Cloud

Conclusion: One IT-as-a-Service strategy remains to migrate legacy systems to SaaS to reduce cost, improve service level and achieve excellence in end user experience. However, large-scale ERP SaaS migrations are still not imminent, primarily due to the significant ERP customisation made by Australian organisations during the last twenty years, which prevent the use of standard SaaS architecture without re-engineering the business processes. However, it is worth noting that there are third party ERP maintenance and support services, which used in the short term may result in up-to 50 % reduction in the current yearly maintenance and support cost.

Conclusion: With the local availability of VMware’s Infrastructure as a Service (vCloud Air), all Australian VMware customers should consider it for self-service dev/test environments, virtual desktops, and more importantly DR as a Service (DRaaS). Savvy CIOs will use low risk, low cost practical experiments to develop in-house skills and experience while delivering new capabilities to the business that leads to real adoption of IaaS over the next 18 months.

The risk to CIOs who do not start adopting IaaS is that IT staff and/or business units embark on their own projects in an uncontrolled fashion leading to IT fragmentation and loss of control over the IT strategy.

Conclusion: IT organisations adopting IT-as-a-Service strategies tend to acquire the best of breed services from the market instead of building them in-house. This leads to increased adoption of multi-sourced services, whereby reliable governance processes are critical success factors to realise the desired business benefits in a timely and cost-effective manner.

Conclusion: the Department of Finance has produced a Cloud Policy that is linked to a paper about Cloud implementation that does not mention modern Cloud architecture, which in turn is linked to an architecture paper that does not mention Cloud.

Agencies looking to adopt Cloud services are advised to look for advice beyond the Australian Government’s Cloud Policy and its supporting documents.

Conclusion: in the publication ‘Running-IT-as-a-Service part 4’, IBRS defined how Service Value Agreements can be constructed by correlating business performance metrics with IT service levels. This note describes how Service Value Agreements can be constructed by aligning IT service levels with business service levels and processes. As a result, meeting or exceeding SLA targets will demonstrate the IT organisation’s contribution to business performance improvement and cost reduction undertakings.

Conclusion: the adoption of Cloud-based applications and data, the proliferation of mobile devices (i.e. Smartphones and Tablets) and the increased interest in BYOD is driving a radical change in end user computing. The old device-centric model, based on a stateful Windows desktop, is being replaced by an application-centric model where device state is transient. While this is not yet the end of the Windows desktop, the beginning of the end has arrived.

Conclusion: IT organisations adopting IT-as-a-Service practices are often challenged by limited resources to meet service demands, especially in the IT Operations space. IT operations groups should develop supply/demand models that link to business priorities and ensure funds allocation. These models will enable IT organisations to meet client necessities, clear workload backlogs, and set the foundation for effective resource management methods.

Conclusion: with the increased adoption of SaaS for business systems (e. g. ERP), new SaaS providers continue to appear in the market. While those providers are offering easy-to-use products and low start-up costs compared to running in-house business systems services, there is a risk that some service providers might cease to do business. As a result, SaaS clients will be at risk recovering services on time and without data loss. To address this issue, several escrow services have been evolving. IT organisations wishing to migrate critical services to public SaaS should explore escrow1 services. Unfortunately, escrow service costs have to-date been fully absorbed by the buyer. In this light, IT organisations should incorporate the escrow services cost into the SaaS migration business case.

Conclusion: When moving from traditional on-premises IT to Cloud it is important to update the Business Leaders and Executive on the risks. Rather than try to quantify the absolute risks, as the first step in gaining acceptance, explain how the risks of Cloud compare with the current on-premises, or MSP, solution. Offer ideas on risk mitigation that might be necessary and liberally apply simple examples and analogies to aid comprehension.

Conclusion: Running IT-as-a-Service requires offering broad IT services tied to external-value that goes beyond meeting or exceeding SLA targets. This is because the majority of existing SLAs are IT centric and vaguely relate to business value. Much of this issue is related to IT Groups’ lack of business analysis skills and IT ad hoc methods to comprehend business strategic requirements. As a result, business lines perceive IT as a support function instead of being a strategic business partner.

Many Australian IT organisations have been implementing Configuration Management practices since 1994. However, with limited success when assessed against the key objectives of Configuration Management process and its associated database (CMDB).

Conclusion: Once an organisation decides its on-premises IT infrastructure model must be transformed into a Hybrid Cloud model the important question becomes “how is this best achieved?” While Cloud Native applications and Dev/Test infrastructure are the typical first steps they do not address the Enterprise applications that are central to most enterprises.

An emerging transformational strategy is one based on Disaster Recovery as a Service (DRaaS). This is a low cost, low risk, incremental approach to transforming on-premises IT infrastructure into a Hybrid Cloud infrastructure. The DRaaS leaders in Australia will be VMware, Microsoft and AWS in that order.

