Run C2D Jobs
Last updated
Was this helpful?
Last updated
Was this helpful?
To proceed with compute-to-data job creation, the prerequisite is
to select the preferred environment to run the algorithm on it. This can be
accomplished by running the CLI command getComputeEnvironments likewise:
Initiating a compute job can be accomplished through two primary methods.
The first approach involves publishing both the dataset and algorithm, as explained in the previous section, Publish a Dataset Once that's completed, you can proceed to initiate the compute job.
Alternatively, you have the option to explore available datasets and algorithms and kickstart a compute-to-data job by combining your preferred choices.
To illustrate the latter option, you can use the following command:
In this command, replace DATASET_DID with the specific DID of the dataset you intend to utilize and ALGO_DID with the DID of the algorithm you want to apply. By executing this command, you'll trigger the initiation of a compute-to-data job that harnesses the selected dataset and algorithm for processing.
For running the algorithms free by starting a compute job, these are the following steps.Note
Only for free start compute, the dataset is not mandatory for user to provide in the command line. The required command line parameters are the algorithm DID and environment ID, retrieved from getComputeEnvironments
command.
The first step involves publishing the algorithm, as explained in the previous section, Publish a Dataset Once that's completed, you can proceed to initiate the compute job.
Alternatively, you have the option to explore available algorithms and kickstart a free compute-to-data job by combining your preferred choices.
To illustrate the latter option, you can use the following command for running free start compute with additional datasets:
In this command, replace DATASET_DID with the specific DID of the dataset you intend to utilize and ALGO_DID with the DID of the algorithm you want to apply and the environment for free start compute returned from npm run cli getComputeEnvironments.
By executing this command, you'll trigger the initiation of a free compute-to-data job with the alogithm provided.
Free start compute can be run without published datasets, only the algorithm and environment is required:
NOTE: For zsh console, please surround [] with quotes like this: "[]".
To obtain the compute results, we'll follow a two-step process. First, we'll employ the `getJobStatus`` method, patiently monitoring its status until it signals the job's completion. Afterward, we'll utilize this method to acquire the actual results.
To monitor the algorithm logs execution and setup configuration for algorithm, this command does the trick!
To track the status of a job, you'll require both the dataset DID and the compute job DID. You can initiate this process by executing the following command:
Executing this command will allow you to observe the job's status and verify its successful completion.
For the second method, the dataset DID is no longer required. Instead, you'll need to specify the job ID, the index of the result you wish to download from the available results for that job, and the destination folder where you want to save the downloaded content. The corresponding command is as follows:
