What is Teradata?

 

What is Teradata?

Teradata is a petabyte-scale, highly customizable MPP database solution. In existence for over 30 years, Teradata has an extremely mature feature set that caters to the most demanding requirements of some of the largest organizations in the world. Its shared-nothing architecture makes Teradata a powerful database capable of linear scalability.

While Teradata has traditionally been installed on-premises with proprietary hardware, Teradata has recently begun offering its database technology as a managed service in AWS or in their own private cloud.know more at Teradata online training

Choosing Teradata as a database

Platform maturity

Because Teradata is one of the oldest and largest database providers in the world, the company has been able to built a completely comprehensive feature set that makes it possible to make Teradata work for almost any use case. A couple notable features include , a federation layer that enables access to datasets stored in Teradata and Hadoop, and Teradata’s powerful Workload Management software.

Platform customization

Given that Teradata bundles both the database hardware and software together, it’s possible to select from a large selection of platforms that vary based on workload. This makes it very easy to select hardware that’s optimized for your use case.

Performance

Some of Teradata’s features, if smartly implemented, can make Teradata an extremely performant database. For example, Teradata implements a hybrid row store / column store, which allows the user to store data either in row or column format. Another feature that boosts performance is the ability for Teradata to  into in-memory storage.

What is Teradata great for?

Fitting into a large enterprise data stack

The value of Teradata’s database solution can really only be realized at scale. Teradata’s massive functionality, integration, and concurrency is designed to be able to fit seamlessly into an enterprise data stack and be leveraged by a large number of users.

Building a custom data stack from the ground up

Given Teradata’s mature feature set, large number of platform offerings, and ability to exist both on-premise and in the cloud, Teradata’s RDBMS is perfect for competent DBAs that want to customize their data stack from top to bottom.know more at Teradata training

Price considerations

Pricing for Teradata is going to vary based on whether the database is hosted on-premises or in the cloud. For Teradata deployed in the cloud, the pricing will depend on the the hosting provider (AWS / Azure / Teradata IntelliCloud).

Teradata on-premises

On-Premise pricing data for Teradata will depend on the type, number, and size of the appliances purchased from Teradata. Specific pricing information is not included on the Teradata site.

Teradata on AWS

Teradata on AWS is sold on the AWS Marketplace, and comes with a pricing calculator that you can use to estimate what the monthly rate for an instance of Teradata would look like. Variables include feature set, storage type, instance size, and number of nodes. It’s possible to scale up to 32 nodes on AWS.

Teradata on Azure

Teradata on Azure is sold on the Azure Marketplace. The only plan that is offered on the Azure marketplace is an hourly plan. Teradata also integrates with Azure Blob Storage to manage backups and store additional data. It’s possible to scale up to 32 nodes on Azure.

Teradata IntelliCloud

Teradata also offers a fully managed software and hardware hosting option on their private cloud, called Teradata IntelliCloud. Pricing options vary for different Teradata and can be found here. Hosted by Teradata, the Teradata database is fully scalable.

Teradata architecture

Like most self-managed MPP databases, Teradata distributes both storage and compute to multiple nodes within its system. This architecture, known as a “shared-nothing-architecture”, means that nodes contain both compute (in the form of CPU and memory), and storage (in the form of disk space).

There are several components to the 

  • Parsing Engine (PE) - Similar to a “leader node”, the parsing engine receives the request from a user, and uses the SQL from the user to construct a query plan that is then distributed to the BYNET
  • BYNET - The communication engine between the parsing engine and the nodes (known as AMPs). BYNETs receive the query plan form the PE, and then decide which AMPs must be utilized.
  • Access Module Processor (AMPS) - AMPs function as the “worker nodes” which contain a section of the distributed table, and process the query for the section of the data that resides on their individual disks. The intermediate results are then passed through BYNET to the PE, which in-turn passes that data to the user.

Types of data

Teradata stores the data in either row or columnar format, depending on the table  (DDL*). It can ingest both tabular and nested data structures originating from(delimited) files, larger character objects (BLOB and CLOB), as well as

*Note that while Teradata can orient table storage in either direction (row or column), it will operate on the data in row format upon computation. Therefore, typical performance gains from column-based storage may be lower when compared to a native columnar database engine (Amazon Redshift or Vertica).

Maximum recommended size of data

Teradata can support 50+ petabytes in the on-premises capacity. You can attain similar scalability with their private cloud, as it will run on the same hardware that powers on-premises installations. Since Teradata is still figuring out the intricacies of the public cloud, Teradata on AWS is a single-node cluster only with support for up to 48 TB of HDD storage (SSD storage limits are lower). know more at Teradata online course

Database management for Teradata

Implementing Teradata

Teradata offers step-by-step instructions for implementing Teradata on and Azure. For details about implementing Teradata on-premise, or on Teradata’s private cloud, it’s best to 

Teradata also provides to help you get set-up with their platform.

Process for new data

You can leverage a standard SQL operation when ingesting data into Teradata. Higher volumes require Teradata’s proprietary ETL tools, called know more at Teradata online training from India

Maintenance

While the hardware component of Teradata can be fully managed, table architecture and query optimization is not. In order to attain acceptable performance out of Teradata, you must apply column encoding, partitioning, indexes, and statistics. General MPP tuning strategy can be applied when tuning a Teradata database.

Comments

Post a Comment

Popular posts from this blog

What is Azure?

  What is Azure? Microsoft Azure, formerly known as Windows Azure is a  Microsoft public cloud computing environment.  This platform provides a wide range of services including compute, storage, networking, and networking. Here the users can pick and scale these services to develop and scale applications (or) run the existing applications in the public cloud. This Azure platform aims to help business challenges and meet their organizational goals.  It offers tools that support all industries including e-commerce, finance, and a variety of fortune 500 companies.  This platform is compactable with various open-source technologies. This platform is capable of four cloud computing platforms like: a)Infrastructure as a Service(Iaas) b)Platform as a Service(Paas) c)Software as a Service (Saas) and d)Server less Besides, this cloud computing platform is an online portal that allows you to access and manage cloud services provided by Microsoft. These services and resour...
Read more

SharePoint Interview Questions

  1) What is MS Share-Point is used for? MS Share Point is used for multiple purposes, and it acts as a back-end system that combines all your employees PC’s and mobile device together allowing them to communicate symmetrically. know more at Share point online training Share documents:  It allows to store and share documents by anyone in the company also it allows to work on a single document by several employees at same time Intranet:  Dashboards can be customized by department and access to all employees are given to check the regular updates into the company Extranet:  With the help of SharePoint, you can provide an outside company a space or a site to upload all the information your company may require from them Websites:  You can use SharePoint to manage and build a publicly facing site as well.  display relationship between different inputs through charts and graphs Collaboration:  Through SharePoint employees can be interlinked with each other a...
Read more

Selenium Interview Questions

  1) What is Automation Testing? Automation testing or Test Automation is a process of automating the manual process to test the application/system under test. Automation testing involves the use of a separate testing tool which lets you create test scripts which can be executed repeatedly and doesn’t require any manual intervention. know more at Selenium online training Q #2) What are the benefits of Automation Testing? Benefits of Automation testing are: Supports execution of repeated test cases Aids in testing a large test matrix Enables parallel execution Encourages unattended execution Improves accuracy thereby reducing human-generated errors Saves time and money Q #3) Why should Selenium be selected as a test tool? Selenium is a free and open source have a large user base and helping communities have cross Browser compatibility (Firefox, Chrome, Internet Explorer, Safari etc.) have great platform compatibility (Windows, Mac OS, Linux etc.) supports multiple programming langu...
Read more