Conclusion: Cloud migration should not be a quick and dirty job just to upload the current business systems with their inefficiencies, only to get rid of the in-house hardware ownership. It should be considered as an opportunity to clean IT and business inefficiencies at the same time. IT organisations wishing to migrate to public Cloud require a new methodology to avoid incurring unforeseen consumption cost and to address business processes overheads. Strategies are needed to measure code inefficiencies and develop a remedy roadmap whilst building the case for public Cloud. Only efficient code should be released to public Cloud unless there are other benefits which make the overall migration cost-effective. This will ensure IaaS usage remains within IT budget.

Organisations migrating to the Cloud and embracing flexible user-based computing have been tied up in knots with Microsoft’s archaic licensing models. On the end user computing side of things, a quick review of my notes on the nuances of Virtual Desktop Access (VDA) licensing and Remote Desktop Services (RDS) licensing are enough to give most people a brain aneurism.

Conclusion: To reduce Service Desk costs and improve resources scheduling, some IT organisations are exploring the potential of Virtual Service Desk Agents to either improve self-service and/or reach to the right subject matter expert at the right time. However self-service success depends on the quality of information available to the virtual agents. It is critical for the virtual agent tool to be enabled by a mature service management engine that describes the service’s known errors and their resolution alternatives. Failure to do so will leave the virtual agent with no alternative but to call the live agents, thereby making the investment in virtual agent technology questionable.

While hyper-scale vendors have been a little slow in opening data centres in the Australian market, the anecdotal evidence is the take-up is very strong:

Conclusion: When implementing enterprise Cloud services, a disciplined and locally distributed approach to user acceptance testing in combination with real-time dashboards for test management and defect management can be used as the centrepiece of a highly scalable quality assurance framework. An effective quality assurance process can go a long way to minimise risks, and to ensure a timely and successful rollout.

Conclusion: IT organisations developing IT policies in isolation from business units1 will face challenges to tie policies to business drivers and limit policies acceptance rate. IT organisations should formulate policies by involving business units at an early stage in policy scope discussion. IT best practices2 should be leveraged to develop reliable and practical policies. The resources needed to develop the new policies should come from both sides and a business benefits realisation plan should jointly be developed and tracked.

Conclusion: Determining the optimum licensing mix involves not only an understanding of Software Assurance, but also consultation with the organisation’s business strategy groups, as well as a firm understanding of potential structural changes, such as mergers, de-mergers, acquisitions, and growth strategies. Getting the wrong mix can result in overspend, or worse, an inability to adopt business strategies such as mobility, activity based working, or bring-your-own-device.

Related Articles:

"Understanding and Optimising Microsoft Software Assurance: Part 1 – The Basics" IBRS, 2014-10-01 20:28:23

Conclusion: Business-centric IT strategies are critical to run IT-as-a-Service1 because they attempt to integrate IT with business strategies. The rationale is to support business operations by implementing new technologies that reduce business risks, create business opportunities and achieve high levels of customer satisfaction.

Business-centric IT strategies focus on addressing the business critical issues by implementing new IT solutions in a timely and cost-effective manner. The proposed IT solutions should provide capabilities that address the current and emerging market forces such as consumerisation, mobility, social media and Cloud. This will signal to business lines that IT is being modernised to meet consumers’ exigent needs.

It is critical for business-centric IT strategies to be developed within two months to accelerate IT-as-a-Service transitioning.

Conclusion: To improve business performance and/or reduce the cost of doing business, forward-thinking IT organisations are trying to run IT as a Service. However, they are challenged by long software implementation timescales, fragmented delivery processes and insufficient skilled resources to meet business demands.

To address these challenges, IT organisations should emulate the commercial practices related to delivering quality IT solutions at reasonable costs.

Conclusion: There are many different Hybrid Cloud approaches, each with different costs, risks and benefits. Organisations should evaluate the alternatives to find which is best aligned to their business requirements, then update IT governance processes to guide the organisation towards the chosen Hybrid Cloud strategy. Failure to align to the right Hybrid Cloud strategy will either result in the creation of new IT silos, which becomes a barrier to the business strategy, or will adopt an approach that stifles business innovation and agility.

Conclusion: Failure to embrace the SMACC stack (Social, Mobile, Analytics, Cloud and Consumerisation) will result in the wider organisation working around the internal ICT provider. Job losses in the ICT team and a reduction in wider corporate capability will result.

Conclusion: While many IT organisations believe that using public IaaS (e.g. AWS, Microsoft Azure, Google) to host business applications is a cost-effective strategy, the lack of IaaS usage planning will most likely increase consumption cost. IBRS recommends that IT organisations undertake a self-assessment of their usage management practices prior to migration to public IaaS1.

Conclusion: IT organisations' lack of IaaS usage planning will most likely increase consumption cost. As a result, IT organisations should work closely with business units to understand usage patterns and track monthly usage against forecasts. This will more likely ensure that IaaS usage levels remain within budget. This note provides the usage management framework. Part 2 planned for release in August 2014 will provide User Management maturity self-assessment approach.

