CDP Public Cloud: August 2020 Release Summary

Cloudera is pleased to announce the latest updates to CDP Public Cloud.

Highlights

• The Cloudera SLA for CDP Public Cloud has been published
• CDW enhancements including a reduced IP footprint and new HUE features
• CML enhancements including the ability to retry failed workspace upgrades and an engine upgrade
• [Tech Preview] Cloudera Streaming Analytics iis now integrated with Kudu and is available on both Azure and AWS

New Or Updated Capabilities

Documentation

• Service Level Agreement for CDP Public Cloud

  • The SLA document is posted here

Cloudera Data Warehouse

• Simplified Virtual Warehouse deployment for Azure with Network Overlay option

  • Network overlays make it easier to deploy a Virtual Warehouse by reducing the number of IPs required. This capability was previously released for AWS, and is now available for Azure.

• Improved Admin and End User experience for SQL Workbench (HUE)

  • Ability to do more advanced configuration options of HUE via this introduced safety valve feature in the admin view

  • Improved load balancer health visibility

Cloudera Machine Learning

• Kubernetes 1.16 Support

  • New ML Workspaces will be provisioned using K8s 1.16; this will keep CML ahead of Azure’s current deprecation plan of K8s 1.15

• Retry of ML Workspace Upgrades

  • If the upgrade of an ML Workspace fails due to an issue that has since been resolved in a newer version, you can now retry the upgrade from the UI to the latest version. This will help eliminate issues and enable more frictionless upgrades.

Cloudera Data Flow

• [Tech Preview] Cloudera Streaming Analytics (Flink) for Data Hub is now available on Azure

  • Streaming Analytics, powered by Apache Flink, is now available on CDP Public Cloud as a DataHub cluster template for Azure (previously it was available on AWS only)

  • This enables powerful analytics on top of streaming data, including Kafka topics on your Streams Messaging Data Hub clusters

  • This tech preview release ships the ​lightweight ​cluster template which is intended for small workloads as well as testing/dev environments

• [Tech Preview] Cloudera Streaming Analytics (Flink) integration with Kudu

  • Flink applications can store data in Kudu subsequently for fast analytics, and can query data from Kudu to load state

Cloudera Search

• [Tech Preview] Data Discovery and Exploration

  • Ability to index data in cloud storage into Solr, using Spark (i.e. the Crunch Indexer Tool)

  • To quickly produce prototype dashboards or demo applications on Solr, you can now use HUE Solr Dashboard

Cloudera Data Hub

• QueueManager ‘Absolute Mode’

  • YARN resource queues in Data Hub clusters can how be configured with fixed CPU/memory capacity using the visual Queue Manager component

• SQL Workbench (HUE)

  • Additional health checks for load balancer status

Data Catalog

• Search experience improvements

  • Find the data assets you are looking for faster by being able to search against all asset types simultaneously with a single query. Refine your search by filtering along facets such as asset types, owners, and classification tags.

• Simplified Tag management flows

  • Improved efficiency for data stewards with the addition of tags to assets and columns as well as the creation of new tags from the Data Catalog interface's asset 360 or search interfaces, all without having to go switch to Atlas.

Management Console & Control Plane

• SSH Key Management

  • A Power User can add and delete SSH keys for all users and users can add and delete their own SSH keys.

  • See Managing SSH keys.

• Choose SDX version at deployment

  • When deploying an environment, you can now select the Data Lake version.

  • See Register an AWS environment and Register an Azure environment.

• Edit some parameters of a previously deployed environment

  • Subnets: See Add subnets to an environment

  • Security Groups: See Add security groups to an environment

  • Root SSH key: See Add root SSH key to an environment

• [Tech Preview] Archive audit logs to third party storage

  • The Control Plane can now write audit logs to third party cloud object storage (e.g, a customer owned ADLS container)

  • This feature can only be configured via the CLI

Full release documentation including release notes for each service is available here (Service Name → Release Notes → What’s New).