BECAUSE EVERYTHING IS ALL-TIME™

Tags


Tag

Data centre

SunGard Availability Services - Data Centre - LTC

We have added a further 8,000 square feet to our highly resilient and secure London Technology Centre (LTC) to meet growing customer demand for managed hosting and, increasingly, cloud-based Infrastructure as a Service solutions.

The expansion will allow more businesses to switch their IT costs from a capital expenditure (CapEx) to an operational expenditure (OpEx) model. Our shared service approach makes high quality data centre space available at a fraction of the cost of any one company trying to buy, resource and power its own dedicated facility, recently estimated by Broad Group Consulting to be between £5 and £7.8m per 10,000 square feet. It also addresses the concerns of many businesses that space and power supplies in the capital will be inadequate as data usage continues to rise.
 
 
 

With the first organisations scheduled to pay the new climate change levy in 2012, many are anxious about what effect the change of government will have on the measures intended to force UK industry to reduce carbon dioxide (CO2) emissions. Unfortunately, the answer appears to be “wait and see” as the Carbon Reduction Energy Efficiency Scheme (formerly known as the Carbon Reduction Commitment) introduced in April 2010 was afforded just one small paragraph in the Comprehensive Spending Review.

Under the original scheme, british businesses with energy use above 6000MWh of electricity that were not subject to any other energy efficiency scheme would be charged a levy to discourage heavy energy usage. This was billed as being a central part of the UK’s strategy for reducing carbon dioxide emissions in line with our commitments under the climate change Act 2008. As the scheme was ostensibly designed to encourage changes in behaviour and infrastructure, firms could apply for a sizeable proportion of the charge to be returned in exchange for implementing certain energy-saving measures.

The coalition government has committed to simplify the scheme to reduce the burden on businesses. While the details are still under discussion, from what we know of the new-look scheme, it seems that while the charges are still in place the rebate mechanism has now been removed entirely (presumably under the guise of simplification). 

SunGard experts are keeping a close watch on the developing situation and any further announcements from the Department of Energy & Climate Change (DECC), which may lead to further legislation. But it seems inevitable that the government intends to levy  an additional charge on industry calculated on the basis of past energy usage.

Whatever form the scheme ultimately takes, it will strengthen the argument for buying into a shared infrastructure that services multiple organisations, which is inherently a ‘greener’ and cheaper way to provide computing and workplace recovery than individual companies trying to buy, resource and power their own standby facilities.

Cutting through the hot air

IT spending surveyWe promised to bring you the full findings of our IT survey of 100 chief financial officers (CFOs) in mid-sized UK headquartered firms conducted by Vanson Bourn on behalf of SunGard Availability Services.

Two thirds (66%) of those responsible for financing it expenditure admitted to not fully understanding the benefits of moving to cloud computing, although this is recognised as being a key technology to reduce  it spend. This is particularly concerning as almost  two thirds (62%) of CFOs expect their it budgets to remain static over the next three years.

More than half (56%) of those polled said they are deterred from outsourcing the management of their it infrastructure by the perceived security risks. The research showed that these fears are exacerbated by high profile media stories about third party it outages or data losses with nearly half of respondents (45%) confessing that such cases make them more inclined to keep their data in-house, despite the cost implications.

While almost half (45%) of CFOs said they aspire to remove data centres from the balance sheet, almost two thirds (60%) had concerns over loss of control in handing data over to a third party. This concern could be attributed, at least in part, to a lack of understanding as less than a third (28%) said they knew the distinction between private and public clouds, which as you will know, differ radically in terms of security and resilience!

Interestingly, the survey highlighted a marked difference in attitudes between home and work use of the cloud. More than two thirds (67%) of CFOs have eagerly embraced cloud-based apps such as Hotmail, GoogleMail and Spotify for their personal use, while  just over a quarter (26%) currently use corporate applications in the cloud.

A solid record and history of resilience in protecting customers’ data was rated the most important attribute in a third party provider by 49% of respondents, followed by a well-known brand name (35%) and impressive ROI statistics cited by just 16%.

The findings suggest that organisations are looking for a solution that offers all the benefits of the cloud – such as cost savings and increased agility – but where data and applications are stored in a fully resilient and secure data centre that allows firms complete control. As this is exactly what a private cloud offers, it is clear that  those with it responsibility probably need to do more to educate financial decision-makers about the benefits of moving to the right cloud environment.

SunGard Availability Services Data Centres

On recently being introduced as a data centre specialist, a US investor piped up, “Ah, the garbage bins of the IT world!”