Conclusion: A majority of organisations around the world and across Australia are implementing or trialling some form of Cloud service whether it be IaaS, PaaS or SaaS. While Cloud services offer many potential benefits to organisations they can increase complexity in a number of areas of IT service management. Organisations may implement a hybrid Cloud model and deliver some services using public or private Cloud.

Business areas may subscribe to Cloud services for the provision of application services with or without the participation of IT. Identifying and managing the schedule of change with a wide variety of providers can be complex but will provide the CIO and the organisation with a clear view of who, what, when, and how changes will be made, the risks involved and the mitigation actions required.

Conclusion: Amazon Web Services (AWS) dominates the IaaS market, witha commanding market share lead over all other vendors. Since there are no clear market forces that will change this in the next few years the question is who will become second and third?

Conclusion: Moving services to the Cloud is a part of nearly all organisational strategic plans. Organisations today are either starting to trial services with one provider, moving from the trial phase to include additional services or heavily focussed on Cloud as part of their service delivery model.

Based on the learnings from organisations that are heavily focussed on the Cloud, CIOs will only be successful if they can successfully develop their maturity to be a competent customer of Cloud services. Developing an objective view of your organisation maturity level and actively seeking learnings from organisations that have already undertaken the journey will assist CIOs in developing an appropriate, actionable plan for their organisation.

Conclusion: WhileI SaaS and PaaS adoption has been increasing during the last two years, most IT organisations are hesitant to migrate their legacy systems to public SaaS. This is primarily due to the applications being highly customised to support the current business mode of operations. As a result, migration to cloud requires significant effort to retrofit the existing systems in public SaaS architecture. One of the options to address the customisation obstacle is to adopt a rapid business process redesign approach.

Conclusion: The proliferation of mobile devices and increasingly mobile staff in the enterprise is driving demand for file sharing and synchronisation services. In the absence of a usable offering from the organisation, users are turning to the ad-hoc use of consumer grade services. This is often referred to as ‘The Dropbox Problem’.

Failure to provide a workable enterprise alternative will increase organization’s risk of data loss or leakage.

While IBM is planning to invest A$1.4 million to grow its global datacentre facilities, its focus remains on private cloud with no serious public cloud offerings, As a result, IT organisations under traditional outsourcing contracts with IBM should examine the feasibility and cost-effectiveness of third party public cloud alternatives prior to renewing the existing outsourcing contracts.1

Conclusion: With the migration to complex hybrid sourcing strategies, traditional IT organisations based on ‘plan/build/run’ models won’t be suitable for acquiring public cloud services in an increasingly changing market. This is due to vague understanding of service total cost of ownership and limited contract negotiation skills. IT organisations wishing to rely on external services must evolve to ‘plan/procure/govern’ structure to emphasise strategic service planning and hire specialised service procurement skills. This paradigm shift requires CIOs to restructure IT procurement with a view to run it as-a-service to other IT groups and business lines.

Conclusion: The rapid adoption of SaaS by HR departments is a herald of the way IT departments will need to reinvent themselves. SaaS means that IT will not control everything, but there is an important role for influencing how the organisation adopts. IT needs to proactively develop relationships with business units in order to establish itself as a partner to business and create an environment which will ensure that advice is sought early, and not only when integration issues arise. This mind shift, from utility to partner, is critical but may not come naturally to the mindsets that have been so good at running traditional IT departments.

Conclusion: IT organisations wishing to maximise Public Cloud return on investment should adopt a Cloud Governance Maturity Model that ensures consistent delivery, builds trust and leverages new technology. This will enable IT organisations to effectively manage their sourcing portfolio by balancing cost and risks, creating value and realising the desired benefits.

One of the most common and contentious infrastructure discussions that I have with clients today is the Public Cloud. The views of most IT executives have shifted from “if” to “when, what and how”. Like other big IT shifts, whether it be Mainframe to Midrange, the PC or Unix/RISC to virtualised Intel, there is much wailing and gnashing of teeth by the old guard about what will, frankly, become the new normal within the next 5-7 years.

Conclusion: In-house IT Service Management (ITSM) initiatives require considerable time and investment (up-to three years, up-to $1.5 million approximately). This has resulted in limited senior management continuous buy-in and reduced ITSM benefits realisation. Therefore, IT organisations wishing to implement ITSM should evaluate a public cloud alternative versus the cost and merits of establishing in-house service management capability.

Conclusion: Some SaaS service providers can exercise ‘exit for convenience’ contractual terms by giving no more than thirty days termination notice. As a result, SaaS users will be at a high risk to recover services on time and without data loss. Therefore, IT organisations wishing to migrate critical services to public SaaS should develop a Contingency Plan and test it regularly. The Contingency Plan establishment cost should be incorporated into the business case for public SaaS migration.

