Subscribe / Unsubscribe Enewsletters | Login | Register

Pencil Banner

Multi-vendor approach to BI in the cloud

Helena Schwenk | Aug. 24, 2009
Integrating the various back-end and front-end components of a BI stack is no mean feat, regardless of whether it is in the cloud or not.

Four vendors spanning the business intelligence (BI) and data warehousing space have jointly announced a combined solution to offer a BI stack in the cloud. The vendors involved are RightScale, a cloud management platform provider; JasperSoft and Talend, open source BI and data integration vendors; and Vertica, a columnar storage data warehousing provider. By combining their efforts, all vendors have the potential to present a more unified BI cloud solution, covering both front-end analysis and back-end data management capabilities, compared to other more piecemeal cloud initiatives.

Due to the fact that it speeds up and lowers the cost of deployment, BI in the cloud is likely to appeal to SME and departmental deployments where cost constraints and risk aversion are considerations. These vendors will need to resolve challenges resulting from integrating their separate technologies, as well as help customers do business with four separate vendors.

A disposable deployment model that suits short-term projects

Ovum logoProponents of BI in the cloud argue that its pay-as-you-go approach (for instance, customers can pay a monthly subscription or even pay by the hour) gives customers the ability to build and dispose of BI deployments on an as-needed basis. For example, it can benefit power users who need to work on seasonal or one-off projects where the shelf life of the data and BI application may be relatively short. In the past, users would be required to call on IT to source the data, but now they have the option to provision the hardware and software on an ad hoc basis; although this could arguably increase the potential for datamart proliferation.

This hosting model speeds up and lowers the cost of deploying BI, and since there are no residual hardware or software licences to pay once the subscription is terminated, this presents an economically feasible and lower-risk way for users to engage in BI projects. I n particular, the model lends itself to prototyping proof-of-concept projects or to more tactical ad hoc BI deployments that are set up in response to market changes or shorter-term project requirements.

Because BI in the cloud suits faster- and smaller-scale deployments, it is not likely to become a replacement for larger-scale and complex data warehousing projects any time soon. Rather it is more squarely focused on smaller datamart projects, typically the domain of SMEs and departmental-style deployments.

The end-to-end BI stack presents a different approach to BI in the cloud

This approach differs somewhat from previous cloud initiatives as it is one of the first to combine both back-end data integration and infrastructure capabilities alongside front-end reporting and analysis tools in the cloud; most other initiatives have focused on data warehousing services in the cloud. In addition, this solution mixes both proprietary and open source BI components in one cloud solution.

 

1  2  Next Page 

Sign up for Computerworld eNewsletters.