Databases Reference
In-Depth Information
how much data needs to be moved to the cloud. Apart from the data, the operational cost of this solu-
tion is hundreds of dollars less expensive compared to the traditional solution and it delivers value
on demand. Scalability can be elastic in this situation and users can pay as the usage increases or
decreases on a monthly or weekly basis.
Applications that can utilize the PaaS option include social media analytics and text mining
analytics.
Software as a service
The most popular option from a cloud perspective adopted in business intelligence is the software as
a service (SaaS) platform. This is application-driven and less intrusive on data, which makes it easier
to adopt. There are a variety of vendors providing an SaaS option today.
For the data warehouse deployment on the cloud we need platforms that are:
●
Scalable
. Elastic scalability is the primary need when we discuss data warehousing on the cloud.
The data volume can dynamically shift on-the-fly.
●
Secure
. Data privacy issues often keep adopters of the technology worried about the security
of data. As a general rule, public or hybrid clouds should never encourage any data that has
compliance requirements to be loaded to the cloud.
●
Available
. Data availability around the clock is a user expectation.
●
Performing
. System performance and query performance are desired characteristics, though some
providers may need to set up and charge extra for those SLAs.
If we can support these four pieces of a 2 × 2 quadrant, the primary requirements of the data ware-
house users are resolved. The next question that comes up is what is the infrastructure components
that make up my cloud.
Cloud infrastructure
A typical cloud infrastructure consists of the components shown in
Figure 9.3
.
●
Applicationscan be classified into:
●
Services—Paypal or Google Maps falls into this category, where the application is a complete
service.
●
Business—Salesforce.com and Microsoft Office 365 enable businesses to run their entire
requirement on a cloud platform, eliminating the need for onsite software and infrastructure.
●
Collaboration—portals and forums are examples of a collaboration suite.
●
Mobile applications—another emerging cloud-based application footprint is the concept of
corporate application stores.
●
Services:
●
Include the different management tools and application suites provisioned to manage the cloud.
●
Software API layers for programming and deployment.
●
Storage—includes multi tiered storage architectures including all types of disks and technologies
●
Hardware—servers and network infrastructure
●
Virtualization/operating system tools
