Data Center Managers Grappling with Issues

by Tabrez Khan    Jan 30, 2009

Data center professionals today are grappling with a number of problems including underutilized storage and servers, complexity, lack of properly trained staff and the difficulty in meeting set SLAs.

With the increasing reliance of businesses on IT and the corresponding increase in storage requirements, server and storage underutilization seem ironical. However, to understand this one needs to see how applications are used today, says Anand Naik, director, systems engineering, Symantec.

"Storage capacities exist in silos in individual applications used by businesses and while each application is allocated a certain amount of storage, which usually exceeds the requirement, this storage cannot be utilized for other applications," says Naik

This leads to wastage of storage capacity even as certain other applications may be suffering due to lack of storage. "Enterprises are waking up to this phenomenon and using a number of storage optimization tools," says Naik.

Another major data center issue is the complexity that has become inherent to any large enterprise data center due to the sheer breadth of applications in use. "Data center complexity has multiple reasons including siloed applications, heterogeneity of software platforms, multiple staffing needs and lack of standardized ways of interaction between various IT tools," says Naik.

"The answer is a standardized approach, increased automation of data centers to minimize manual reliance and intervention, a correct governance model and the right-skilled staff that has skills across multiple software and hardware platforms."

There are a number of governance frameworks such as ITIL, COBIT, ISO etc that cater to various aspects of IT governance and enterprises need to adopt these as per their requirement, points out Naik.

Staffing has emerged as another huge challenge for data center managers. "Technology is changing rapidly, for instance you have things like virtualization now getting deployed, however, staff skills have not kept pace with this rapid evolution of IT," says Naik.

Moreover, you have multiple software platforms being used in a single enterprise data center, while the staff may have skills pertaining to a single platform. Multi-platform skills are not commonplace yet, so staffing remains a problem.

This has given rise to another set of problems; the failure of IT to meet service levels promised by it within an organization. The reliance of business on IT is growing and so are data centers.

"Data center growth without an equal growth or increase in qualified staff is the main reason for IT professionals’ difficulty in meeting service-level agreements (SLAs). Pervasive internal SLAs are difficult to reach due to growth; staffing issues caused by this growth make SLAs difficult to achieve worldwide," says Naik

IT budget growth has also not been able to match data center growth, points Naik. "Therefore, data center professionals are seeking different cost-cutting measures such as virtualization, server consolidation and standardization to deal with SLAs," says Naik.

So with all these problems facing data center managers today, wouldn’t outsourcing data centers be a more cost-efficient and effective alternative? "For some skills that are difficult to acquire outsourcing makes sense. However, on the whole IT remains the nerve center of businesses today and they may not want to give away control over IT by outsourcing their entire data centers," says Naik.

"And with no clear answers to the suitability of cloud computing model to enterprise IT, things may remain that way," concludes Naik.