Conclusion: IT organisations wishing to migrate in-house services to public cloud should ensure that service providers understand the complexity of the in-house architecture candidate for cloud migration. This can be achieved by identifying the in-house service failure points within the legacy applications and their associated infrastructure. The service providers’ lack of understanding of the existing operational weaknesses will most likely extend the transition period and delay achieving the expected service levels in a gradual and cost-effective manner.

Conclusion: IT organisations managing a multi-sourced environment and wishing to reduce unscheduled service downtime, should establish end-to-end Change Management Frameworks. This will ensure that business operations remain unaffected by service providers’ system changes.

Conclusion: Driving value from Infrastructure as a Service (IaaS) requires more than just a technical evaluation. IT Organisations must get clear understanding of the features and benefits of the billing model and how these are aligned to, and can be used to drive, the business’ objectives (e.g. faster time to solution, rapid scale-up and down, infrastructure costs to usage and revenues).

Achieving this understanding will require IT organisations to elevate the evaluation of the IaaS billing model to the same level of consideration as other key non-functional requirements such as availability, recoverability, and security. Organisations that fail to do this may find themselves locked into costly, inflexible IaaS contracts that are not aligned to the business objective and which fail to deliver the full potential of the Cloud.

Conclusion: Given that the public cloud value is maximised when end-to-end SaaS is reached, IT organisations’ misunderstanding of SaaS building blocks, business applications architecture integration and lack of mature multi-sourced environment governance will limit SaaS public cloud adoption. CIOs should establish a cloud sourcing strategy to assess the feasibility and cost effectiveness to gradually migrate business applications to public cloud. Failure to do so might minimise public cloud opportunity to improve enterprises’ performance and/or reduce the cost of doing business.

Conclusion: Infrastructure as a Service (IaaS) is starting to be used in dev/test and production environments by many ANZ organisations. CIOs who get IaaS right will create significant benefits for their organisation, both in cost and agility, and greatly improve the perception of IT with their peers in the organisation.

However, a general lack of experience with Cloud, and the use of outdated infrastructure purchasing approaches, may result in poor IaaS contracts that can cost the organisation hundreds of thousands, if not millions of dollars per year in excess fees. To combat this, CIOs should adopt a just-in-time approach to IaaS, eliminating vendors that lack the contract terms, or depth of infrastructure resources, to accommodate this approach.

Conclusion: In a few years from now the Cloud services we use today will look as quaint as the highly static Web of 1997 in the rear view mirror. In the wake of the global financial crisis the hype around big data is still on the increase, and big data is perceived as the new oil of the economic engine. At the same time, many of the data management technologies underlying Cloud services are being consolidated, creating new kinds of risks that can only be addressed by the adoption of a different data architecture.

Conclusion: Cloud infrastructure and platforms have started to alter the landscape of data storage and data processing. Software as a Service (SaaS) Customer Relationship Management (CRM) functionality such as Salesforce.com is considered best of breed, and even traditional vendors such as SAP are transitioning customers to SaaS solutions. The recent disclosure of global Cloud data mining by the US National Security Agency (NSA) has further fuelled concerns about industrial espionage in Europe and has significantly raised citizen awareness with respect to privacy and data custodianship. Any realistic attempt to address these concerns requires radical changes in data architectures and legislation.

Conclusion: While on-premises is still the dominant IT delivery model, Cloud is increasingly viewed as a robust complement or alternative. When evaluating new IT system and services ensure IT staff evaluate the use of Cloud as an alternative delivery model. The evaluation should include non-cost benefits, such as time-to-solution, rapid scale-up and scale-down, pay-as-you-go, as well as traditional metrics such as Total Cost of Ownership (TCO) and risk.

Rather than ask “Should we move to the cloud”, IT executives should ask “Why, What and When”, and then use these three questions to create a guidelines for comparing Cloud as an alternative delivery model to on-premises.

Conclusion: VMware’s new strategy of directly entering the IaaS market will create confusion and ultimately decimate VMware based IaaS vendors. IT organisations should manage the risks this creates with their current (or future) VMware based IaaS partners. In the long run the new strategy will benefit all customers by creating a global scale, VMware based, IaaS that reduces costs, increases service quality and drives greater innovation.

VMware does not lead the IaaS market and faces massive competition from non-VMware based hyperscale IaaS providers, such as Amazon, Rackspace and Microsoft. VMware centric organisations should not blithely assume that a VMware based IaaS is the best option and should evaluate the IaaS alternatives from hyperscale providers, especially Amazon.

The Cloud is a significant long-term trend that should not be ignored.Like the introduction of the PC and Open Systems in the ‘80s/‘90s,an IT organisation can either selectively embrace the Cloud, orfind itself bypassed by the business units who will introduce Cloudbased solutions to suit their needs.

Organisations that do not embrace the cloud risk losing control ofthe IT Architecture, which leads to an overly complex, cost andineffective environment. Even worse, while individual business unitsmay gain some temporary benefits, the overall organisational agilitywill decrease and the alignment of IT to strategy breaks down,creating longer-term problems for the organisation as a whole.

