December 29, 2011 | Written by: Biren Gandhi
Share this post:
Note: Through the end of the year, we’ll be posting one blog per day from our top 10 “greatest hits” from Thoughts on Cloud since we launched in September. This post is #9 and was originally published on Oct. 20.
Our modern information environment is more complex than ever. Data volumes are burgeoning and coming in from all mediums including sensors, blogs, podcasts, wikis, and tweets. The pace of everything is accelerated. Information must be analyzed, contextualized, and shaped for decision-making and right-timed action. At the same time, globalization demands better sharing of information not only with colleagues down the hall but also with those around the world.
Each of these people requires multiple levels of information and detail to make decisions that impact performance. Using highly visual scorecards, dashboards, reports, and real-time activity monitoring, decision-makers gain immediate insights regarding the health of the business and can understand what is happening in their area of the business. Analyzing trends, statistics, correlation and context, decision-makers can understand what leads to the best outcomes and discover why things are on or off track.
Knowing what is likely to happen equips decision-makers with the foresight that they need to intervene. Simulation through predictive modeling and what-if analysis enables decision-makers to predict and act and change the course to improve the outcomes. Financial and operational planning, budgeting, and forecasting deploy resources in the right place and sets targets for those allocations.
Yes, business intelligence and analytics are not new; quantitative and analytics methods have been in existence in specialized forms such as financial markets trading, for some time. What changed today is the maturity of the technology and tools that provide the business intelligence and analytics services.
Business analytics (BA) today is about creating value at the intersection of multiple business domains and their underlying application landscapes. It’s about business optimization to drive a more sustainable competitive advantage in the marketplace while reducing costs. This means moving from just using ERP and financial applications to providing increased financial risk insight for better business decisions, moving from just managing your supply chain to enabling more dynamic demand planning, and moving from just managing your call center and customer relationships to providing increased insight to improve customer service to drive greater profitability from your customers. It means moving from application agenda to information agenda.
This migration of focus to information creates an optimal opportunity to share the application landscape within the organization, because BA is about consuming data and information locked in specialized applications or ERP systems. Current business intelligence (BI) tools such as IBM Cognos BI Platform and advanced analytics tools such as SPSS can process and present data from a variety of sources efficiently and effectively. But, the greatest value is generated when these initiatives are scoped enterprise-wide and are coupled with cloud to make insights and alerts easily accessible across the entire organization and not just dedicated user groups. The cloud-based deployment of these powerful tools unleashes industrialization or commoditization of the BA inside the walls of the organization and across the value-chain.
Picture an organization where pertinent information, perhaps customer or inventory data, is shared effectively and in a timely fashion. Individual employees, instead of each of them storing similar information in their own spreadsheets, can generate customized reports using a common service. Through that and similar scenarios, businesses can take advantage of business intelligence and analytics methodologies to improve information sharing and bring fact-based analytics to mainstream business and, in a way, commoditize the business analytics across the organization.
Transformation to cloud computing changes the economics of business intelligence and analytics. Cloud computing is both a business delivery model and an infrastructure management methodology. The business delivery model provides you with a standard offering of services, such as business analytics that are easily accessed and rapidly provisioned. Steps, such as producing hardware, installing middleware and software, and provisioning networks, are dramatically simplified. The infrastructure management methodology is built on virtualized resources and provides better economics and increased ability to scale. It makes high-volume, low-cost analytics possible.
But, not all clouds are created equal. Important attributes, such as location, ownership, access, targeted users and workload (application types), varies across an array of clouds. The following figure shows a variety of cloud models to meet unique needs and priorities.
Key reasons for the cloud to be the catalyst for the business analytics are as follows:
- Rapid service provisioning enables a variety of new analytic data management projects and business possibilities. Innovations and new technologies can be introduced in less time.
- Can take advantage of large number crunching capacity on pay-per-use model. One can rent a thousand servers from external or internal cloud service provider only for the number crunching. This is particularly true for real-time data crunching and creating insights from tones of data from devices such as traffic signals, electricity grids, health-care devices, an elastic model of being able to consume processing power and memory makes great case for the commoditization of the business analytics.
- Offers consolidated business intelligence and software product sets. By deploying the business intelligence and software products on the cloud, one can bargain a good deal with software vendors, consulting firms and service providers.
- Offers expanded amounts of data sources that a single user can tap into. As I mentioned previously, the real advantage of business analytics is creating key performance indicators and generating insights from the intersection of data from different lines of business. For example, to implement an effective predictive maintenance strategy in the process industry requires data from varied real-time devices that are deployed in the industrial plant, maintenance schedules of the vendors, OEM warranty information, spare parts information, historical data of the plant performance, and maintenance logs. Typically one can imagine this data is owned and maintained by various lines of business; by being able to create a data mart on the cloud or even better being able to access all of these sources dynamically as needed, the user is able to access the data from multiple sources from a single system. Of course, the cloud strategy also helps to consume the processing power according to the need.
- Is able to serve over a large number of users. By deploying the BI applications and analytics tools on the cloud, one is able to provide the same applications and tools to a larger audience. Personalized role-based dashboards and ability to create reports, score cards on the same data for varied business roles is available. Because all of these capabilities are hosted on the cloud the number of consumers of this data in different presentation modes can be very large, it could be as large as 200,000 users within and out side the organizational boundaries.
- Offers cost savings from hardware, software, and operational efficiencies. By clubbing multiple BA initiatives in the organization to a larger enterprise-wide initiative, one can create great savings from using common hardware, software, and services for the implementation. Operational efficiencies creates even more cost savings, because the data sources, analytics capabilities will be consumed as a service by multiple lines of business, thereby truly commoditizing the BA capabilities.
- Improves standardization by introducing single points of control for departmental business processes, corporate security, and compliance. Figure below depicts how the concepts of the SOA Security and Secure Virtualized Runtime supports maintenance of the identity, isolation of data and business process components and compliance to platform, application and business process regulations.
- Reduces the capital and operating expenses that are needed to support enterprise-wide services.
Is the promise of virtualization and cloud real? Good news is YES! There are lots of real and successful instances of BA’s deployment on the cloud, one of them being IBM’s internal deployment with around 200,000 users.
IBM also offers complete end-to-end deployment of such BA solutions on the cloud with its offering IBM Smart Analytics System. The IBM Smart Analytics System provides the ideal delivery vehicle for rapid deployment of BA capabilities and accelerates delivery of new analytic innovations including those from IBM Research.
The following figure depicts core components of the IBM Smart Analytics System.
An IBM Redbooks publication from the ITSO organization is also an excellent resource; it covers a wide range of information starting from key business case elements, various forms of architectural artifacts (such as functional architecture, operational architecture, and cloud management architecture), detailed installation and application implementation steps, and running the cloud (activities such as service life cycle, on-boarding, provisioning, monitoring, metering, and billing).
Yes, cloud is the catalyst for commoditization of the business analytics and is probably the only way one can economically cater the computing needs of the real-time data analysis, particularly around smarter cities, smarter energy, smarter healthcare and others.