By Brian Wood, Director of Cloud Marketing
The Teradata Everywhere and other Teradata PARTNERS Conference-related press announcements include a number of cloud-specific highlights for Teradata in the Cloud. This post summarizes the key takeaways.
First, a reminder for context: Teradata Hybrid Cloud is a customer environment using a mix of on-premises, private cloud, managed cloud, and public cloud resources orchestrated to work together. Hybrid Cloud’s ease-of-use benefits include automatic synchronization, optimized query routing, and end-to-end management across the ecosystem. This new level of flexibility and workload portability enables organizations to quickly and seamlessly optimize their data warehouse resources to match continuously changing business requirements.
Now on to the Teradata in the Cloud – specific announcement highlights
- Teradata Database on VMware – now MPP, scalable to 32 nodes
- Teradata Managed Cloud – expansion into German data center
- Aster Analytics on AWS – now available, scalable to 32 workers
- Teradata Database on AWS – now MPP, scalable to 32 nodes
- Teradata Database on Azure – available in Q4, MPP scalability
Top takeaways specific to Teradata Database on AWS
- MPP scalability up to 32 nodes
- Automatic node failure recovery
- Direct backup and restore to and from Amazon S3
- Direct querying of data in S3 via QueryGrid Teradata-to-Presto Connector
- One-click deployment automation via AWS CloudFormation
Let’s drill down into the new feature updates specific to Teradata Database on AWS
MPP Scalability to 32 Nodes
First, Teradata Database on AWS is now MPP (massively parallel processing, or multi-node) and can scale up to 32 nodes. This is a huge step forward compared to the previous single-node-only option available since Q1 2016. Thirty two nodes is a lot, but rest assured that Teradata will be “raising the roof” of scalability even more in the quarters to come.
Automatic Node Failure Recovery
Second, Teradata Database on AWS now includes automatic node failure recovery, which applies to MPP instances using Elastic Block Store (EBS). Node failures may be related to hardware, software, or the operating system. When a node fails, Node Failure Recovery:
- Automatically spins up a virtual hot standby node instance
- Detaches the EBS storage of the failed node
- Configures a new instance
- Reattaches the EBS storage to the new instance
- Reinstates the configuration
The replacement node is based off an image recorded when the system is first launched, or an updated image if the proper tools are run after a software upgrade. The replacement node has the same secondary IP, elastic IP, and identifiers as the replaced node. By default, the downed node continues running and can be used for diagnostic purposes.
Backup and Restore to/from Amazon S3
Third, in addition to using Elastic Block Store (EBS), customers may now directly backup and restore to and from Amazon Simple Storage Service (S3). Using Teradata Data Stream Utility, this capability applies whether the source is a cloud or on-premises Teradata system and also whether the restore target is a cloud or on-premises Teradata system.
The benefit of using low-cost Amazon S3 is that customers pay only for what they use and storage scales automatically. It is easy to create and manage jobs in Viewpoint for backup and restore at the database or object level, and incremental backup means that only changed blocks are streamed from the database. Multiple parallel data streams between each Teradata node and the Amazon S3 storage bucket minimizes backup and restore windows.
Direct Querying of Data in Amazon S3
Fourth, customers may now directly query data in Amazon S3, which is a welcome feature addition because it helps save time and money by avoiding egress charges and effort when moving data from S3 to Teradata Database.
Direct querying of Amazon S3 is achieved by using the Teradata-to-Presto QueryGrid Connector, which interfaces with the Teradata distribution of Presto running on Amazon Elastic MapReduce (EMR), the AWS Hadoop-as-a-Service offering, or either Hortonworks (HDP) or Cloudera (CDH) on Elastic Compute Cloud (EC2) instances.
The Teradata distribution of Presto is open source (and thus free of charge), but to get support for Teradata Presto nodes customers must engage Teradata Customer Services.
Automated Deployment via AWS Cloud Formation
Fifth, Teradata has harnessed AWS CloudFormation templates to help streamline the provisioning process for bringing up the numerous Amazon Machine Images (AMIs) involved in a Teradata ecosystem.
We’ve established three ecosystems: Test and Development, Minimum Production, and Full Production, each of which contains increasing quantities of Teradata software. To use the launch templates, customers first subscribe to the AMIs in their desired ecosystem. Once they do and once they launch the Teradata Ecosystem template, it’s a straightforward fill-in-the-blank step-by-step process to bring up the AMIs much easier than doing so manually.
Onward and Upward
What does it all mean for you?
Progress. Innovation. Investment. Acceleration. Responsiveness. And so much more.
I’m pleased to report that the Teradata team is FULLY ENGAGED in quickly developing and deploying the full capabilities of Teradata analytic ecosystem in the hybrid cloud environment.
It’s an exciting time to be part of a high-revving execution engine with the vision, resources, and leadership to make a real difference in customers’ businesses.
Please view our website pages – starting with Teradata Hybrid Cloud – to learn more.
Brian Wood is director of cloud marketing at Teradata. He is a results-oriented technology marketing executive with 15+ years of digital, lead gen, sales / marketing operations & team leadership success. He has an MS in Engineering Management from Stanford University, a BS in Electrical Engineering from Cornell University, and served as an F-14 Radar Intercept Officer in the US Navy.