Ever Since E.F. Codd  Coined the term  OLAP (Online Analytical Processing) it has gained high popularity in enterprise computing, in simple terms  OLAP  is  synonymous  with concepts, tools that make Data Ware House data easily accessible.

OLAP enables analysts, managers, and executives gain insight into data through fast, consistent, interactive access to a wide variety of possible views of information.  They also answer the Who ? and What ? and What if ? questions on the data.

The key attributes of the OLAP are

• Multidimensional views of data
• Calculation intensive capabilities
• Time Intelligence

The multi dimensional nature of  OLAP requires an analytical engine to process the underlying data and create a multi dimensional view and  the success of OLAP has resulted in a large  number of vendors  offering OLAP servers using different architectures.

MLOAP : A Proprietary multidimensional database with the  aim on performance.

ROLAP : Relational OLAP is a technology that provides sophisticated multidimensional analysis that is performed on open relational databases.  ROLAP can scale to  large data sets in the terabytes of  range.

HOLAP : Hybrid OLAP is an attempt to combine some of the features of MOLAP and ROLAP technology.

Is There a COLAP  (Cloud OLAP) ?

Most business cases for CLOUD is about,  dynamic infrastructure, multi tenancy, elasticity and other attributes that target  reducing  cost of computing by reduced capital expenditure and operational expenditure.

While OLAP  on the other hand is viewed  for the needs on processing large amounts of data (usually in TERRA BYTES ) in a  most efficient manner from time perspective, and performance is a key distinction here.

So how do they both  meet and  a OLAP application is indeed a good candidate for Cloud. In other words can a OLAP Server  be hosted on Cloud Platform, either by a SaaS Provider or through a PaaS.

In OLAP a CUBE is a logical construct. It allows a client application to retrieve values as if every possible summarized value existed in the cube.  Like a fact table,  a cube contains one columns for each dimension and one column for each measure.

An example of a CUBE in a OLAP Server that stores  Sales information is visualized here.

Important considerations for a efficient OLAP Server are :

• Storage Optimization : How efficiently the map, detail, aggregate components of a CUBE are stored
• Processing Optimization: How efficiently a Server processes a CUBE with respect to changing dimensions and fact.
• CUBE Drilling Optimization: How efficiently the CUBE information can be queried by multiple clients.

Let us see whether a CLOUD architecture is fit for OLAP implementation.

Remember A CLOUD Is a GRID Too !!
Grid computing is a type of parallel and distributed  system that enables the sharing, selection, and aggregation of geographically distributed "autonomous" resources  dynamically at run time depending on the work load  and availability requirements.

Cloud computing evolves from grid computing and provides on-demand resource provisioning. Grid computing may or may not be in the cloud depending on what type of users are using it.

In other words, we could implement Grid Computing  in a cloud computing environment and it is a valid generalization to say  that "every cloud is a grid".

Grid Computing  Attribute of CLOUD In Processing OLAP CUBES
With appropriate support from  Clod Computing Platform and with work load management, a Cloud platform can be effectively  use it's  incarnation as a Grid to effectively process the OLAP CUBES and other  CPU intensive Analytical processing  steps to be positioned as a effective OLAP Platform.

These concepts are new,  leading OLAP enabled databases  like DB2  have long been utilizing features like

• Inter-Partition Parallelism , which is each Virtual Machine retrieve and process the requested rows they own in parallel
• Intra-Partition Parallelism , where by within a Virtual machine depending on the CPU availability certain operations can be parallelized further.

So if properly configured if there are N Virtual Server s are configured for this Service and each virtual server can perform M degrees of Parallelism at the end of the Cloud Platform could perform the Analytical operations with N X M parallel tasks and the other tenants of Cloud like VM Migration and work load migration all play towards a efficient and successful OLAP platform.

As Evident, Cloud Platform is very effective for OLAP Servers too provided the implementation support the GRID qualities of CLOUD in effectively utilizing the multiple physical and virtual servers of CLOUD to achieve the goal.