sure, but the real benefit is taking a multi-day effort and finishing it in an afternoon. Coiled changed the character of our work."

Matt Plough

Software Engineer, KoBold Metals

"Easy and reliable API"

"Coiled is a super simple way to launch a cluster with a simple API that's hard to break."

Eric Jeske

Head of Data Science, Telemetry

"Amazing Support"

"Coiled support is amazing. I’ll run into an issue and before I have a chance to mention it I have an email in my inbox. You don’t get this kind of support with large companies."

Katya Potapov

Software Engineer, Floodbase

"Burst to the cloud"

"Quite literally ‘burst to the cloud from your laptop’, everything I've been dreaming of since grad school."

Eric Ma

Data Scientist, Moderna

"Dead Simple"

"Dask was simple. Coiled made it simpler."

"Set up is half the battle when working with cloud...Coiled made this easy."

Kenneth Nguyen

CTO, Tasq

"From weeks to hours"

"If I did this on my computer it would take a week. Instead it takes an hour."

"I love package sync"

Mohamed Akbarally

Data Scientist, With Marmalade

"The speed is nice, sure, but the real benefit is taking a multi-day effort and finishing it in an afternoon. Coiled changed the character of our work."

"Super Simple"

"Coiled is a super simple way to launch a cluster with a simple API that's hard to break."

Eric Jeske

Head of Data Science, Telemetry

"Burst to the cloud"

"Quite literally ‘burst to the cloud from your laptop’, everything I've been dreaming of since grad school."

Eric Ma

Data Scientist, Moderna

"Simple and Easy"

"Dask was pretty simple, and Coiled made it even simpler. Set up is half the battle when you’re working with cloud. Coiled definitely made it much simpler if I just want to run something very easily."

Kenneth Nguyen

CTO, Tasq

"Very Fast"

"If I just did this on my computer it would take days. Instead it takes an hour. Setup was honestly very fast."

Mohamed Akbarally

Data Scientist, With Marmalade

"Amazing Support"

"Coiled support is amazing. I’ll run into an issue and before I have a chance to mention it I have an email in my inbox. You don’t get this kind of support with large companies."

Katya Potapov

Software Engineer, Floodbase

Cloud Computing
for Data People

Use the Libraries you Love

Part of the Python Ecosystem

Parallelize your code in 4 easy steps

1. Code

2. Parallelize

3. Scale Up

4. Optimize

Extend your existing Python code

No Docker Required

Parallelize with Dask

Manage cloud infrastructure with ease

Observe and Optimize

Trusted by data science teams, large and small

Companies using Dask

Dask adoption by the Python community

Companies Using Dask

Python Users Switching from Spark to Dask

"Changed how we work"

"Easy and reliable API"

"Amazing Support"

"Burst to the cloud"

"Dead Simple"

"From weeks to hours"

"Super Simple"

"Burst to the cloud"

"Simple and Easy"

"Very Fast"

"Amazing Support"

Dask and Coiled in Production

Try Coiled for Free

Set up in your cloud for Free

Use Cases and Tutorials

See Pricing

Cloud Computingfor Data People

Use the Libraries you Love

Part of the Python Ecosystem

Parallelize your code in 4 easy steps

1. Code

2. Parallelize

3. Scale Up

4. Optimize

Extend your existing Python code

No Docker Required

Parallelize with Dask

Manage cloud infrastructure with ease

Observe and Optimize

Trusted by data science teams, large and small

Companies using Dask

Dask adoption by the Python community

Companies Using Dask

Python Users Switching from Spark to Dask

"Changed how we work"

"Easy and reliable API"

"Amazing Support"

"Burst to the cloud"

"Dead Simple"

"From weeks to hours"

"Super Simple"

"Burst to the cloud"

"Simple and Easy"

"Very Fast"

"Amazing Support"

Dask and Coiled in Production

Try Coiled for Free

Set up in your cloud for Free

Use Cases and Tutorials

See Pricing

Cloud Computing
for Data People