On the other hand, if the Cloud is selectively embraced as yetanother IT sourcing strategy, and if best practice IT managementfunctions are retained and expanded to provide appropriate governance,the Cloud can be a positive agent for change that increasesagility and creates greater transparency in cost.

Conclusion: DevOps is a grassroots movement that is only a few years old but has quickly spread across the globe, and its influence is present in virtually all organisations that operate popular Cloud services. DevOps is a portmanteau of software system Development and Operations, referring to the desire to bridge the gap between development and operations that is inspired by agile techniques, and that is driven by the need to continuously operate and upgrade Cloud services. The DevOps movement is having a profound impact in terms of the tools and techniques that are used in the engine rooms of Clouds, leading to order of magnitude changes in the ability to perform hot system upgrades.

Conclusion: Cloud services are not similar to a highly virtualised internal environment. Nor are they similar to the tightly controlled experience of time-sharing on a mainframe back in the 1970s. The supposed elasticity of the cloud has become a point of vulnerability because the elasticity is only partial, and only at certain points. The outcome is a service which is believed to be highly resilient, but which can actually prove to be surprisingly brittle.

Conclusion: Early adopters of cloud services often swept aside security and risk concerns, as these adopters were more interested in the end – a better IT service – rather than the means. But now organisations with mature risk and governance processes are looking at cloud services and risks are being identified and assessed for their potential impact. Cloud services can dramatically improve the IT service experience of an organisation, but organisations must be completely clear on what services they are, and are not getting as part of the engagement. As with all commercial engagements, the devil is in the detail.

Related Articles:

"How do you catch a cloud and pin it down? Part 1" IBRS, 2012-05-28 00:00:00

Conclusion: Cloud offerings, particularly Software as a Service, have many technical risks to iron out before they are palatable for any organisation that has a mature governance requirement. The vendors know this, and because they don’t want to raise these issues in the minds of less mature organisations, their master subscription agreements are typically thin on details or accountability, and heavy on indemnity. Given these unaddressed risks, CIOs should ensure that business executives are informed of the potential for service failure, as well as the implications and potential business impact.

Related Articles:

"How do you catch a cloud and pin it down? Part 2" IBRS, 2012-07-27 00:00:00

Conclusion: IBRS has observed that many organisations struggle with mobility, implementing fragmented mobility solutions that service narrow areas of the business. This leads to higher complexity and costs, and lower levels of user satisfaction. Instead, organisations should take time to build a holistic mobility strategy, driven and grounded by use-cases, and shaped by a concise set of use-case categories.

Related Articles:

"Coping with Mobility - Part 3: aligning generic use cases to application development approaches" IBRS, 2012-04-30 00:00:00

"Coping with Mobility - part 1: mobile architecture and the enterprise" IBRS, 2012-02-28 00:00:00

"Coping with mobility - part 4: governance" IBRS, 2012-05-31 00:00:00

"Coping with mobility - part 5: developing the strategy" IBRS, 2012-10-28 00:00:00

"Coping with mobility Part 6: Work context" IBRS, 2013-06-26 00:00:00

Conclusion: Organisations looking to develop and deploy mobile applications must realise that the mobile device market will undergo significant change over the next five years. This creates serious challenges for developers within enterprises, who must either create applications for specific devices and different form factors, or attempt to develop cross-platform applications that will still meet end-user expectations. IBRS has identified four high-level architectures for developing mobile applications, each of which has specific strengths and weaknesses. This research provides an overview of these architectures.

Related Articles:

"Coping with Mobility - Part 3: aligning generic use cases to application development approaches" IBRS, 2012-04-30 00:00:00

"Coping with Mobility - Part 2: First steps towards a holistic mobility strategy" IBRS, 2012-03-29 00:00:00

"Coping with mobility - part 4: governance" IBRS, 2012-05-31 00:00:00

"Coping with mobility - part 5: developing the strategy" IBRS, 2012-10-28 00:00:00

"Coping with mobility Part 6: Work context" IBRS, 2013-06-26 00:00:00

The Cloud is a significant long term trend that you ignore at your peril. Like the introduction of the PC and Open Systems in the ‘80s/‘90s, you can either selectively embrace the Cloud or find yourself bypassed by the business units who will introduce Cloud based solutions to suit their tactical needs and political agendas.

Unless you embrace the cloud, albeit in a controlled and limited way, you risk losing control of the IT Architecture, which will lead to an overly complex, costly and ineffective environment in the long run. Even worse, while individual business units may gain some temporary benefits, the overall organisational agility will decrease and the alignment of IT to the business will break down, creating longer term problems for the organisation as a whole.

Three recent events have cast serious doubts on the viability of public cloud computing in the Australian marketplace. These events have raised critical concerns about the security, reliability and regulatory aspects of emerging cloud platforms in both public and private sectors.

