MDM in Cloud Computing
What is Cloud Computing?
National Institute Of Standards and Technology (NIST) defines it as "Cloud computing is 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." There are service providers who provide services like Infrastructure (IaaS), Platform (PaaS), Software as a Service (SaaS) and others and there are service consumers who use these services to achieve a business need. The service provider and the service consumer are usually bound by a common SLA and the consumer only pays for whatever and how much of the service they have consumed.
Energy Utilities and Smart Grid
Today electric utilities across the world are dealing with key issues that are becoming increasingly critical. Physical infrastructures in the electric power grids are aging and will require extensive replacement, and those costs are expected to outstrip historical capital spending rates. Then there are environment considerations like reduction in greenhouse gas (GHG) emissions and movement towards Renewable Portfolio Standards (RPS). Customers are also demanding higher levels of service including more visibility and control on how they are using power so that they can influence usage pattern and cost incurred.
We find that utilities' IT teams are struggling with these delivery problems today:
- High dependency on third party systems.
- Data complexity and variability that has an impact on integration and testing efforts.
- Need for data storage has grown manifold as with smart meters in place utilities will be storing thousands of readings each month instead of one reading for an entire month for each meter.
So in the world of smart grid utilities need to periodically upgrade their data centers and applications to access and analyze that data. Effective use of Smart Grid will benefit both the utilities and the end customer. For utilities it will bring in optimization in grid usage, better alignment of demand with supply constraint and grid congestion, reduce their carbon footprint and bring in green energy tax benefits. For end consumers it will provide more control and transparency, they will be able to control their consumption patterns and consume energy much more efficiently and intelligently than what is possible today.
Cloud computing in smart grids
In this section we will explore how cloud computing solutions can provide technology capabilities for some smart grid scenarios:
Data storage on the cloud: Utility companies today collect only one reading for an entire month from electric meters installed in the premises of end users. However with smart meters, utilities will have to monitor and report end user consumption details much more frequently, often collecting data from multiple smart home appliances on a second-to-second basis. This is the extreme requirement of utilities to transform their business in such environment and enhance their meter data interoperability system. Thus the volume of data that must be handled will be in orders of magnitude much beyond the current capacity of utilities. They simply do not have the proper infrastructure in place now to handle such large data volume and they will have to scale it manifolds' in the future.
Cloud computing creates the platform for delivering data, storage, communications, applications and services, at scale, through an as-needed, pay-as-you-go business model, For this utilities can leverage on-demand data storage infrastructure from cloud providers. This will not only limit the initial expenditure but will also allow them to scale up as the adoption of Smart Grid increases. Moreover cloud providers will take care of associated functional and non functional requirements like SLA assurance, data security and availability, disaster recovery, physical security of the data centers. Smart grid and cloud computing need to go hand in hand in order for the customer focused aspects of smart grid to be successful. Cloud computing offers great opportunities for both utilities and customers. The objective of cloud computing is to make full use of the meter data to empower utility and their consumers to make smarter decisions and truly enable demand management, demand response/load control capability.
Key benefits of cloud based on-demand data storage solution:
- The data can be access anytime from anywhere over the internet with valid credentials.
- Reduced OPEX; no cost incurred for hardware/software maintenance, administration, power, real estate etc.
- Reduced CAPEX; utilities will not have to procure additional infrastructure.
- Operational efficiency from consistent infrastructure.
- Utilities only pay for the amount of storage that they use.
- By shifting to cloud computing utilities can have meet the continuously increasing demand for storage than upgrading dedicated data centers periodically.
- Pay-as-you-go business model can result in reduced TCO (total cost of ownership).
Predictive analytics: Turning huge volume of usage data that the smart meters and sensors captures into knowledge and acting upon it in real-time will provide the actual benefits that smart grid promises. For this utilities will have to use predictive analytics which involves building statistical models that learn patterns hidden in vast amounts of data and use the same to detect similar patterns as they occur, enabling the grid to predict whenever non-optimal or unwanted conditions are about to occur. These predictive models are thus able to apply their knowledge of the grid to improve its day to day performance. This however requires a lot of computation capabilities.
For example, various functionalities of MDM including VEE, reporting, auditing, etc. can be designed to harness the computing within the cloud. Hence the utilities can have the freedom to go for an entirely cloud based MDM product or a MDM product using only the storage service of the cloud. Similarly reporting can use the existing BI capabilities of the cloud environment.
Key benefits of cloud based predictive analytic solutions:
- Use high performance computation systems to perform high compute intensive jobs (e.g. Hadoop).
- On Demand storage and compute power only when required.
- Cost effectiveness.
Cloud computing will enable smaller utilities to go for advanced AMI technologies like MDM as the high license and hardware cost will not be a deterrent anymore. This move will also benefit the MDM software vendor as they will be able to reach out to many more customers. The system integrators can have a resource pool catering to cloud service providers, MDM vendors as well as the utilities. This creates a win-win situation for all the parties involved.
User friendly data access through pervasive devices: In order for end customers to take informed decision regarding their power consumption, they need to have access to data in real time in a user friendly manner. Users need to have right to choose on delivery interface, format and time based on their convenience.
Applications and data hosted on a cloud can be access through pervasive devices like smart phones, PDA's and other hand held devices. These devices have the capability to capture and stream data from remote locations and using either satellite or telecommunication network facilities to contact applications hosted on the cloud.
Another possibility before utilities is to provide application programming interface (API) that provides access to meter data with customer's permission. This can open up an ecosystem of consumer side cloud based services and application which can provide different services to the end users.
- Any time any where access.
- Users get to choose from multiple devices and multiple formats.
MDM in Cloud: Proposed Architecture
The promise of cloud computing with its elasticity and the low cost availability of such compute, storage, ability to connect to pervasive devices makes it an ideal candidate for many industry segments. Smart Grid initiatives undertaken by utilities to modernize aging electric grids are one such segment which can benefit from cloud computing model. Pay per use model in cloud computing will be a big boon to utilities which otherwise have to invest enormous amount to procure infrastructure required to store and process large data set in smart grids. Besides high compute jobs required for predictive data analysis and providing a user friendly interface to end customers are other potent areas where cloud computing can be used.