01 Informatique's article gives a good factual overview of Bime and explains some of the innovation that has made user-friendly BI possible. They talk of Bime's "attractive user interface and simple usability" before listing some of Bime's main features and explaining some of the technological tricks we use to improve calculation speed. Of course there are thing's we would like to add, such as our full range of connections and a comment on the more advanced visualizations, but the article serves as a good introduction to what Bime does.
The original article is only available in French but you can read the translation below.
"Business Intelligence in the clouds for all." This is the promise made by the start-up We Are Cloud through their online tool Bime, with its attractive user interface and simple usability. Before founding their start-up, Nicolas Raspal and Rachel Delacour worked in BI and shared the same vision: that it was too complicated, too expensive and too difficult to set up.
Innovation in SaaS (Software-as-a-Service) and Cloud Computing presented them with a solution to these traditional problems. "SaaS BI will interest companies for whom the software giants' BI suites are too expensive, especially SMBs" explains the CTO, Nicolas Raspal.
15 Visualization Types The online service fulfils their objectives of simplification. Once the datasource has been specified by the user, the scheme of analysis in the data is automatically detected by the application. Next, dimensions (e.g. temporal and geographic) and measures (e.g. turnover and profits) are dragged and dropped on to the axes of the results sheet. There are 15 visualization types on offer, for example PivotTables, Graphs and Piecharts. The processed data is retrieved from on-premises sources such as Excel files or online data such as Google spreadsheets and Salesforce.com.
Users can share a dashboard, even if only one of them has access to the datasource. The service is only delivered as a SaaS application, but there is a desktop application to install which automatically refreshes your data each morning.
Using Tricks to Speed Things Up Raw data can be uploaded to the online datastore, Amazon S3. The caching mechanism that they use, Déja vu, stores and encrypts the datafiles online to facilitate their recovery. The data is then stored in aggregated form: the system presupposes plausible requests and pre-calculates them, to present responses to the user much more rapidly. For example, the sales per store will have been calculated before the user requests it. The start up uses a calculation technique called in-memory processing and has data organised by column rather than row. For larger calculations they use other techniques such as a Hadoop framework capable of distributing processing requests across multiple servers.
Marie Jung, 01informatique, 04/15/2010