12. Funding, citing, and anonymous usage statistics
12.1. Citing Pegasus in Academic Works
Please see the Pegasus Website
12.2. Usage Statistics Collection
12.2.1. Purpose
Pegasus WMS is primarily a NSF funded project. As part of the requirements of this funding, Pegasus WMS is required to gather usage statistics of Pegasus WMS and report it back to NSF in annual reports. The metrics will also enable us to improve our software as they will include errors encountered during the use of our software.
12.2.2. Overview
We plan to instrument and augment the following clients in our distribution to report the metrics.
pegasus-plan
pegasus-transfer
pegasus-monitord
For the Pegasus WMS 4.2 release, only the pegasus-plan client has been instrumented to send metrics.
All the metrics are sent in JSON format to a server at USC/ISI over HTTP. The data reported is as generic as possible and is listed in detail in the section titled “Metrics Collected”.
12.2.3. Configuration
By default, the clients will report usage metrics to a server at ISI. However, users have an option to configure the report by setting the following environment variables
PEGASUS_METRICS
A boolean value ( true | false ) indicating whether metrics reporting is turned ON/OFF
PEGASUS_USER_METRICS_SERVER
A comma separated list of URLs of the servers to which to report the metrics in addition to the default server.
12.2.4. Metrics Collected
All metrics are sent in JSON format and the metrics sent by the various clients include the following data
JSON KEY |
DESCRIPTION |
|---|---|
client |
the name of the client ( e.g “pegasus-plan”) |
version |
the version of the client |
type |
type of data - “metrics” | “error” |
start_time |
start time of the client ( in epoch seconds with millisecond precision ) |
end_time |
end time of the client ( in epoch seconds with millisecond precision) |
duration |
the duration of the client |
exitcode |
the exitcode with which the client exited |
wf_uuid |
the uuid of the executable workflow. It is generated by pegasus-plan at planning time. |
12.2.4.1. Pegasus Planner Metrics
The metrics messages sent by the planner in addition include the following data
JSON KEY |
DESCRIPTION |
|---|---|
root_wf_uuid |
the root workflow uuid. For non hierarchal workflows the root workflow uuid is the same as the workflow uuid. |
data_config |
the data configuration mode of pegasus |
compute_tasks |
the number of compute tasks in the workflow |
dax_tasks |
the number of dax tasks in the abstract workflow (DAX) |
dag_tasks |
the number of dag tasks in the abstract workflow (DAX) |
total_tasks |
the number of the total tasks in the abstract workflow (DAX) |
dax_input_files |
the number of input files in the abstract workflow (DAX) |
dax_inter_files |
the number of intermediate files in the abstract workflow (DAX) |
dax_output_files |
the number of output files in the abstract workflow (DAX) |
dax_total_files |
the number of total files in the abstract workflow (DAX) |
compute_jobs |
the number of compute jobs in the executable workflow |
clustered_jobs |
the number of clustered jobs in the executable workflow. |
si_tx_jobs |
the number of data stage-in jobs in the executable workflow. |
so_tx_jobs |
the number of data stage-out jobs in the executable workflow. |
inter_tx_jobs |
the number of inter site data transfer jobs in the executable workflow. |
reg_job |
the number of registration jobs in the executable workflow. |
cleanup_jobs |
the number of cleanup jobs in the executable workflow. |
create_dir_jobs |
the number of create directory jobs in the executable workflow. |
dax_jobs |
the number of sub workflows corresponding to dax tasks in the executable workflow. |
dag_jobs |
the number of sub workflows corresponding to dag tasks in the executable workflow. |
chmod_jobs |
the number of jobs that set the xbit of the staged executables |
total_jobs |
the total number of jobs in the workflow |
In addition if pegasus-plan encounters an error during the planning process the metrics message has an additional field in addition to the fields listed above.
JSON KEY |
DESCRIPTION |
|---|---|
error |
the error payload is the stack trace of errors caught during planning |
Note
pegasus-plan leaves a copy of the metrics sent in the workflow submit directory in the file ending with “.metrics” extension. As a user you will always have access to the metrics sent.