Run any Python function on 1000 computers in 1 second.

Burla is the simplest way to scale python, it has one function: remote_parallel_map It's open-source, works with GPU's, custom docker containers, and up to 10,000 CPU's at once.

A data-platform any team can learn in minutes:

Scale machine learning systems, or other research efforts without weeks of onboarding or setup. Burla is open-source and can be depoloyed in your cloud with a single command.

How it works:

Burla only has one function:

from burla import remote_parallel_map

my_inputs = [1, 2, 3]

def my_function(my_input):
    print("I'm running on my own separate computer in the cloud!")
    return my_input
    
return_values = remote_parallel_map(my_function, my_inputs)

With Burla, running code in the cloud feels the same as coding locally:

• Anything you print appears in your local terminal.

• Exceptions thrown in your code are thrown on your local machine.

• Your local python packages are automatically synchronized with the cluster.

• Responses are pretty quick, you can call a million simple functions in a couple seconds!

Attach big hardware to functions that need it:

Zero config files, just simple arguments like func_cpu & func_ram.

from xgboost import XGBClassifier

def train_model(hyper_parameters):
    model = XGBClassifier(n_jobs=64, **hyper_parameters)
    model.fit(training_inputs, training_targets)
    
remote_parallel_map(train_model, parameter_grid, func_cpu=64, func_ram=256)

Simple, flexible pipelines:

Nest remote_parallel_map calls to build simple, massively parallel pipelines. Use background=True to schedule function calls that keep running after you close your laptop.

from burla import remote_parallel_map

def process_record(record):
    # Pretend this does some math per-record!
    return result

def process_file(file):
    results = remote_parallel_map(process_record, split_into_records(file))
    upload_results(results)

def process_files(files):
    remote_parallel_map(process_file, files, func_ram=16)
    

remote_parallel_map(process_files, [files], background=True)

Run code in any Docker image, using the latest GPU's:

Public or private, just paste a URI to your image and hit start. Burla works with any linux based Docker image.

Watch our Demo:

Get started now:

signup | Burladocs.burla.dev

---

Questions? Schedule a call, or email [email protected]. We're always happy to talk.