Conclusion: While it appears that every known test to evaluate cloud computing has been done, there are two which determine the accuracy of any savings claimed. Indeed, they could be applied to any evaluation of IT savings and not the cloud alone.

To a large degree the tests discussed here challenge some processes of cost assessment, but IT executives ought to look for better ideas and arguments. It should be possible to ask questions of consultants and vendors in order to obtain better answers.

Conclusion: Analysis of Microsoft’ recently announced licensing model for education suggests that up to 60% savings are possible for K-6 schools, with 30% savings for 7-12 education. Furthermore, Microsoft’s new cloud-based offerings provide similar opportunities for licensing rationalisation. Educational organisations planning desktop migration must carefully assess these new licensing and deployment options in order to gain the most advantage of Microsoft’s new licensing models. The licensing costs involved also raise questions regarding the pedagogical value of take-home netbooks.

Conclusion: Demand for storage capacity continues to grow at 60%+ per annum, requiring ongoing capital investments in incremental capacity upgrades, or worse, a capital intensive rip and replace upgrade every 3-4 years. Since “cloud” is the current IT buzzword, IT organisations are being asked to look at how the use of cloud storage can reduce cost and transform lumpy capital expenditure into a more uniform “pay as you go” operational cost.

In dealing with the many issues around the cloud it will take a delicate balance of political skill backed by a strong communications strategy to negotiate and collaborate with business. Offering informed and contextual guidance in an open minded discussion is a strong position to adopt. Technology managers should reflect on, and if necessary, modify how they are managing the cloud, with their business colleagues. In some cases a formal approach may be required: presentations, roadmaps, evaluations and information packages delivered to a business audience. In many other cases, a revised approach may be informal, and involve a collaborative attitude to enable an organisation to make better choices.

Software, and or, middleware solutions to enable collaboration and other activities via the cloud are growing. Fortunately for organisations, the features, benefits and costs of the solutions offered by vendors provide plenty of choice. As this area of cloud applications and tools is growing rapidly, organisations should take a longer view, up to three years for a total cost of ownership evaluation, and assess the requirements for file sharing solutions. Social networking connection tools can be chosen without too much cost or risk attached.

Conclusion: In recent months several Tier-1 Australian and New Zealand vendors have announced, and in some cases delivered, locally hosted Infrastructure as a Service (IaaS)1. These announcements will reduce Business and IT Executives’ perception of the risk of adopting IaaS, and result in greater interest in using cloud as a “lower cost alternative” to in-house infrastructure.

While the cloud is often assumed to be an inexhaustible supply of low cost virtual machines, that are available on a flexible pay-as-you-go model, organisations that have looked beyond the hype found it was not as cheap, or as flexible, as you might think.

The cloud computing economic model is expected to bring significant rewards – apparently. Those rewards may be possible, but the quality of analysis to demonstrate that the cloud paradigm will yield an ever-growing margin is far from assured. The assumptions underlying the economics of the cloud are tenuous and therefore the promotions and promises should be treated with caution. More analysis has to be done to evaluate how and where advantages are achieved, and at what cost and margin. Without sufficient rationale, empirical data and analysis, the hype will burst once the ambiguous and unclear economic outcomes emerge, just like other technology bubbles before.

One of cloud computing’s apparently key advantages is reduced operational costs. On deeper investigation, however, the purported savings are achieved by removing obvious waste, which represent the bulk of the headlined ‘savings of 50%’ that cloud computing allegedly offers.

Conclusion: Defining the Cloud is proving to be elusive. This is because vendors are trying to neatly define Cloud around their products and services. This creates competing, product based definitions for what is actually an aspiration to create a “better IT environment”.

Viewed as an aspiration, the Cloud becomes a journey to create an IT environment with specific characteristics, such as cost transparency, utility pricing, capacity on demand, commodity pricing, self-service, location and device independence. Like all journeys there are many different paths which depend on where you start and where you want to go.

Conclusion:Organisations that that plan to deploy or extend their WAN Optimisation Clients (WOCs) should strongly consider virtual WOCs. Virtual WOCs will carry less financial commitment, and an organisation deploying virtual WOCs will not be encumbered after 2-3 years with outdated appliances which cannot be repurposed. The importance of not over-committing to WOC appliances will become increasingly important: as WOC capabilities get baked into application-specific integrated circuits (ASICs); and as organisations move towards web applications, which will require their own accelerators.

Conclusion: vCloud Express is a new entry level Infrastructure as a Service (IaaS) offering based on self-service portals, credit card payments and VMware’s enterprise class virtualisation products.

CIOs should look at vCloud Express as a low cost, low risk way to learn how to use public cloud infrastructure. Since vCloud Express may be seen by some groups (dev/test, business units) as a way to side-step the perceived bureaucracy of the IT Organisation, CIOs should develop a strategy to embrace this use as a way to retain control and ensure relevancy with dissatisfied customers.

