UK Managed Service BrightCloud Opts for HDS AMS 2100

Client Background
BrightCloud provides managed solutions including hosted applications, on-demand computing, virtualisation, online backup, business continuity and network optimisation. The services are managed by the customer or by in-house team. As businesses have had to reduce IT budgets, BrightCloud has taken the lead as a quality Managed Service Provider (MSP), able to reduce costs and improve IT reliability for its clients.

BrightCloud was acquired by Open Reality in 2009 and is using subsequent investment to open new UK data centres. The acquisition has enabled Open Reality to offer virtual services to its own client base through BrightCloud.

Key challenges
BrightCloud was originally an SME IT solutions provider, delivering services to support the client’s IT strategy, with the majority of the IT infrastructure on the client side. Over time BrightCloud’s in-house work highlighted that having IT equipment on-site is often less efficient for the client or end user, due to redundancy, management and data protection considerations. Therefore, a process of collocation with existing customers began. This process naturally morphed into the development of virtual services and the first of five data centre ‘phases’.

The MSP commenced ‘Phase 1’ and opened its first data centre in 2005 at the Telecity Docklands site. As the company began to expand and its services became more popular with its SME customer base, the available storage at the ‘Phase 1’ site decreased and the original array began to wear out, although not unexpectedly at five years of age.

BrightCloud had already set its storage capacity ‘trigger’ at 50-60% in order to maintain site resilience, and the Open Reality investment meant the MSP could launch ‘Phase 2’ of its data centre expansion plan. As the original storage array was purchased over five years ago, Duncan Little, BrightCloud managing director, knew the SAN market had developed significantly and was therefore keen to source a solution that was as suitable for future development and expansion as it was for ‘Phase 2’ implementation.

"While the original array had proved sufficient for the initial five years of hosted services, we knew we would need considerably improved technology and engineering for the SAN at the second site, in order to cope with the rapid client-base expansion the company was experiencing, and the service level agreements our customers expected as standard", explained Little.

BrightCloud sourced the second data centre, selecting the installation at Park Royal in North West London, and then engaged its parent company’s value-added distributor Zycko, in the new SAN selection.

The solution
"It was less about what was wrong with the original solution, and more about what was available now", claimed Little. "Storage technology has come a long way in the past six years, and we identified several key purchasing criteria including high availability of service, input/output (I/O) performance, future proofing and quality of engineering, that were all essential for the new data centre and paramount in our preferred solution."

After considering offerings from several leading storage vendors including HP, NetApp and EMC, BrightCloud opted for the Hitachi Adaptable Modular Storage 2100 from Hitachi Data Systems.

Hitachi Data Systems’ platform is for medium to large businesses, as it is designed for use with SMB applications like Microsoft Exchange Server and VMware, and will easily scale to 120 disk drives.

Little continued: "After our technical team held comprehensive conversations with Zycko’s pre-sales support, it became very clear, very quickly that the solution from Hitachi Data Systems was the next level up. Take the I/O performance for example: on the AMS 2100 you can automate the process of I/O load balancing on two controllers – even with more expensive storage platforms you need to do masses of configuration, and then if you have a controller failure, you’re in trouble. Another aspect we considered was the quality of engineering; Hitachi Data Systems re-engineers SATA drives for example, with its own firmware, in order to increase the drives’ reliability. They spin down when not in use – this progressive approach to engineering was a key contributor in our selection."

BrightCloud also purchased Hitachi Data Systems’ TrueCopy and performance management software. It enables array-based replication, speeding up the MSP’s ability to copy and back up partitions on the SAN, a requirement for high-availability service level agreements. The performance management software operates with almost "red, amber, green-like simplicity", according to BrightCloud’s technical team. "If a particular disk LUN is running hot, then the software automates a response. The application de-skills the technology management process and means we can dedicate more time to directly servicing the customer", Little explained.

The cloud computing provider noted that while support wasn’t a primary driver as it has a technical team, the service provided by Hitachi Data Systems was refreshing: "While high-level support is expected, it’s nice to see it in action", claimed Andy Grover, Technical Director at BrightCloud. "We’ve not had a failure yet but were impressed when Hitachi called through whilst we were experimenting with the SAN in preparation for BrightCloud’s third phase. The unusual readings resulted in an engineer alerting us to what appeared, remotely, as array anomalies. It makes a massive difference knowing that Hitachi’s engineers are constantly monitoring the equipment, as it provides us with a second line of support resilience." 

Benefits
The AMS 2100 solution went live in October 2009 and the second data centre opened for business in November 2009. "Scalability was an important initial factor", noted Little. "Not only did we need a solution that could easily be installed into our new sites – Zycko’s engineers carried out the installation in a day – but we also needed to be able to increase capacity at each site over time. The ‘Phase 2’ AMS has already doubled in size with no teething problems whatsoever, and it’s reassuring to know that we have the ability to upgrade the technology with just a controller change
 
"As BrightCloud developed from an IT consultancy company to a true cloud computing provider, one of the most important considerations for any new infrastructure was its reliability. Hitachi Data Systems’ model boasts the famous ‘five nines’ guarantee which equates to no more than five minutes downtime per year. This is an essential aspect of the technology that we believe will be held in higher and higher regard as cloud services develop, and clients expect ever more rigorous SLAs."

David Galton-Fenzi, Group Sales and Marketing Director at Zycko, concurred: "Our existing partnership with Open Reality meant we had already established a trusted high-value relationship. When the network and infrastructure management specialist introduced us to BrightCloud, we knew Hitachi Data Systems’ AMS range would be an excellent fit. It not only fulfills the MSP’s current requirements, but will also grow with BrightCloud as it opens new data centres and looks to upgrade existing sites."

Future development
The ‘Phase 2’ data centre is over ten miles away from the original site and boasts advantages including power capabilities of up to 30kW / 120 amps per rack, security, excellent connectivity and well-trained on-site staff. BrightCloud is currently migrating customers to the new data centre and will eventually rebuild the ‘Phase 1’ site with another AMS.

The MSP’s expansion plans don’t stop there. The company is considering a new tier 4 data centre in Milton Keynes as the location of ‘Phase 3’.

Little concludes: "In order to increase the resilience of its managed solutions, BrightCloud doesn’t want too many eggs in one basket. We currently service over 100 customers with around 600 virtual machines and 30-40 terabytes of data per site. Our rule is to start building a new site as soon as 50-60% capacity is reached. We intend to eventually open five sites and then expand each storage array, welcoming new customers on board as we go, while providing current customers with more robustness and resilience than ever before."