Startup Aster Offers Software For Clustering Commodity Servers - InformationWeek

InformationWeek is part of the Informa Tech Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them.Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

IoT
IoT
Government // Enterprise Architecture

Startup Aster Offers Software For Clustering Commodity Servers

Aster says its technology can turn commodity servers into a massively parallel processing relational database for analytics.

Startup Aster Data on Tuesday launched clustering software that can turn commodity servers into a large analytics database.

Aster nCluster is built on a series of patent-pending algorithms and processes that control the placement, partitioning, balancing, replication, and querying across clusters of intelligent nodes, the company said. The system is capable of scaling to hundreds of microprocessors.

By using Aster technology to turn commodity servers into a massively parallel processing relational database for analytics, companies have a less expensive alternative to traditional data warehouses, Aster executives claim. Besides spending less through the use of off-the-shelf computers, companies have a system that's scalable not just in the amount of data that can be stored, but in other areas too.

For example, a cluster can be partitioned into specific tasks, so if more processing power is needed for extracting, transforming, and loading data, a customer can dedicate more servers to that activity. "Usually when people say scalability, they think data size," Tasso Argyros, chief technology officer and co-founder, told InformationWeek. "But it's actually much more than that."

The nCluster user interface provides a single, unifying view of the system, so the cluster appears to be one data warehouse. The user interface makes it possible for one-node-at-a-time incremental scale in a "plug-and-play" manner, according to Aster. Similarly, nodes can be removed to phase out failed or older generation hardware.

Ncluster manages data in locally attached storage on individual nodes. A cluster is divided into three separate classes: Queens, Workers, and Loaders. Queen nodes provide the external single-system interface to the data warehouse, giving system administrators monitoring capabilities. The Queen nodes also handle query processing, result aggregation, and failover.

Worker nodes are responsible for storing partitions of data and replicas of data. They also participate in query processing and maintenance tasks, such as indexing, backup, and load balancing, as invoked by the Queen nodes.

Loader nodes are responsible for partitioning and loading new data into the Worker nodes, In addition, the former can also export data for use in other systems, disaster recovery, and data backup.

MySpace, the largest social network on the Web with 58.7 million visitors in April, according to Nielsen, is a customer of Aster. MySpace went into production in January with Aster software, which is being used to analyze user behavior, advertising opportunities, and other areas related to business intelligence. The social network has built a system of more than 100 nodes, according to Aster.

Aster was founded in 2005 by three colleagues in Stanford University's Computer Science Department. Mayank Bawa, chief executive, developed algorithms for querying distributed systems, and CTO Argyros researched large-scale data clusters. The third co-founder and chief scientist George Candea worked with Oracle and in research for IBM and Microsoft.

In launching its product, Aster is entering into a crowded data warehousing market that includes large players like Hewlett-Packard, Microsoft, and IBM. Aster is partnering with business intelligence software makers, including SAP-owned Business Objects. The BI software runs on top of the Aster system.

Pricing for the Aster nCluster starts at $100,000 and is based on the size of the user's data.

We welcome your comments on this topic on our social media channels, or [contact us directly] with questions about the site.
Comment  | 
Print  | 
More Insights
Slideshows
10 Ways to Transition Traditional IT Talent to Cloud Talent
Lisa Morgan, Freelance Writer,  11/23/2020
News
What Comes Next for the COVID-19 Computing Consortium
Joao-Pierre S. Ruth, Senior Writer,  11/24/2020
News
Top 10 Data and Analytics Trends for 2021
Jessica Davis, Senior Editor, Enterprise Apps,  11/13/2020
White Papers
Register for InformationWeek Newsletters
Video
Current Issue
Why Chatbots Are So Popular Right Now
In this IT Trend Report, you will learn more about why chatbots are gaining traction within businesses, particularly while a pandemic is impacting the world.
Slideshows
Flash Poll