Get instant live expert help with SQL
“My expert helped me in less than 20 minutes, saving me what would have been 5 hours of work!”

Enter your problem description Our Experts are available now

Your message must be at least 40 characters
Select your SQL Environment:
Your message must be at least 40 characters
By submitting this form you agree to Got It's privacy policy.

What is Teradata and its Main Features – Querychat

In this article, we will discover what Teradata is, the features of Teradata SQL, and why this database differs from other relational DBMSs.

What is Teradata?

The main feature of this database is that it is parallel, although this technology is far from new.

What does parallelism mean? Teradata downloads, archives, and processes data in parallel. Depending on the configuration, Teradata allows us to perform hundreds, thousands of operations at the same time.

Teradata exists from 1978, and this is a highly parallel, Teradata-SQL system that shares no resources on the basis of mass-produced microprocessors, disks, and memory. Teradata systems act as SQL servers for user programs running on regular computers.

Teradata features

Teradata systems can have over 1000 processors and many thousands of drives. Functionally, Teradata processors are divided into two groups: interface processors (IFP) and access module processors (AMP). Interface processors communicate with the host machine, parse and optimize queries, and coordinate AMP during query execution. Access model processors are responsible for executing queries.

Teradata DBMS is a large database server that communicates with many clients through the TCP / IP protocol or a connection to the IBM universal computer channel.

Teradata has installed many systems with over 100 processors and hundreds of drives. These systems demonstrate almost linear acceleration and scalability on relational queries and significantly outperform traditional main-class machines in processing large (terabyte) databases.

Each relationship is hashed based on a hash between AMP subsets. When inserting a tuple in relation to the primary key of a tuple, a hash function is used to select the AMP where this tuple will be stored. Once the tuple enters the AMP, a second hash function is used to determine the position of the tuple in its fragment of the relation. 

The tuples in each fragment are arranged according to the value of this hash function. Given a key attribute value, we can find a tuple in a single AMP. AMP checks its cache memory, and if it doesn’t have such a tuple, it selects it for one disk read. Secondary hash indexes are also supported.

From Teradata Database, we can send requests for access, filtering, and returning datasets from Hadoop without the help of IT professionals.

Teradata works on MP-RAS UNIX, Microsoft Windows 2000/2003, and Server SuSE Linux.

The Teradata SQL is a regular SQL language, with some built-in functions that complement the main language.

The reasons to choose Teradata

There can be several reasons to choose this type of database:

  1. The database was designed and built on a parallel architecture from the very beginning.
  2. It allows us to support large amounts of information – more than 400 TB in one area.
  3. It supports modular expansion from small databases (10GB) to large (100+ TB).
  4. It provides a parallel-aware optimizer that eliminates the need for complex settings to receive a request.
  5. It automatically distributes data, excluding complex indexing schemes and laborious reorganizations.
  6. It supports ad hoc queries that use the ANSI SQL standard and include SQL database management information (log files), which allows us to submit queries to other database management systems in Teradata.
  7. Single point of administration for database administration (Teradata Manager).
  8. Several scripting languages are available for development: Perl, Ruby, Python, and R.

Instant livechat to an Expert!

Most of the time, the problem you will need to solve will be more complex than a simple method. If you want to save hours of research and frustration, try our live SQLExpert service! 

Our SQLQuerychat Experts are available 24/7 to answer any questions you may have. We guarantee a connection within 30 seconds and a customized solution within 20 minutes.

Leave a Reply

Notify of
Trusted by people who work at, Inc
Facebook, Inc
Accenture PLC
Siemens AG
The Allstate Corporation
United Parcel Service
Dell Inc