My, perhaps unsurprisingly, preferred analogy is with the PBX or office switchboard – it may not be sexy or high-profile, but it is central to any business and it needs to be incredibly reliable.

However, perhaps my analogy is also becoming less accurate. Unlike the humble PBX, overtaken by VoIP and the convergence of it and telecoms, the data centre is gaining in prominence and importance. While it could yet struggle to be described as ‘sexy’, it is the critical element to many sexy parts of the it world. As the CEO of data centre provider, telecity, describes it to investors, “The only non-virtual part of the virtual economy”. Or, as the CFOs of the likes of Google and Microsoft see it, easily the biggest capex line in the company.

The final point also highlights an interesting trend. Cloud computing, rather than being a substitute for data centres, actually increases the requirement. While the notion of cloud computing infrastructure being anywhere is attractive, for security, legal, auditing and business reasons, most corporates wish their data  to remain in the same region. Thus meaning there won’t be a rush on cheap out of country data centre space to host the latest cloud.

Of course, what really ensures the attention of the data centre at the forefront of executives’ minds is their importance, both to the business needs and the bottom line. Even in the economic challenges of 2009, many data centre providers saw annual growth of 20-30%. European telecoms traffic going through internet exchanges is continuing to increase at 50-60% per year. Further drivers going into 2011 include video (and particularly the potential for HD and 3D tv), smart grids, high speed broadband and mobile, and regulations particularly in the financial sector. The desire for vendors to take part in this demand can be well illustrated by the recent bidding war for data centre storage firm, 3PAR.

But, while the importance of data centres may have increased, the role of data centre managers and importance of data centres in business and IT thinking has yet to keep pace.

Rather strangely, there is no obvious voice or organisation for data centre managers. They often belong to broader organisations, such as facilities management or computing society, or more specialist groups, based on areas such as cabling or cooling. IT strategy decisions are often made with scant regard to the data centre, and decisions can also be shared between a number of departments such as property, facilities management and IT.

There is also the element of control and centralisation – as so often in it, the political and cultural reasons are bigger inhibitors than any technological one.

As someone at a bank recently put it, “I had to change my title from global enterprise architect, because I was getting so much negativity”. Basically, business units resented any sense of control, centralisation and interference. Such issues, as well as historical attitudes, explain why only around a third of servers are actually in proper data centres, and nearly 90% of data centre requirements are kept in-house.

Outsourcing is often simply not on the agenda. On asking a data centre manager at a multi-national recently about whether they had considered outsourcing their data centre, they replied, “Of course not, that it is too business critical”. Reasons include control, security and cost.

In a recent discussion with an IT director about data centres, his view was that, “We can do it a lot cheaper in-house”. But are they really understanding all aspects of the financial calculation – impact of depreciation, operating costs (of which power could easily be  30-40%), risks (whether technology or environmental obsolescence, or simply the fact of how a data centre will evolve in the 15 years they would hope the facility  to remain in use)? it is also use of finance – last year, we saw some outsourcing simply due to the need to move costs from capex to opex. Data centres can cost  around £5-7.8m per 10,000 sq ft to build and fit-out, and is this really the best use of company resource? equally, is it best to allow a specialist in the area, with all the associated skills, experience and economies of scale to manage and run data centre services? 

Many data centre managers have rebutted attempts at outsourcing. Even companies, who regularly outsource activities from it to catering, often still keep their own data centres. The thinking, experience and expertise from other sourcing decisions – from cost comparisons of in-house vs outsourced to SlA and t&c issues – have rarely been used in conjunction with data centres.

Clearly, outsourcing does not work for all companies, and a hybrid route between in-house and external data centres can often make most sense. This is particularly true as, by meshing such facilities, far greater reliability and redundancy can be achieved. Indeed, a number of the larger banks have recently gone through exercises to rate applications as to the sort of data centre they need – such as specification, location and connectivity. The conclusion of such analysis has often been that they need to be less dependent on highly expensive facilities near capital cities.

On a final note, choosing a third party provider really requires a separate article. One it manager recently told me the reasons he had never considered a third party was that, “We looked five years ago, and facilities out there were very low quality”. At that time, many of  the largest third party facilities were built in the early 1990s as bankers remained cautious for five years after the dot com crash (which led to the demise of 17 of  27 pan-european players). Since that time, there has been much, high quality new build. However, the credibility and expertise of the data centre outsourcing provider remains key.

