Spotlight on data centre testing from professional services

Matrium Technologies Pty Ltd
By Patrick Barry, Product Manager - Professional Services, Spirent Communications
Wednesday, 16 May, 2012


Costly business disruption and remedial work can be attributed to performance problems in data centres, operational network systems and software applications. These issues often go undetected prior to launch due to the difficulty of conducting realistic performance testing. Quantifiable consequences of undetected performance problems include: revenue loss, customer base erosion and brand damage. All of these issues call into question a business’s long-term viability. Since the advent of network testing, Spirent test engineers have helped hundreds of clients save thousands of testing hours, which resulted in bringing products and services to market faster and with higher quality.

Spirent’s industry-certified test experts offering packaged test plans, and employing a world-class service delivery process, have delivered reliable test results on time, on budget, worldwide. Spirent Professional Services has developed a range of test services, with special expertise in data centre testing, which have been designed to detect and mitigate the full range of deployment challenges before customers launch new infrastructure and applications.

Using Spirent’s Europe region as a sample for an overview of Professional Services’ testing success in the data centre space, three brief customer case studies of data centre testing challenges and the solutions provided by Spirent are presented below.

Enterprise - pharmaceutical

Challenge: The system integrator for this large French multinational enterprise was responsible for building two new data centres and connecting them to the customer’s existing centre. The compatibility of the new devices with the original data centre needed to be assessed before cut-over. The customer needed to select the devices that delivered the right performance and high availability. Solutions from each vendor needed to be verified with the live data centre for performance, QoS and failover scenarios.

Solution: Employing Spirent TestCenter, Spirent Professional Services engineers implemented a rigorous test plan for throughput, latency and failover time in various scenarios. In addition, they tested the impact of network congestion on critical traffic, especially voice.

Benefit and outcome: Spirent testing revealed firmware and configuration defects in all vendor solutions, especially when under high load. Test reports that included vendor performance data provided both objective and critical comparisons of the different vendor solutions. With this report, the system integrator was able to recommend the solutions that best fulfilled the customer’s requirements for the data centre deployment. After selection, the customer requested two phases of retesting to ensure that all defects had been addressed. As a result, the systems integrator was fully confident that the network architecture would work as designed. When the customer went live they had no critical loss of service and are very confident that their data centre solution meets their future requirements.

Enterprise - construction group

Challenge: With 70,000 employees, the French multinational company’s new data centre had critical business applications, including Microsoft SharePoint, which had to support 32,000 local and remote users. Application performance and availability in the new data centre infrastructure needed to be verified for local and remote users in reading and writing modes. The new data centre needed to be launched without interruption to workforce productivity or the company’s tight construction schedules, which were intensely budget-sensitive.

Solution: Employing Spirent Professional Services, remote sites were simulated and 20 test scenarios were executed to determine the maximum load supported by the system. This test plan included integrating existing and new user behaviour scenarios with the applications as defined by the customer.

Benefit and outcome: Spirent testing validated the data centre infrastructure design as being in accordance with the exacting customer requirements. As a result, no additional upgrades were required and the customer launched their data centre on schedule and on budget, maintaining uninterrupted productivity of their workforce. As a result of the confidence inspired by working with Spirent Professional Services, the customer expanded their testing strategy to encompass web portal and firewall assessments, to ensure that these IT initiatives were also as ready for launch as was their data centre initiative, ensuring budgets, operations and schedules were not adversely affected.

Enterprise - financial institution

Challenge: A major bank in the United Kingdom, in conjunction with a key global network equipment manufacturer, planned to implement a new data centre network infrastructure. Network performance, availability and functionality of other network features needed to be ensured before launch. In addition, issues had been observed with ARP performance and traffic distribution under the QoS configuration in the financial institution’s live network. How network failure affected live applications running on the network needed to be identified before launch.

Solution: Spirent developed a test plan using Spirent best practices and test methodology and employing Spirent TestCenter. A series of core network tests were executed. These included testing the server farm (both stand-alone and virtualised), ARP performance, load balancing module performance and failover tests.

Benefit and outcome: Spirent testing identified configuration and firmware issues in the preproduction network. It also revealed the strengths and weaknesses of the various configurations tested. This evaluation allowed an objective decision to be made between the virtualised or stand-alone servers. The availability of servers with various hardware and operating systems combinations was successfully evaluated facilitating successful deployment of the data centre and meeting the launch deadline.

Observations on trends with virtual data centres

In the boom years companies tended to add new servers each time they bought new business applications, rather than risk compromising the performance of the already installed system. As a result of this proliferation, most servers in a typical data centre run at just 5-10% of capacity. Given the power of today’s servers, this approach doesn’t make sense. Virtualisation makes it possible to unite that processing power into one massive server, which then gets broken down by virtualisation software into many smaller virtual servers working at a healthier 60-70% capacity, and the remaining processing power can be switched off and power saved. Even though the virtual servers run on shared hardware, virtualisation software preserves their independence so they retain all the safety features of running applications on separate servers.
Staffing costs are saved by concentrating a highly skilled team into one central data centre instead of scattering them across the enterprise. This trend towards data centre virtualisation is an important solution for the numerous enterprises that are running out of infrastructure as this outdated and inefficient model of limited processing capacity utilisation remains operative in their existing networks.

This is especially true for banks and other financial institutions who cannot afford to take chances. With the process of planning a new data centre taking as long as 18 months, recognising the benefits of exercising full due diligence in that process is essential, and testing is a logical part of any important IT initiative.

The quantitative importance of testing

Nemertes Research reports in its issue paper Strategic IT Initiatives Need Strategic Testing: “Without proper testing, such strategic initiatives can fail, with serious unforeseen consequences, including significant hard-dollar and opportunity costs.” The Nemertes findings also state that gains from strategic initiatives can be minimised or erased if testing is not implemented before an initiative goes live, that testing should be performed throughout the life cycle of a strategic initiative, and that budgeting in advance for testing should be standard practice, allocating 2-5% of the overall budget, including capital expenditures and operational costs.

The customers above clearly recognised the benefits of testing their initiatives before launch to ensure the quality of their solution. Indeed a number of our customers continue to partner with Spirent to ensure that their evolving solutions meet their and their customers’ exacting standards of excellence.

Related Articles

Powering data centres in the age of AI

As data centres are increasingly relied upon to support power-hungry AI services and...

Smart cities, built from scratch

With their reliance on interconnected systems and sustainable technologies, smart cities present...

Smart homes, cities and industry: Wi-Fi HaLow moves into the real world

Wi-Fi HaLow's reported advantages include extended ranges and battery life, minimised...


  • All content Copyright © 2024 Westwick-Farrow Pty Ltd