Aster Data FAQ

Why is an analytic platform architecture needed for
advanced analytics on big data?

The continuing explosion in data sources and volumes strains traditional analytic architectures that rely on moving data from where it is stored to where it is processed. As data volumes continue to grow, pulling terabytes and more of data through the data pipeline to analytic applications results in too much latency, cost, and overhead, making it impossible to include full data sets and fresh data in timely analysis.

At the same time, a new generation of analytics has emerged with new requirements:  analysis of large volumes of multi-structured data, ultra-fast results, and deep data exploration through ad-hoc and investigative analysis. However, it is difficult or impossible to perform these analytics in SQL, and user-defined functions (UDFs) can limit the type of analysis that can be done with acceptable performance and complexity. An analytic platform lets you move analytic applications to the data for maximum performance and the necessary complexity of analysis.

What is needed to handle advanced analytics on big data?

To eliminate the complexity and cost of moving data to analytic applications, you must move application processing to data. This requires a rich environment for embedded analytic processing that overcomes the limitations of SQL. Additionally, both data management and application processing must be treated as "first-class citizens" in the design of the analytic platform.

What is an analytic platform?

An analytic platform is a massively parallel software solution that embeds analytic processing with data stores for big data analytics. It not only stores large volumes of data but also processes data and analytic applications, enabling advanced analytics for deeper business insight. Like a traditional database it provides high-performance, scalable data management. As an analytic platform, it provides a rich environment for in-database processing of analytic applications so that applications can be co-located with data, eliminating data movement and improving performance and scalability. Both data management and analytics processing are treated as first-class citizens within the database.

What is Aster Data nCluster?

Aster Data nCluster is the leading analytic platform, a massively parallel software solution that embeds MapReduce analytic processing with data stores for big data analytics that incorporate new multi-structured data sources and types. nCluster includes an embedded database platform with capabilities including an ANSI SQL interface, ACID transactions, a cost-based query optimizer, indexes, Information Lifecycle Management, compression, security and other features. Its embedded analytic processing engine provides a rich application processing environment-- multiple language support via SQL-MapReduce®, workload management, security, statistics collection, error logging and other critical features.

What are the technical characteristics of an analytic platform?

Architected to serve the needs of users and analysts performing advanced analysis that scales to big data, an analytic platform delivers these key characteristics:

  • Scalability: an analytic platform can scale to load, store, query, and analyze terabytes and petabytes of data without encountering performance bottlenecks
  • Continuous availability: an analytic platform avoids unplanned downtime with fault tolerant features and eliminates planned downtime and performance degradation for routine administration and maintenance
  • Embedded analytics engine: an analytic platform provides a rich environment for processing custom and packaged analytics inside the database rather than being forced to move data to a separate application processing platform.
  • Hybrid row and column storage: a hybrid row-based and column-oriented architecture allows an analytic platform to meet the computation needs of ad-hoc, interactive analytics which are best optimized through a row store while also delivering high performance on more predictable, reporting-style queries best optimized by a column store.

What are the benefits of an analytic platform?

Because applications and data are co-located inside a massively-parallel processing (MPP) analytic platform, data access by applications is local and application processing is done in parallel. This allows you to analyze fresh data as soon as data enters the database and also lets you analyze data more frequently because performance of analysis is improved by 10x-100x over other architectures. The end result empowers your applications and users to rapidly generate insights and updates that help create competitive advantage.

How does Aster Data nCluster enable applications to perform analytics on big data?

Aster Data nCluster enables any packaged or custom analytic application logic written in a variety of languages (e.g. Java, C, C++, Perl, Python, etc.) to be embedded within the analytic platform without changes to the analytic logic. nCluster automatically parallelizes processing of embedded applications using its native in-database MapReduce and patented SQL-MapReduce® framework. As a result, applications running within nCluster can read, write, and update data with ultra-fast performance and massive scalability without having to summarize, window, or sample the data.

Do applications need to be re-written to embed them within
Aster Data nCluster?

No, application logic does not need to be re-written to be embedded within nCluster. Aster Data customers and partners have embedded their existing applications (written in Java, C, C++, Perl, R, etc.) without any change in their application logic. Aster Data’s tools and wizards, including Aster Data Developer Express, make it easy to take application logic, package it for nCluster, and push down the application into the nCluster database.

Do you have any other questions? Please ask.

Want to Learn More?
    1.888.Aster.Data Demo
Top Picks
Whitepaper: A New Approach for Large-Scale Data Management and Data Analysis
Data Sheet: Aster Data nCluster Analytic Platform
Whitepaper: Deriving Deep Insights from Large Datasets