Source BroadGroup

Find out more about SunGard's Colocation and Managed Hosting solutions

The global economic downturn has forced companies to refocus on their core business. As a result, many are now reconsidering the merits of outsourcing on-core business areas, such as management of the organisation’s IT infrastructure, to specialist third party providers. SunGard availability Services believes the role of the in-house IT department will increasingly become one of governance and architectural design and that this will become the ‘new normal’ due to the numerous business benefits created by an Infrastructure as a Service (IaaS) model:

In-house data centres SunGard Availability Services

Cost-savings
Research conducted by SunGard into the relative costs of ownership of IaaS, Managed Services and in-house solutions revealed that on average organisations can reduce total IT expenditure by as much as 55% when moving into SunGard’s secure, virtualised, cloud environment. These savings come from many sources: the soaring cost of data centre development and operational costs is one obvious area, while the savings to be gained from using a shared infrastructure rather than creating new platforms on an application by application basis is another example.
 
Removes management burden
While a decision to outsource IT infrastructure may originate as a cost saving exercise, the real benefits of outsourcing actually come from saved management time, which can then be used to focus on more strategic activities.
 
CapEx to OpEx
This is particularly relevant in the current economic climate where banks are reluctant to lend and funding is scarce. The cost of building or upgrading a data centre today is huge, requiring a multi million pound investment to be diverted into a non-core business area where it will be tied up for years to come.
The OpEx model of outsourcing gives back balance, control and predictability of costs, enabling organisations to continue to invest in the revenue-generating areas of the business.
 
Business demand driven agility
Whether businesses are conserving resources to weather a feared ‘double dip’ recession or want to be in a position to take advantage of the hoped-for upturn, the ability to scale services and the workforce support up or down according to business demands is a prized feature. With outsourcing, IT assets are not left unused in ageing data centres and companies pay only for what they need now, rather than tie up valuable capital in anticipation of future growth.
 
Greater resilience
The levels of redundancy built into SunGard’s highly resilient data centres require capital and operational investments that often cannot be commercially justified by one company alone. With utility costs alone for a 100,000 sq ft data centre averaging £3.7m a year*, the gold standard of data centre management practised by specialist hosting and managed service providers such as SunGard is beyond the pockets of all but the largest firms.
For instance, even the incidental expense of an annual deep clean of a single data centre, incorporating zinc whisker and air particulate tests, runs to tens of thousands of pounds.
 
Environmental benefits
The shared services model is inherently a more environmentally responsible form of computing. In the light of issues associated with real estate, power draw, cooling and other resources, a shared data centre that can service multiple organisations is intrinsically a more sustainable model than companies trying to buy, resource and power their own standby facilities.
 
Despite the clear benefits of outsourcing management of a data centre to a third party and harnessing cloud computing technology, some still doubt the security and reliability of public clouds. There is no question that choosing a trustworthy cloud partner with a demonstrable track record of security, resilience and financial stability is vital. For this reason SunGard’s own private cloud is proving popular with organisations that want to reap the benefits of an enterprise-class, cloud computing infrastructure without any of the perceived risks.
 
*Datacenter Journal 1/9/9

 

Historically, cost-effective recovery has been constrained by the limits of technology. Now it is possible to costeffectively recover more people via super connectivity and one seamless estate through virtual and physical environments. Here’s how to exploit new capability and benefits for a flawless recovery.

Calamity comes unannounced and undefined, making recovery planning a dicey exercise. Despite a careful calculation of the odds, it is nearly impossible to foresee what type of disaster will actually strike and what problems it will drag in its wake. Nevertheless, there are strategic steps that can preserve and protect the enterprise in nearly all scenarios. The key is to see mundane details as pivotal and to view complex business issues in orderly subsets of vital functions.
 
Unfortunately, many a company views recovery only in terms of the technological aspect. While it is true that technology has made recovery efforts far easier and more efficient than in years past, information
is only one part of a company’s total composition. In other words, retaining data through technological means alone will not save the enterprise overall. Retaining data does not equate to continuing full
operations unabated.
 
Alternately, there are companies that focus merely on the physical side of recovery believing the key to resuming profitable operations is more a matter of moving its people to a safer, yet still centralised, place. While in theory that sounds perfectly logical, it is often more perplexing in practice. Traffic can be frozen in place; road systems destroyed; water and food can be contaminated or in short supply; electrical power can be off for weeks or even months.
 
The problems then can range from how to move people to another company location to how they can work when they get there if there is no power, no water and no safety.
 
In short, both the technological and physical strategies are one-sided efforts that can fall victim to the most mundane of troubles.
 
