BI on Cloud – Key Benefits and Challenges

There has been a rise in the number of enterprises that have started implementing Business Intelligence (BI) on Cloud platform instead of on- premise BI. Is it offering better benefits compared to traditional BI models? Let’s explore this model in detail.

What is BI on cloud?
Hosting an enterprise Data warehouse Database, ETL & Reporting tools on the enterprise cloud server or using an existing cloud infrastructure as a service (IaaS) like Amazon Web Services (AWS) for BI.

NIST defines Cloud Computings as “a model for enabling convenient, on-demand network access to a shared pool of configurable computing resources (e.g., networks, servers, storage, applications, and services) that can be rapidly provisioned and released with minimal management effort or service provider interaction”.

For example for our existing client, we have started designing the DWH and Vendor evaluation of ETL & Reporting tools. Then we looked upon the time line for servers procurement, setup, installation & Database license cost.
We realized that it’s not advisable to wait for such a long time and spend huge amount for initial setup. Hence we decided to go ahead with AWS for our entire BI (DWH, ETL & Reporting) to host on the cloud.

By hosting BI onto cloud enterprises can save money, time & effort. Below are the drivers for moving to a cloud hosted BI solution:
• Reduce the Hardware/Software licensing & Maintenance costs.
• Reduce the time spent on procuring the hardware/software, setup & installation.
• Decommission most part of in-house developed ETL/Reporting code to avoid person dependencies.
• Increase the availability of business intelligence platform for various business analytical needs through internet.
• Improve the performance by increasing the Compute nodes as and when needed.

Service Models
Software as a Service (SaaS): Companies host applications in the cloud that many customers access through Internet. The services offered are typically complete end-user applications. Some examples are
Zuora, IBM Cloud, Google Docs, Facebook, salesforce.com and Oracle on demand

Platform as a Service (PaaS): The ability to provide a computing environment and related development and deployment stack needed to deliver a solution to the customer. Some examples are
Birst, Snap Logic, IBM Cloud, Azure Services Platform, Oracle SaaS platform and force.com

Infrastructure as a Service (IaaS): Infrastructure, including a management console and associated software, provided to companies from the cloud as a service. Customers/Users buy these resources as services on “as-and-when” needed basis. Some examples are Amazon Web Services (AWS), IBM Cloud, Windows Azure, GoGrid and Google Base

Some Key Characteristics of cloud BI
On-demand self-service: As the name describes, we can expand the storage or service at a click of the button without any human help.

Information over the net:Information is available over the network and can be accessed anytime through the net by different modes like laptop, moble, ipads etc.

Resource pooling: Provider resources are grouped and used efficiently by multi-tenant model. Resources include storage, memory, VMs etc.

Rapid elasticity: Resources (both hardware & software) can be increased or decreased efficiently and effectively in quick span of time. Customers can purchase the resources for any quantity and at any time.

Cost effective: Resource usage can be monitored and would be charged on the basis of usage. This system is very transparent which makes the provider and the user more comfortable to adopt it.

Deployment Models of Cloud BI

Private cloud: A set of computing resources generally within a company that serves its internal staff but is set up to operate in a cloud like manner in terms of its management.

Community cloud: A set of computing resources shared by several organizations and supports a specific community that has shared needs. It may be managed collectively by the organizations or a third party and may exist on premise or off premise.

Public cloud: A resource available to any consumer either for a fee per transaction service or as a free service.
Some examples are Amazon Elastic Compute Cloud (EC2), IBM’s Blue Cloud, Sun Cloud and Windows Azure

Hybrid cloud: A computing environment that includes the use of public and private clouds often where there are one or more touch points

Key Benefits of Cloud BI

Adopting a cloud based solution for BI space can provide tremendous benefits for any organization.

First and foremost benefit is: IT Departments can effectively concentrate on other business development solutions than on BI Systems maintenance activities. It can be an effective SLA based outsourcing of IT systems to Third-party vendors, who provide cloud services. Enterprises do not need to worry too much about maintenance, up-gradation, monitoring etc.
Scalability: Cloud platform is designed to automatically provision the resources based on the workloads. So the user doesn’t have to worry if data loads starts to increase, cloud automatically provide for storage as well as increase its processing capability by increasing compute nodes as on when needed.
Savings: When cloud vendors are used to host for DWH/ETL/Reporting infrastructure there is no need to worry about capital expenditure involved for setting up servers and admin resources as all this is taken care by cloud vendor, and the user will only pay as per usage. This really brings down the cost of having a BI solution. Organizations can also leverage upcoming technologies at lower costs and can experience a hassle free maintenance and upgrade.

Speed: It’s been observed that data retrieval speed of the cloud based database is many times better than most RDBMS. This is because of the way the data is organized in the cloud environment. In AWS data is stored in column wise rather than rows wise. This different design and architecture enables cloud vendors to maintain very fast retrieval rate from tables with billions of rows.

Customer Experience: Since customers can access the applications through internet without any hassles it improves the customer experience. No need to install any add-ons/plug-ins & VPNs while accessing via internet.

Challenges and Concerns of Cloud BI

Below are some challenges and concerns that are still in the minds of oraganizations when they think of BI on cloud platform.

1. Data security: Enterprise key data will be hosted on cloud there is concern of the security of the data. Vendors are coming up with standard mechanism to avoid data security issues and providing proper disaster recovery mechanism the customers data.
2. Governance Issue: When enterprise data is residing on 3rd party system the question of how access to data is restricted comes up. Service providers generally have access structure based on the user credentials and also on IP address level. So it Controls who accesses, what data and from where.

3. Data Integration: Cloud technology is relatively new and there is possibility of some existing solutions like ETL/ Reporting tools may not be able to connect /integrate/access with data on the cloud. Now a days most cloud vendors are supporting existing softwares if required by providing add-ons/configuration files.

Conclusion

To summarize, BI on cloud provides IT capabilities to enterprises through the Internet, which is highly scalable solution. Also keeping analytics in the Cloud provides advantages like Economy, Scalable/elasticity for lots of data, Flexible for experimentation.