Optimising data centre performance

In an attempt to optimise IT infrastructure, Indian organisations are looking at performance optimisation of their data centres through application performance management. By Chirasrota Jena

Shrinking IT budgets, outsourcing, increasing energy costs and the dramatic changes in technology landscape over the last five years have put CIOs under pressure to come up with innovative alternatives to traditional systems management frameworks that are effective at detecting and alerting system level and network issues such as servers and applications with high utilisation, high packet rates, or numerous retries. However, they do not provide a view of what the user actually experiences at the desktop nor do they give direction on where a problem is or how it can be corrected. Astronomical application fees, maintenance contracts, energy (heating and cooling) bills, hardware costs and staffing issues make it difficult to prove the business value of systems deployed because these expenses eat away a significant portion of the annual IT budget. The concept of IT Performance Optimisation (ITPO) is gaining momentum nowadays to overcome all these problems. The motive behind ITPO is to extract the optimum performance from IT infrastructure starting from software applications, to the hardware as well as the network and servers and services provided while supporting IT. However, optimisation is only one aspect of management, which also includes administering and maintaining control over IT in order to deliver services. Many large businesses are now looking at performance optimisation of their data centres along with network and service management. Existing complex infrastructure navigated by application data-access requests can significantly impede the process of diagnosing performance problems and undermine day-to-day management. Bottlenecks can appear at any connection in the data path. With limited visibility into a constantly changing infrastructure, IT must rely on automated tools that streamline the identification and isolation of performance problems wherever they occur.

Madan Mohan, director (Consulting), Information & Communication Technologies, Frost and Sullivan says, “Companies cannot afford to ignore an inflexible, sub-optimal (both in capacity and performance) infrastructure. Companies want their infrastructure to become vendor agnostic and scalable. The data centre market is an emerging one and more traction is required before we witness the emergence of competitive products and services. Companies are open to server consolidation because of the heterogeneous server systems that they have built up over time. On the storage front, more solutions are non-priced with vendors often offering these at the pre-sales stage.” Organisations continue to work to defend the availability and performance of critical data and applications. By leveraging tools to apply best-practice approaches for optimising performance and ensuring availability, organisations not only protect against costly downtime, but they also safeguard the quality of service their customers demand in today’s highly competitive business environment.

Need for performance optimisation

Data centres are becoming more complex with the deployments of newer applications and tools. Today’s mission-critical applications hosted in a data centre perform a number of important business functions. They generate revenue, run supply chain management and the delivery of essential government services, to name just a few. Says Kasturi Bhattacharjee, principal consultant, PricewaterhouseCoopers, “The latest thrust areas in data centre performance optimisation are into consolidation, capacity management, vulnerability management and performance enhancement. An organisation typically looks for performance optimisation to reduce cost by reducing complexity, improving manageability and service levels with enhanced security. Apart from optimisation the other areas of concern for most CIOs includes system and application availability management, disaster management and storage capacity management.”

Suresh Menon, business manager, Business Critical Systems, HP India adds, “For organisations deploying these applications, the success or failure of the business depends on the ability of end-users or customers, to complete transactions reliably and the ability of IT teams to find and fix problems rapidly before service level agreements (SLAs) have been breached. Ensuring transaction integrity and user satisfaction are key to achieving business goals through Web applications. These applications have led to the need for more computing and storage power.” The deployment of application performance management tools allows organisations to immediately identify, prioritise, and assign an owner to application slowdowns, which can significantly reduce business cycle times. Organisations are also better able to understand the actual response time users are experiencing and respond accordingly, thereby improving customer service and satisfaction. Application performance management tools represent a critical component of an overall strategy for high availability, since performance is a natural extension of availability, and applications represent the user experience to IT. Enterprises must manage performance degradation as well as major failures that might result in downtime. To ensure data centre uptime, organisations should consider implementing data protection, replication, performance management, and clustering technologies.

Effectively managing for customer success is no easy task. The increasing reliance on Web-based applications means that direct customer interactions have shifted from business owners to IT. Rajendra Dhavale, consulting director, CA, India and SAARC says, “Now business managers have limited visibility into customer experiences and the success of business processes. As the first line to the customer experience, IT must provide the business with visibility into customer transaction performance in addition to isolating and eliminating complex problems throughout the Web application environment in the data centre that includes Web servers, application middleware, databases, operating systems and networking devices. Hence business organisations need to look for various technologies to optimise performance of their data centres.” By leveraging tools to apply best-practice approaches (such as ITIL and BS7799) for optimising performance and ensuring availability, organisations not only protect against downtime, but also safeguard the quality of service their customers demand.

CIO challenges

Data centre costs and complexity grow in proportion to the number of servers you deploy, yet you must still reduce costs, meet service level requirements, and satisfy the increasing demand for resource capacity say industry pundits. At the same time, you must also be able to rapidly deploy new applications without risk to service or further increasing cost and complexity. Most CIOs are currently focusing on power density and heat dissipation issues rather than improvements in application performance or supporting a greater number of users per server. Capacity planning and vulnerability assessment are the key issues faced by CIOs. Bhattacharjee adds, “Some of the key challenges for a CIO would be to address the areas of exploding cost and rising complexity in terms of configuration and security that are added with the introduction of each new service. The other areas are meeting the continuous compliance requirements and maintaining quality of service levels.”

Explains Sanjeev Gupta, product manager, Site and Facilities Services, Global Technology Services, IBM India, “Excessive heat and insufficient power are among the top three issues faced by CIOs. Moreover, an IDC 2006 study predicts that power and cooling spend will exceed server spend within the next two years. So for CIOs, the issue is about handling inexpensive dense computing and increasing power costs which are shifting requirements and spending in data centre. CIOs will have to keep looking into data centre strategy and planning which will make a positive impact in reducing power consumption and maximise hardware utilisation, better leverage IT resources and reclaim real estate in the server room.”