Conclusion:Cloud computing is promoted as the next disruptive technology in the organisational use of IT. If this does happen, no matter what else changes there are some verities which must not change, in particular meeting legal requirements. There are at least seven areas where a move to cloud computing should not be contemplated unless the legal requirements can be demonstrably satisfied.

Conclusion: Cloud computing is not a new environment, merely the extension of a number of technologies that support IT outsourcing (bureaux, ASPs, SaaS, IaaS, PaaS). Cloud computing is technology-driven but will be and is likely to become “the next disruptive technology” in sourcing. Rightly or wrongly, many have been jumping on the cloud bandwagon because they’ve been driven to reduce costs. However, many potentially significant legal problems and their consequences have yet to be addressed. These legal issues extend beyond, and can be more complex than those that apply in traditional outsourcing agreements. Organisations considering outsourcing business applications to cloud computing must consider all relevant legal challenges at the same time as they explore the technologies.

Conclusion:With infrastructure vendors jumping on the cloud bandwagon, their sales and marketing teams are increasingly using the terms “Cloud”, “Cloud Computing” and “Infrastructure Cloud”. From discussions with clients we have observed these terms are not well understood and mean a wide range of different things to different people.

This confusion is driven by a war between vendors to establish a definition of these terms that best suits their specific products, technologies and architectures. Until “Cloud Computing” and “Infrastructure Cloud” become commonly defined, which we expect to take at least until the end of 2010; be careful to define what you mean, and seek to understand what others mean by these terms to avoid significant misunderstandings between staff, vendors and partners.

Conclusion:Privacy and data protection laws in Australia and NZ hold organisations, rather than their subcontractors, responsible for the activities of their subcontractors. Before committing to outsourcing any corporate data to a cloud computing vendor any organisation must ensure that all relevant legal constraints are agreed and in place so as to avoid any subsequent litigation. Ensuring and monitoring this may not be easy.

Conclusion: Before any organisation outsources any of its operations to a cloud computing provider it must be fully cognizant of, and have addressed to its satisfaction, the many potential legal problems and their consequences. These extend beyond, and can be more complex, than those that apply in traditional outsourcing agreements. Organisation must ensure that all legal issues have been addressed before committing any core systems to a cloud environment.

Conclusion: Proprietary web services are raising concerns about strong lock-in. Those raising the alarm bells paint a simplistic picture based on the assumption that services such as Facebook are representative of the web service landscape. Upon closer examination it appears that the doomsday prophets have a vested interest in prolonging the use of localised IT infrastructure. In reality the concept of web services opens new possibilities to unbundle and mitigate lock-in, allowing internal IT to focus on the core business and to outsource the operation of non-core functionality.

Conclusion: There is still more hype in the media about cloud computing than uptake. Advocates promise dramatically improved ease of use, lowered costs driven by economies of scale, and much greater flexibility in sourcing and adapting to change. Nicholas Carr in his latest book2, predicts that cloud computing will put most IT departments out of business. "IT departments will have little left to do once the bulk of business computing shifts out of private data centres and into the cloud," Such arguments make it likely that organisations will increasingly place some or all of their IT supported services in “the cloud”. This makes these organisations dependent on the reliability of the vendor’s cloud offerings. If an organisation moves all or part of its IT services to a cloud environment it must first identify and understand the new risks it may be exposed to.

Conclusion: Business departments not getting the services they want from their IT departments see Software as a Service (SaaS) as a very attractive alternative to get applications currently not being provided or to replace unsatisfactory or poorly supported applications. CIOs must be prepared to find (possibly renegade) SaaS applications in use in their organisations. They must set in place the relevant support and appropriately skilled staff to manage this potentially disruptive technology.

Conclusion: Although Microsoft’s new range of servers products have been available for some months, it is only now that the software giant’s marketing engine will go into full throttle. Understanding Microsoft’s likely marketing strategy that will endevour to shift the market from virtualisation to what it calls “Dynamic IT,” and being prepared for this strategy’s ramifications and the pressures it will create for IT managers will be essential to managing expectations from the business and smooth technology deployment.

Conclusion: A perfect IT storm is looming, driven by merging category 4 storms such as Utility (or cloud) computing, and the Red Shift growth in massive computing. The force of the storm will be exacerbated by rising energy costs and their impact on the data centre energy budget. As a consequence, in a few years many mid to large organisations have at least all their non-differentiating applications running on remote shared SaaS-like sites. This will have a significant impact on the IT department and it’s CIO.

Conclusion: For most corporate IT departments, concepts such as Cloud Computing seem light years away from current day-to-day reality. Yet the number of commercial providers of such services is growing fast, and even more far-fetched ideas such as global software service supply chains are emerging on the horizon. The distance between innovators and late adopters of modern techniques and technologies is growing. In this scenario it is essential to know when not to remain amongst the late adopters, to avoid being left behind in the dust and struggling with evaporating profit margins.

This article is the first in a series of three on technologies and techniques that are leading to fundamental changes in the architectures used to construct software applications and software intensive devices. First examples of these changes are already visible today, and over the next five years, many of the current rules for architecting business applications will be re-written.

