Have you ever stepped back, looked at your current IT infrastructure, and thought, "No human in their right mind would ever design a datacenter like this!"? If you’re like most IT professionals, you’ve either inherited a jumble of technology patched together over the years, or you’ve had to throw together quick solutions to meet business demands. Most of us don’t have the luxury of building a datacenter from scratch—and even if we did, we’d likely end up adding features, functionality, hardware, and software on as-needed basis. It’s nearly impossible to find the time for planning when you’re just trying to keep up with day-to-day needs.  

Datacenters are usually built in a reactive manner--Oh! I need more capacity. Oh! That server is down and I have to get it up. You need what capability by when? Are you out of your mind?! It will take months just to figure out how to do it.  

The reason why IT infrastructure gets so messy is that IT and business are not in sync. IT enables and empowers the business. But business is focused on earning more revenue, creating new product lines, and finding opportunities to generate new cash. Business managers often consider IT as a support department that is responsible for capital expenditure.   

However, you can change this situation. When IT and business have a common way of looking at IT infrastructure and can see how technology expenses can be lowered, then IT can be transformed from a cost center to business enabler. By supporting business proactively and in a reliable manner, IT can have a direct impact on business success. This transformation comes with standardization in the delivery of datacenter services through end-to-end management of physical and virtual environments. You can reduce operational costs and get optimum utilization of data center resources by considering your entire IT platform in every decision you make.  

The Platform Vision architectural model provides that common perspective on IT as a whole. The Infrastructure solution pattern represents the foundation for any IT platform and encompasses datacenter management, client devices, identity, networking, and security. By understanding the elements that make up the Infrastructure solution pattern, and by considering the dependencies and relationships that exist throughout the IT platform, you can give business management a way of seeing the cost and Return on Investment (ROI) implications of introducing any technology into your infrastructure. And you can look at your datacenter and see it as something that is evolving and growing in a sensible, planned way.  

This quick overview of the rational for and benefits of the Platform Vision model is just a starting point. To provide an in-depth explanation of the model, describe the components of each layer, and explain the dependencies and relationships among the components and layers, the Platform Vision team will create a white paper drilling down into each layer of the model. Stay tuned for the first of these white papers, which will focus on the Infrastructure layer.  

We invite your input into the model. Help us grow the Platform Vision model and make it more useful. Let us know about your experiences, as well as your needs. Email at Vikram.Jain@Advaiya.com

After years in IT, working in various areas of software development, application deployment, and infrastructure and security solutions architecture, I think that security is the most important platform investment for any organization. I think you should invest in security before investing in any of the other Platform Vision solution patterns because if your environment is not secure, it is vulnerable to failure at every level. From an enterprise perspective, when you think about security, it covers everything – identity management, protection of intellectual assets, application and network security, and physical security. Security means protection of organizational assets in all ways. And security is implemented with a combination of technology, processes, and training.

Numerous technologies are available for implementing security at various levels. Some of the forms these solutions can take include hardware and software solutions like firewalls, intrusion detection systems, anti-malware solutions, identity management, authentication mechanisms, encryption, gateways, and VPNs. Also, supplemental technologies are available to protect collaboration, integrated communications, business applications, and development of secure applications from the ground up.

One important aspect of implementing security technology is that the implementation should be governed by the use of proper IT policies and procedures with operational controls. Creating IT policies and procedures is the first and most critical component to any security and management program.  The benefits of this approach include the following:

·         Creating a baseline from which to operate

·         Communicating management’s intent

·         Describing acceptable uses of various systems, expectations from users

·         Providing best Practices to base the solution on

·         Establishing a framework for business continuity and disaster recovery

So to establish a roadmap for implementing security in an organization, you need to:

·         Create an IT strategy

·         Perform a comprehensive risk assessment

·         Establish a control framework

·         Determine IT policy

·         Implement your security solution

·         Set up programs for user awareness and training

·         Perform regular audits

Implementation of security with proper IT policy and user training is the key element to creating a solid foundation for all other solution patterns. Security is also key to enabling all of the Platform Vision solution patterns: user experience, Business Intelligence (BI), data management, collaboration and ECM, data management, SOA, next-generation Web solutions, and cloud solutions. A good security strategy and implementation allow you to get the most business value for your organization. When all business functions are surrounded by a security layer, the business can focus more on core business functions rather worry about security issues.

But no matter how much you focus on security, maintaining a secure environment is always an ongoing process. I still remember a quote from a security expert: “Security cannot be measured in absolute terms – either you are less secure or more secure; there is no term like ‘most secure.’” 

When designing and estimating the cost of a data center, the business needs, and IT, people generally overlook factors that affects Total Cost of Ownership (TCO) for the data center in the long run. While doing Business Value Assessments (BVAs), which include TCO studies for large enterprises, my colleagues at Advaiya and I have found that there are no recognized standards for calculating the TCO of a data center. To address the need for guidance on TCO, we've added up various capital and operating expenses and have done extensive analysis of data center costs. Our study shows that a typical data center costs  $6000 -$6500 per server per year. This cost includes many factors, such as:

·         Building facility – floor space leased or rented

·         Power consumption

o   Raw

o   Standby – UPS and generator

·         Lighting

·         HVAC

·         Raised floor

·         Fire suppression

·         Security costs

o   Digital threats

o   Physical threats

·         Administrative costs

o   Administrators and technical staff

o   Project management

We also found that to improve data center TCO over a longer span of time, you should plan a practical approach for designing a data center so that from the design itself, it is scalable enough to support additional racks and servers and power and HVAC equipment. The most important benefit of practical sizing over traditional sizing is that your organization can save on increasing costs for power (electricity and cooling) over the next 5 years , as power is the largest cost factor in the data center and takes a 30-40 percent chunk of operational costs.

Thus an ideal datacenter would have the power and cooling infrastructure needed at the present moment but that is scalable for future requirements. We recommend that you consider such a practical approach for creating a data center that is scalable and modular to obtain significant cost savings.

In my next blog, I'll discuss how you can further reduce operating costs and the management complexities of your data center by using virtualization and other green IT tactics.