New Features and Changes in Cloudera Data Science Workbench 1.7.1
Major features and updates for Cloudera Data Science Workbench.
- Supported upgrade paths to CDSW 1.7.1
- Cloudera Data Science Workbench only supports upgrades to version 1.7.1 from version 1.5.x and 1.6.x. If you are using an earlier version of CDSW, you must first upgrade to version 1.5.x or 1.6.x, and then upgrade to version 1.7.1.
- Analytical Applications
- Cloudera Data Science Workbench now gives data scientists a way to create
long-running standalone ML web applications/dashboards that can easily be shared
with other business stakeholders. Applications can range from single
visualizations embedded in reports, to rich dashboard solutions such as Tableau.
Applications stand alongside other existing forms of workloads in CDSW (sessions, jobs, experiments, models). For details, see Analytical Applications.
- Monitoring CDSW with Grafana
- CDSW now leverages Prometheus and Grafana to provide a dashboard that allows you to monitor how CPU, memory, storage, and other resources are being consumed by CDSW deployment. For details, see Cluster Monitoring with Grafana.
- Feature flag overrides
- This is a new property available in the CDSW service in Cloudera Manager. It can be used to enable/disable experimental features (such as quotas) and disable metric collection in diagnostic bundles.
- Quotas
- CDSW site administrators can now enable CPU, GPU, and memory usage quotas per
user. You can set default quotas for each user on the deployment as well as
overriding custom quotas for specific users. For details, see Configuring Quotas.
- Usage Metrics Collection
- By default, CDSW 1.7.1 now gathers highly redacted information on which feature
is being used on your deployment. When you create a diagnostic bundle, this
information is packed alongside the diagnostic information.
You can use the Feature flag overrides property in Cloudera Manager to disable collection of usage metrics.