Related Articles:

"The Industrialised Web Economy - Part 3: Automation and Model Driven Knowledge Engineering" IBRS, 2008-05-28 00:00:00

"The Industrialised Web Economy- Part 2: Software Supply Chains" IBRS, 2008-04-28 00:00:00

Conclusion: IT managers planning business-to-business integration, or with the need to couple old-school EDI (Electronic Document Interchange) and legacy ERP (Enterprise Resource Planning systems) with modern web-based architectures must look towards a uniform message-based middleware infrastructure. If the organisation is already moving down the .Net deployment path BizTalk R2 is now a contender along with the more traditional products, such as Tuxedo and Tibco.

Conclusion: While wikis are certainly an important new approach to information management, they should not be considered as a replacement for enterprise content management systems (CMS). Instead, wikis should be considered an adjunct to content management, providing added flexibility and collaboration where needed. Understanding the differences between CMS and wikis is vital.

Conclusion: Legal firms acting on behalf of copyright holders of images are sending out thousands of copyright infringement notices globally, seeking hefty fees from companies whose web sites contain copyrighted images. CIOs need to understand some of the legal nuances and be directly involved in planning and implementing measures to minimise the financial risks of web content. 

Utility Computing is gaining higher levels of customer interest and acceptance, amid significant investments by systems vendors to build on their individual "brand" (e.g., Adaptive Enterprise, On Demand, Real Time Infrastructure) and the resulting cacophony of terms, definitions, and strategic directions. However, there still remain several crucial "missing-technology-links" in the evolutionary chain -- these will be addressed by both traditional software and systems vendors, as well as Open Source providers.

Widely-available, relatively cheap technology is catching up with the long-standing desire of end-users and businesses to use and pay for technology as demand arises, rather than them being forced to buy entire software packages or infrastructure, and then use just a small percentage of the overall capability. At the same time, user business environments have become more nimble, requiring more flexibility in IT delivery and usage, in licencing and payment structures, and in vendor business models. By being aware of new provisioning models, users will be able to gain long-sought improvements in costs and service.

''On-demand''—or ''adaptive,'' ''agile'' or other terms being used by major vendors includes the combination of business models, processes and operations that are enabled by and which require these IT resources. This concept is therefore more of an overall business strategy, including not just the availability of IT resources ''on-demand,'' but the ability to build, change, adapt and manage business operations using and leveraging the ready availability and variable capabilities of utility computing.

The next ''Big Thing'' in business computing is more about new IT deployment, sourcing and management models than about the evolution of technology itself. Vendors such as IBM and Hewlett-Packard have continued to fine-tune products and services that were announced in 2001 to a point where they are now viable choices for technology buyers. Given labels such as ''On Demand'' by some, or the Adaptive or Agile Enterprise by others, they can all be considered part of the Utility Computing paradigm.

Conclusion: IT organisations wishing to select quality services at competitive prices should rate themselves against an IT procurement maturity model to leverage economies of scale. This will enable IT organisations to reduce cost while meeting business needs in a timely and cost-effective manner.

Conclusion: It is no longer viable for telecommunication providers to simply offer Session Initiation Protocol (SIP) trunks for voice connectivity or Multi-Protocol Label Switching (MPLS) links to connect office and data centre locations. Nor does it make good business sense for the telco or for the customer.

The modern architectures of Cloud and Software-as-a-Service (SaaS), mixed with the need to maintain on-premise for critical elements are key components that support most digital strategies. Using older telecommunications architectures with fixed connections and physical infrastructure for routing and switching can be costly, and can stifle agility and therefore productivity.

However, modern telecommunication architectures bring an ability to virtualise connections and network switching. The abstraction of these capabilities allows dynamic management of the services providing substantial agility, as well as potential productivity gains and cost savings to the customer.

Conclusion:

As-a-Service machine learning (ML) is increasingly affordable, easily accessible and with the introduction of self-learning capabilities that automatically build and test multiple models, able to be leveraged by non-specialists.

As more data moves into Cloud-based storage – either as part of migrating core systems to the Cloud or the use of Cloud data lakes/data warehouses – the use of ML as-a-Service (MLaaS) will grow sharply.

This paper summarises options from four leading Cloud MLaaS providers: IBM, Microsoft, Google and Amazon.

Conclusion

DevOps, business intelligence (BI) and data, machine learning (ML) and artificial intelligence (AI) are all driving rapid change within IT departments. The challenge will be finding Cloud certified people to meet the rising demand.

Leaders have two main choices. Upskill their existing teams, or embark on a recruitment campaign that brings in Cloud certified professionals to manage Cloud migration and provide the ongoing support and optimisation needed to bring the full value of Cloud to IT operations.

For organisations who suddenly realise how far they are behind on the Cloud value curve, pressure will mount to deliver results quickly. Make sure staff are certified and ready to address your hybrid or multi-Cloud environments.