Additional challenges appear as companies turn to the Web as the primary channel for consumer and business applications. The compaction of IT equipment and simultaneous increases in processor power are creating challenges for data centre managers in ensuring adequate distribution of cool air, removal of hot air and sufficient cooling capacity. The requirement to deploy high-density servers within single racks is presenting data centre managers with a challenge. To avoid the financial burden of building a data centre some CIOs are opting for outsourcing. While virtualisation is becoming increasingly widespread, many applications have not yet been tuned for virtual environments. Along with this the security and compliance issues are also poised as major challenges before CIOs. Menon says, “CIOs are facing challenges in each and every area of data centre performance starting from system utilisation to power consumption to cooling requirements and space optimisation.” Engineering parameters such as heat and power density management are the two other issues. Organisations must be able to achieve measurable application performance, productivity, and availability improvements that directly translate into bottom-line savings. They must be able to detect, diagnose, and correct performance problems before service levels are affected.

Spot ‘hot zones’ and eliminate power wastage

Enterprise data centres had infrastructure reliability and high uptime as key drivers for technology selection and deployment up till 2002. In successive years, IT equipment refreshes have brought about increased power densities, and this in turn has driven new approaches to powering and cooling IT equipment. A fundamental technology shift started with the need for increased computing performance per kilowatt. Informs Mohan, “From an Indian company’s perspective we can see three broad trends in the data centre. Firstly, companies are relocating ‘rudimentary’ data centres from multiple sites to a centralised and focused data centre. The other trend is of server and storage consolidation. This trend is more visible in banks and in the government sector where the focus is on gaining efficiencies, improving utilisation rates as well as reducing complexity in managing server sprawl. Finally, firms are also looking at asset optimisation, a life-cycle approach of planning and management of key assets.”

With a 42U server rack holding multiple blade servers consuming power from 6 kW to 24kW, now going beyond 24kW, data centre cooling has moved closer to the rack. Since it is still unclear when equilibrium between IT hardware performance and the physical capability to support the hardware environment will be reached, critical physical infrastructure technology designs have taken the route of scalable, modular and efficient design which helps an organisation deploy as per its present requirement but still be ready for future IT equipment refreshes. It can make entire data centre design and performance highly optimised as it reduces the total cost of ownership. Remote monitoring of a data centre’s physical infrastructure along with its integration with network monitoring also helps in overall performance management. Says Gupta, “Enterprises have started focusing on obtaining better energy efficiencies in their data centres. It starts with the use of servers with efficient power supplies, server virtualisation, more blade servers and their energy management technologies. Equally important is data centre planning and modelling which creates a master plan that takes cooling best practices into account. This master plan is used as blueprint for achieving high energy efficiencies throughout the life of a data centre.”

An upbeat scene

With every Indian and MNC corporate looking to IT for business enablement, setting up data centres, disaster recovery centres or consolidating its existing data centres—this is becoming an important part of the annual IT budget. Informs Gupta, “The last four years have seen the Indian data centre market growing robustly. IBM India has already executed data centre projects exceeding 2.5 lakhs square feet for over 55 clients. The data centre business in the small and medium business (SMB) market has been growing steadily. Indian organisations are looking to IBM to assist in moving them from a server room environment to a data centre. CIOs and IT managers of organisations are updated on mission critical environment needs for which their data centre must be designed. But risk mitigation and service management is still the priority while selecting a data centre partner organisation.” Growth drivers for the offerings for data centre can be categories into three broad markets. Large organisations looking for tier 3 and 4 design data centre ranging from 5,000 to 50,000 square feet are the segment that most vendors are eyeing. Nowadays SMBs are looking to move from a server room to a data centre. These firms look for an end-to-end solution with service and support apart from a scalable and modular design to make the best use of their IT budgets. Some organisations want to have an innovative design as well as an environment friendly facility. Site and facilities services product portfolio help IBM clients evaluate, design, implement and manage their IT infrastructure and physical environments. Services products are designed to deliver best design practices though various assessment services to target the SMB sector.

Industry leading application management solutions offer IT organisations deep insight into the whole Web application environment that is hosted in the data centre. They also bring the customer experience back into focus for business managers who lack real-time intelligence about customer satisfaction and the health of critical business processes. Menon opines, “The whole scenario has changed in India. Earlier the data centres were set up in an ad hoc manner. Now the whole system has become more systematic. The business requirements have forced Indian customers to rethink around the consolidation of storage. There is a growing concern found among the organisation with the growing complexity of data centres. Each and every player is providing its solutions to ensure best performance coupled with cooling and virtualisation.” While elaborating on the ITPO, he further added it is about improving how applications work in organisations and about raising the bar with regard to performance. It is about hardware as well as software, because the platform upon which the application runs plays a crucial part in how well it can function. Perhaps the most important part of this optimisation process is the management that can be applied, including the administration and control over IT that is needed to deliver a service to where it is required.

Vendors in this space have positive expectations from the Indian market. Dhavale says, “From the Line of Business (LOB) Managers that fund strategic applications to IT operations that deliver the applications to customer service that is accountable to customers, these offerings provide facts about your internal and external customer’s interaction with your critical applications in business terms that all key stakeholders understand.” CA’s Wily Technology Division provides a comprehensive suite of products and services specifically designed to achieve customer success through monitoring and management of end-user experiences and web infrastructure performance. Application performance management solutions from Wily are an essential element of CA’s comprehensive Enterprise IT Management (EITM) approach that enables customers to securely unify and simplify the complex and evolving infrastructure.