The PiCloud Platform gives you the freedom to develop your algorithms and software without sinking time into all of the plumbing that comes with provisioning, managing, and maintaining servers.
# your program running locally $ my_program data/input1
# to offload to PiCloud, first push your data $ picloud volume sync data/ my_volume: # then execute your program (returns job id) $ picloud exec -v my_volume "my_program data/input1" 84
$ picloud result 84 # grab result of computation
An environment is the filesystem your jobs see as they run. We give you the ability to fully customize the environment.
You setup the environment once. When you run a large workload, we automatically distribute your environment to all the cores that need it.
Our object store makes it easy to keep your data in the cloud for efficient processing by our nodes.
# store a file into bucket $ picloud bucket put data_file data_obj # list contents of bucket $ picloud bucket list data_obj # retrieve the file $ picloud bucket get data_obj .
# store a file into bucket >>> cloud.bucket.put('data_file') # list contents of bucket >>> cloud.bucket.list() ['data_file'] # retrieve the file >>> cloud.bucket.get('data_file')
Volumes let you synchronize directories with the cloud. Your jobs see them as mounts on the filesystem.
# create your volume $ picloud volume create volume_name /mnt_path # push data to the volume $ picloud volume sync local_dir volume_name: # pull data from the volume $ picloud volume sync volume_name: local_dir
# create your volume >>> cloud.volume.create('volume_name', '/mnt_path') # push data to the volume >>> cloud.volume.sync('local_dir', 'volume_name:') # pull data from the volume >>> cloud.volume.sync('volume_name:', 'local_dir')
We consolidate all of your history into a single interface. You can see a task's:
We believe that the simpler we make cloud computing, the more powerful it is for you. Developer productivity is our number one goal.
See our documentation for a taste →
If your application needs compute cycles, we can help. For example:
Our latest feature gives you a distributed, fault-tolerant message queue with a scalable message processing system. Learn more →
The PiCloud Notebook is a Python console in your browser optimized for data analysis and collaboration. Check it out →
To give back to the community, we are awarding three researchers 10,000 free core hours. Apply today →
We have published our Spotathon Application to share our insights into the spot instance market with the greater community. Learn more →
> Geordie Rose, CTO, D-Wave Systems.
> Gary Rose, Head of R&D, Zinc.TV
> Stephen Elliott, Senior Product Manager at Amazon.
> Joemar Taganna, Flanders Institute for Biotech.
Quantum computing pioneer, D-Wave, has sped up machine learning workloads 1,000 fold using PiCloud. Read it here →
Internet television dashboard Zinc.TV aggregates daily video content from over 500 web properties using PiCloud. Read it here →