Not All Analytics Fit for the Cloud
January 25, 2012 – Not all analytics systems are a great fit for cloud computing, though there are a handful of promising deployments surrounding big data, customization and innovative applications, according to a new report from MWD Advisors.
The report from the U.K. research and analysis firm, entitled, “Analytics in the Cloud: A Practical Approach,” explores the complexities and capabilities of analytics in public, private and hybrid deployments.
Among MWD’s half-dozen common analytic adoption scenarios in the report is one, like the cloud, receiving quite a bit of attention lately: big data. Public or hybrid deployments can boost analytics for big data with newfound capacity for high-performance capabilities such as preparing data for data mining, scoring data and producing data models. In addition, off-premise environments may be better built to handle unstructured data from Facebook posts or log files and enable more flexible and powerful analytics processing. MWD reports that these analytics capabilities are gaining credibility among end-user organizations.
In another instance, customized analytics applications, MWD points to a platform as a service offerings as uniquely positioned for designing and developing customized analytics applications. Taking on a PaaS, the cloud provider is in charge of tasks like on-demand provisioning and the maintenance of software and hardware.
The public or private cloud can also provide advantages for prototyping or running proof-of-concept projects for new analytics technologies. Here, the cloud works as the evaluation and benchmarking environment without having to find and customize hardware to install test configurations, MWD reported.
SMBs, long a target market for widespread cloud adoption, may find distinct new advantages in deploying certain analytics capabilities. As many SMBs lack the IT infrastructure for an in-house cloud, the public cloud model offers lower upfront cost barriers for analytics or BI and faster access to analysis options, MWD reports.
Regardless of organization size, certain departments or operations more naturally lend themselves to the application of analytics and BI in the cloud, says Helena Schwenk, principal analyst for analytics and information management at MWD and author of the report.
“Sales and marketing is a prime example, as companies look to use the cloud to analyze customer data, gain visibility into the sales pipeline or understand the impact of social media on marketing initiatives. Similarly, customer service, operations and supply chain also make good use cases,” Schwenk says.
Other prime use cases outlined by MWD include:
- Scaling out enterprise BI applications, with a focus on the option to use a hybrid cloud to offload some analytic processing or to move a downstream data mart.
- Short-lived ad hoc analysis, as a way to respond quickly – but only on a temporary basis – to new business conditions like an integration of an acquisition.
To access the report, click here. Information Management readers can receive a 25 percent discount by entering the following code at checkout: IM0112