The better strategy is to focus on how the business should actually operate during a disaster. Determine what precise business functions need to be restored first and in what order. Second, determine the core group of personnel you need to accomplish that directive. It is important to identify which personnel are most useful to your recovery in terms of technological skills and end customer experience. The ultimate goal is to retain customers and accommodate new customers who may arise during a crisis—if for no other reason than they cannot reach your competitors. It is therefore imperative to restore customer communications as fast as possible.
 
Connectivity, then, is crucial during a crisis. Even more so given the rise in the mobile workforce. The ability to connect to your data and telephony wherever you are recovering from is essential if you are going to give your customers a businessas- usual experience.
 
Otherwise, where do the inbound calls go? How are they managed? Customers will first notice that they can’t reach you. Make sure all channels of customer communications are resumed as fast as possible. Look for a vendor that provides connectivity across all channels and in spite of public utility outage.
 
Both the technology and physical parameters should be planned accordingly. The ultimate goal should be to plan recovery efforts with maximum flexibility and the least number of vulnerabilities.
 
Here are the best practice tips for disaster recovery efforts to succeed despite the circumstances of any given catastrophe:
  1. Think short and long-term 
    Most disasters are short-term events lasting on average seven to eight days but there is also the threat of disruptions that can last six to nine months or even longer. Your disaster recovery plan, therefore, should include an immediate phase, a short-term phase and a long-term phase. The ultimate goal is to achieve a seamless environment for customers and workers in all three phases.
     
  2. Keep customer communications open 
    Discern how your customers communicate with your company and plan now on how you will have those channels up and running as fast as possible. The longer communications are down, the more likely your customers are to become alarmed or lose faith in your company entirely. Look for a vendor that can provide complete communications using your company’s own phone numbers, website urls, email addresses, etc. Said vendor should have multiple trunks and redundancies to ensure telephony and internet connections are protected and viable.
     
  3. Ensure key employees can report in
    Ensure you have more than one method to reach key personnel during a disaster. Certainly cell phones are useful but they too are subject to cell tower damage and other technical problems. Therefore, have a minimum of three unrelated methods that employees can use to check in and get instructions.
     
  4. Ensure employee access to data
    Rather than depend on dongles, CDs and other physical hardware, consider using a disaster recovery vendor that provides a dark site—that is a web site that is not visible on the general internet and is not activated until a disaster occurs—where employees can check in to get instructions or access company information. Dongles and CDs and the like can be lost or simply be beyond reach if an employee cannot get to their office, home or car—or wherever they stored the item. Additionally, updating information on these tools is awkward and difficult to accomplish on a regular basis, even more so after a disastrous event. This means whatever data is stored on these tools is likely to be out-ofdate. By accessing a dark website, information is current and everything is accessible to the employee (according to their clearance level).
     
  5. Take a holistic approach
    Physical and virtualised desktop recovery environments have their limits. A combination of the two means you can minimise most known risks. Plan a physical disaster recovery site with travel routes and mass transportation in mind. Think hard about how your employees can get to the site before you choose a location. Expect the “perfect properties” to be unavailable as other companies will be looking for the same optimum sites. Instead, you may want to consider a vendor that has multi-tenant disaster recovery centres already built. Don’t stop there, however. Also plan how your people can work from home or while mobile. Adding the flexibility of fixed, remote and mobile workstation options will ensure the best outcome for your company.
     
  6. Choose standardised technologies
    Whether you build your own disaster recovery office or obtain one through a vendor, you want to use standardised technologies in your DR plan so that parts are easier to find and repairs are easier to make in an emergency environment. By comparison, many proprietary technologies add obstacles to recovery efforts. That is not to say you should choose simplistic technologies, however, as many of your business functions are complex. Rather, it is to say that, as a general rule, standardised technologies are the better choice for crisis management. You will also want technologies that pose no compatibility issues. Make sure your DR plan calls for standardised technologies for these reasons.
     
  7. Think employee retention
    Call centre agents and other typically low wage workers are often key to a company’s operations. However, low wage workers are hard to retain in the best times; harder still in times of disaster. The motivation to travel far from home, for example, is muted or absent. Or, other factors may affect these employees through no fault of their own. For example, babysitters and elderly care may not be available during a disaster, leaving the worker with few options. Plan for this by adding flexible work options. For example, you may want to allow these employees to work from home or from satellite workstations throughout the area rather than requiring all key employees to travel to a central location. Whatever you decide, plan it now and make the options and the processes known to key employees in advance.