1010data Updates Hosted Analytics and Reporting Platform
Version 5 adds power, flexibility, richer user interface
- By James E. Powell
- June 18, 2009
1010data has begun shipping Version 5 of the company’s namesake hosted analytics and reporting platform which adds a richer user interface and features that help users more quickly and efficiently solve a wider range of data analysis problems. 1010data Version 5 increases the platform’s lead over SQL-based business intelligence (BI) solutions while providing an intuitive and visual user experience that is typical of desktop rather than browser-based applications.
According to Joel Kaplan, CEO of 1010data, “The new back-end and query features enable effortless ‘surfing’ across many dimensions of data and the types of time-series analyses that are challenging or just are not possible in SQL-based BI systems, while the front-end changes provide a better overall user experience and boost productivity.”
1010data offers a complete approach to performing the deepest analysis and deriving insight directly from raw data, at a fraction of the cost and time of other solutions. Delivered as a cost-effective Software-as-a-Service (SaaS) solution, 1010data makes it possible to deploy a sophisticated BI and data warehousing solution in just a few weeks. It combines a powerful back-end database with a flexible front-end tool, enabling fast and reliable querying of detailed data. The high costs and long lead times that are typical of BI solutions are eliminated because 1010data does not require the installation of software, or lengthy database planning and design cycles. 1010data’s solution gives users immediate insight for better business decisions and rapid ROI.
In addition to significant performance improvements, Version 5 offers the following enhancements:
Omni-dimensional analysis allows for analysis across different dimensions in a single query. The capability was introduced as part of Version 4; Version 5 adds new built-in functions.
Relating time-dependent data between intervals is a common challenge in data analysis. 1010data's ability to efficiently analyze time series has been a distinguishing factor for many years. What would otherwise be a complicated and inefficient process with most systems is trivial in 1010data. At its core, this type of analysis involves exploring data across rows. Importantly, cross-row analysis is not limited to time-series.
Examples of what is new include functions to enumerate unique values of and calculate the frequency of values in a column, functions to allow moving-window sums, averages, and other situations where the window size is a specific amount of time rather than a specific number of rows, an improved interface to least-squares regression, and a variety of new statistics functions.
As-of linking: Connecting data in one table or database with another is another common theme in data analysis. 1010data has always provided exceptional functionality in this regard, allowing tables to be joined on an ad-hoc basis, large numbers of tables to be joined without performance degradation, and tables to be joined using an inexact match. The latter, called as-of linking, is a unique capability that allows analysts to match time-dependent data even when the times don't exactly "line up." In Version 5, 1010data's as-of linking facility has dramatically improved efficiency with fewer restrictions.
Row expansion is new in Version 5; it provides the ability to add rows to a table by replicating existing rows. This feature can be used for advanced analysis including pair-wise comparisons, filling in gaps in time series, and forward modeling (e.g., by creating new rows in a table that correspond to future dates and evaluating a predictive function at those rows).
ODBC driver: 1010data's ODBC driver now transfers data up to 10-15 times faster.
SDK: A new software development kit is available for connecting to 1010data. Native interfaces to the SDK are available for Java, .NET (C++, C#, VB, etc.), Perl, and Mathematica.
Tendo is 1010data’s new application that can automatically run queries and download the results at high speeds. It is similar to 1010batch but has many more options and has a 10-15 times faster data transfer rate.
More information is available at http://www.1010data.com.
James E. Powell is the editorial director of the Business Intelligence Journal and BI This Week newsletter.