Ovum research predicts that companies will spend over $9.1bn on Business Intelligence (BI) by 2014 which represents a 6% of Compound Annual Growth Rate (CAGR) over the period. Underlining this growth are significant changes to the way in which BI systems are built and deployed. Till now, only large enterprises could afford to build sustainable enterprise analytic capabilities due to Capital expenditures (CapEx) involved in BI systems. Hence gaining competitive edge through BI has been a forbidden fruit for small and medium businesses (SMBs). Cloud computing is a disruptive trend in recent times which uses pay-per-use billing model and brings various characteristics like scalability, elasticity, emphasis in Operating expense (OpEx) (rather than CapEx model) onto the plate. Cloud computing will be the best solution to BI for SMBs.
Obstacles to BI for Small and medium businesses:
- Cost of BI technology: Supplementary overheads on IT infrastructure leads to more financial commitment for driving business.
- Scalable BI hardware and software: Scalability of Enterprise ETL (Extraction, transformation and loading), database, reporting and analytics hardware and software have been a barrier to adoption of BI.
- Proficiency in People, Processes and Technologies: BI solutions requires specialization in technologies like ETL and data integration, parallel processing, data modelling, statistical analysis, and data mining. Lack of skilled staff or sub-optimal staff is a barrier.
- Efforts: Effort required in consolidating and cleaning enterprise data.
- Easy to recognize, difficult to upgrade: It is not easy for enterprises to upgrade their enterprise systems is a way that does not disrupt their current processes.
These obstacles result in either scarcity of resources or wasteful under-utilization of investments.
Enter the Cloud
Service-based architectures in the cloud SaaS, PaaS, IaaS will drive BI strategy by default. With the advent of Cloud computing, BI will become affordable and available to SMBs.
Cloud computing can be a key enabler for BI for the following reasons:
- Data is moving into the cloud; mainly, with the burst of Cloud Storage options and RDBMS availability in the cloud.
- Analytical and Reporting applications are increasingly required to be used as service to meet the varied demands.
- BI technology needs to be experienced in enterprises before full acceptance and for this sandbox testing cloud is the best choice.
- Data and Computing resources can be planned in incremental basis in cloud and paid for usage.
Business Intelligence as a Service (BaaS) can be a focal and evolving term. In a Cloud environment, business and IT users can access a set of business intelligence services over the internet. Mobile and Tablets can be used as a client as well.
BaaS Security Concerns
Data security becomes more important when using cloud computing at all service levels. Confidential data hosted in a cloud must be protected using a combination of access control, contractual liability and encryption. Compliance, security, consistency, synchronization, bandwidth and latency are Data related issues in Cloud. In case of BI solution on cloud, if information is very important and confidential to the business then hybrid cloud model would be a better implementation model. Confidential information can reside on private cloud and BI application can access it for further analysis. Virtual Private Cloud will be a good candidate for processing information in public cloud.
Final Thoughts
Resource provisioning in split second, Scalability, Elasticity and Pay as you Go billing model of Cloud computing have made the life of SMBs as well as Large organizations easier. Cloud computing is the space where BI can be considered as a feasible use-case which used to be a forbidden fruit for SMBs.
Reference:
http://www.infoworld.com/d/cloud-computing/bi-steps-cloud-085
http://www.thebusinessintelligenceguide.com/bi_solutions/Cloud_Computing_BI.php
http://media.techtarget.com/searchSMB/downloads/LoshinBusIntelchap2.pdf
http://www.information-management.com/issues/20030601/6800-1.html
*If you find something is misleading or not correct then please throw some light on it.
