Hitachi Data Systems' new unified compute platform (UCP) is a turnkey solution for service providers that covers servers, storage, network infrastructure and applications, Adrian de Luca, director, presales and solutions for Australia and New Zealand told iTWire.

A key differentiation from other players in the market is that UCP includes orchestration software to simplify infrastructure management to the extent of providing a self-service portal for the end user. That orchestration layer will also include an SLA engine to automate systems management in order to deliver services according to the SLA.

The initial release will focus on Microsoft software, including Windows Server 2008 R2, SQL Server 2008, System Center, and Hyper-V. More diversity is planned, and "VMware will be the first cab off the rank," said Tim Smith, senior marketing manager for Australia and New Zealand.

HDS is providing the necessary infrastructure as a service to service providers (SPs), not directly to end users. Different SPs are expected to address the needs of different sized clients in various vertical markets, including government.

The model offered by HDS scales up and down: the more you need (performance, storage, etc), the more it costs. Smith pointed to a survey that found 52% or organisations were unable to change their IT environments to adjust to the global downturn, and observed that the UCP model also accommodates seasonality.

It's not necessary to junk all the existing equipment - please read on.

---

Another important aspect of UCP is that it can accommodate equipment housed in the SP's data centre as well as that provided in the cloud by HDS. This is a key part of the way data and applications can be migrated from existing systems.

Lindsay Campbell, professional services manager for Australia and New Zealand, told iTWire that the basics of the migration methodology were plan, design, and implement.

The 'plan' stage includes a survey of the client's existing situation and assessing the scope of the migration. Proprietary tools help the company provide good value to the client during this stage.

It also involves working with the customer to identify specific pain points and priorities.

The 'design' stage includes determining exactly what is required and how it will be implemented. "We spend more time on the plan and design than on the migration," said Campbell.

And when it comes to implementation, "we offer a seamless transition," he said. One action item that can require a small amount of downtime is the introduction of HDS's virtualisation hardware if it isn't already present, but that can be done with an interruption of 15 minutes or less. Once that is done, the physical migration can be done with 100% availability.

"15 minutes is probably a safety margin," said Smith.

Why can HDS complete a migration relatively quickly? Find out on the next page.

---

By virtualising the existing installation behind one of HDS's enterprise servers, everything can be controlled from one place, even if the SP is using storage arrays from multiple vendors. If necessary, the virtualisation server can be dropped into the SP's data centre on a temporary basis to allow the migration.

Either way, this does not require that existing storage is reformatted, which is a time consuming and risky operation that Smith said can mean migrating at the rate of one to two months per storage array as opposed to the 13 arrays in a week that has been achieved by HDS.

Once virtualisation step has been carried out, the rest of the project can take place at the SP's pace. "It is irrelevant how long that migration may take," said Campbell, noting that people usually prefer to get rid of old equipment promptly to avoid maintenance and running costs.

Equipment that hasn't reached the end of its useful life can be kept in action, for example to provide a second or third storage tier.

Generally speaking, HDS recommends cleaning up the existing infrastructure before migrating, but there are times - such as a project recently performed for the NSW Government - where it makes sense to leave some of the housekeeping until after, for example to take advantage of features provided on the destination hardware.

"We've been deploying cloud infrastructure solutions for five years," said Smith, citing Telstra as one of the company's customers.

Want another example of HDS's migration experience? See page 4.

---

Another example involved the transfer of CSC's mainframe applications from the US to Australia - HDS worked on the data migration side of that project, using the same tools and methodology that are applied to UCP migrations: it doesn't matter if the storage hardware is side by side or across the world," said Campbell.

There's more to UCP than just storage, which is HDS's traditional strength. So the company has partnered with Microsoft, Brocade, Cisco and others to provide a complete end-to-end system, said Campbell.

"Together with Microsoft, we are uniquely addressing customer challenges by looking at the data centre from the business process perspective and driving innovation down to and across all components," said John Mansfield, senior vice president, global solutions strategy and development.

"The Hitachi unified compute platform will give customers an automated, virtualised, cloud-ready and sustainable data centre infrastructure and provide the unique advantage of a flexible foundation for long-term growth and sustainability," added Mansfield.