In 
Frank Huang, Data Centre Professional Technologist, Raritan Inc., finds that the most difficult part of data centre management is to bring all components under a single framework. In conversation with Manu Sharma, Huang says it is critical to align the data centre with business needs and growth
What are the data centre related challenges IT heads and teams face with regard to power, space, resource allocations and optimisation?
IT heads and teams are now more focused on the consolidation and centralised management of data centres. They are deploying exclusive and individual management tools to manage power, space, and other components. However, the challenge is to bring all the solutions under a single framework or roof, which can be managed through a single portal.
The toughest of all challenges is to handle data growth amongst the hardware, where research groups like Gartner have predicted a rampant growth. Gartners survey indicated that data growth is the biggest data centre hardware infrastructure challenge for large enterprises. Besides, most enterprises have reported that the investment on data archiving solutions would increase.
What are the best practices that you would recommend IT managers follow to drive data centre efficiency?
As per industry reports, data centre space can consume about 100 to 200 times more electricity as compared to standard office spaces. As such, they are prime targets for energy-efficient design or solutions that can save money and reduce use of electricity. However, the critical nature of data centre loads elevates many design criteria chiefly reliability and high power density capacity far above energy efficiency. Short design cycles often leave little time to fully assess efficient design opportunities or consider cost versus life-cycle-cost issues. This can lead to designs that are simply scaled up versions of standard office space approaches. Though past strategies worked well, they may not suit current requirements of energy efficiency.
The four key trends nowadays for IT managers to follow to drive data centre more efficiently are: Create energy efficient data centres
Consolidation Fuse programs so that all servers are run from a central location, or can be managed by a singular data centre operating system.
Private Cloud There is a shift towards private cloud solutions which are cheaper and provide easier management of data capacity.
Storage Tiering To manage more data while minimising costs, follow the tiering process to archive inactive to less-active data, while reserving top tiers for data that is used constantly.
What are some software tools and technologies that you recommend to enhance data centre efficiency? Could you elaborate on effective ways to implement it?
I would recommend IT managers to opt for intelligent PDUs along with appropriate sensors and Power IQ tools to analyse the garnered raw data. The analysis would help in bringing out effective management solutions. For, it would enhance data centre efficiency in the areas of power management, environment management and energy management. Besides, data centre tracking tools would help IT managers to address change management, capacity management and asset management needs.
What kind of innovations are happening in the data centre which will help IT managers transform their IT environment?
Asset management solutions and tools are making an impact on the data centre efficiency. They help in tracking assets remotely and give information in real time about the changes made dynamically.
Can you explain effective strategy around increasing the RoI within a data centre and lowering total cost of operations?
Based on the premise that you cant manage what you cant measure, data centres are undertaking steps to measure IT device-level power consumption and rack environments in an effort to minimise power use. Using PUE as a driver, and thereafter measuring the Total Cost of Ownership (TCO) metrics (which factor in the cost of the server plus total cost of the energy that it will consume while running a specific workload over its lifespan) is one strategy which can help monitor and hence manage the ROI better.
IT managers can request for ROI calculators from solution providers which is very common in the current scenario that can be used to determine which solution best suits the company and budget accordingly.
Is there a data centre maturity model that IT managers can adapt to? Please elaborate with examples.
The Green Grid released the Data Centre Maturity Model (DCMM) in March of 2011 to provide a road map for data centre designers and operators on current best practices and the path to best practices, five years into the future. The Models Level 2 documents industry best practices and Level 5 serves as a target for organisations to reach in the future.
One example is eBays Project Mercury which incorporates many current and future best practices documented in the DCMM. This provides a real world example of how companies can save resources and money by using the model from the beginning of design through the operations of their facilities and IT infrastructure.
Another example: assessing its current data centre portfolio, eBay determined that the existing model was unsustainable. An inventory audit showed that hundreds of different hardware configurations (SKUs) were still being ordered and deployed in data centres, broadly conforming to the Uptime Institutes Tier IV data centre specification. Further investigation showed that 15 of these SKUs made up 80 per cent of the hardware in the data centres and a majority of application servers did not require Tier IV. This over-provisioning of resiliency resulted in an infrastructure that was far more costly and complex than necessary. eBay used measures including total cost of ownership (TCO) and The Green Grids Power Usage Effectiveness (PUE) to highlight the inefficiency of housing servers in expensive, high resiliency space unnecessarily.
One of the principles of The Green Grids DCMM is to align business needs with the most appropriate level of data centre resilience and efficiency. A frank assessment of eBays application needs, as recommended by the DCMM section 3.2.2, determined that 80 per cent of the application portfolio required only Tier II reliability. By aligning these applications to the appropriate level of redundancy, eBay freed capacity in its Tier IV data centres, effectively extending their lives. Moreover, eBay also was able to augment its portfolio by adding new high-density, low-cost, and highly efficient Tier II space.
Another core benefit was that the total cost of ownership for the data centres plummeted, supported by a reduction in capital investments and operating expenses of the new Tier II data centres by more than 50 per cent. Looking at the DCMM, the right sizing of individual applications to the most appropriate level of resilience is a five-year-out best practice. By implementing this best practice now, eBay is on the leading edge when compared to most typical IT organisations.
Amidst the hype around cloud, what challenges are data centres throwing up? Please elaborate on the effective ways to address these challenges.
The heart of all IT operations is the data centre with its complex infrastructure space, power, cooling, cables, servers, storage and communications. It is capital intensive, energy intensive and technology intensive.
The challenges for both the data centre operations management and the organisations executive management responsible for the IT function come from all directions. For example, minimise risk of downtime; satisfy demand for new applications to facilitate business/revenue growth; satisfy internal user needs as overall organisation grows and changes; comply with audits both internal and government regulatory; reduce/control energy usage/costs; maximise utilisation of current resources; improve staff productivity; incorporate processes to address green initiatives; deal with the rate of technological change; live within tighter budget constraints; finance major capital investments.
Simply stated, the challenge is to properly align the IT function operation with the organisations overall business plan and operational requirements.
Devices are getting power hungry and assets need to be monitored and measured more to optimise use of power. This is a high priority among many IT managers across the world, while at the same time, they balance this need for increasing reliability and security for its mission critical solutions.
In the area of Data Centre Infrastructure, the total Data Centre Infrastructure Management solutions will help IT managers to manage smarter and more productively to meet these challenges:
- IT Infrastructure Access and Control
- Secure access solutions meeting corporate and government standards
- Multi vendor management
- Energy and Environment Management
- Power distribution and Management
- Asset Management
- Capacity Management
- Change Management
However, customers can avoid the hassle of handling these challenges by opting for a cloud model, where the service provider could carry the overheads and ease the work of IT managers.
Add new comment