Richard Curran is the Director of the Enterprise Server and Software Enabling Group at INTEL, EMEA. Mr Curran also leads a team of Strategic Relationship Managers responsible for enabling Intel’s Global ISV and GTM activities with its key software partners. Prior to this role he served in a number of high-level posts including as co-leader of the EMEA Customer Solutions Marketing and Sales teams, and as head of Intel’s Services Organisation in EMEA. Mr Curran joined Intel from Compaq where he worked in server executive roles including as Director, Pre-Sale Centers EMEA and Director, Services Product Support & New Products, EMEA. Before Compaq, Mr Curran was Country Manager of Concurrent Computer Italy, managing day-to-day sales and services operations. Richard Curran is an Electronics Engineer and a graduate from Cork, Ireland.
Cloud computing is neither a passing fad nor a cure-all; it will grow strongly in coming years. Massive data centres, located ‘out there’ on the Web will increasingly provide virtual infrastructure, applications or software on a pay-as-you-go basis. It can deliver impressive savings and access to an astounding range of IT facilities, but there are real security concerns, the savings for some companies are not clear, and there may be significant governance and compliance issues involved.
Top-line overview The year 2008 cloud computing whipped the IT press into a frenzy. Within the space of a few months, cloud computing went from being a relatively unknown concept to the year’s most touted technology. Company after company jumped onto the cloud computing bandwagon, launching new ‘cloud’ services, in many cases just re-naming existing offerings to include this new buzzword. It was, in short, hype of the highest order. Now, calm is being restored and the IT press, as well as analysts and the industry itself, are taking a much more realistic view of the cloud. Gartner, for example, is going as far as to predict that cloud computing will go through the ‘trough of disillusionment’ - the phase in its hype cycle where technologies fail to meet expectations and quickly cease to be fashionable. Despite the hype - and unlike many technologies that have gone before it: videophones, paperless office and mobile TV to name just a few - the cloud is here to stay and will evolve over years. It will change the way people work and how companies operate, as they shift to consume services in a more cost effective way. Cloud services are particularly attractive for small businesses or start-ups who cannot afford upfront IT investment. Ultimately, though, it is unlikely that larger organisations will completely abandon on-premises models, or that they will replace IT that is core to their competitive advantage with services through the cloud. Many businesses will still require a level of security, performance or customisation in their applications that cannot yet be supported through the public cloud. For this, they may create their own private cloud architecture, behind corporate firewalls, to harness the efficiency of the cloud but with more security and control. In short, cloud computing will be neither a passing fad nor an e-business revolution. Instead, the majority of businesses will likely have a mixed IT environment, where applications, infrastructure and in some cases complete business processes will be delivered through both public and private, on-premise clouds. What is cloud computing? Everyone is talking about cloud computing, but it seems everyone has a similar, yet different definition for it, depending on their position. Someone involved in services, for example, will have a different definition for the cloud than someone working more on the architecture side. Here are definitions from Gartner and Forrester: “A style of computing where massively scalable IT-related capabilities are provided ‘as a service’ across the Internet to multiple external customers.” Daryl Plummer, Gartner “A pool of abstracted, highly-scalable, and managed computer infrastructure capable of hosting end-customer applications and billed by consumption.” Forrester We, too, have our own definition of cloud computing, preferring to talk in terms of cloud architectures and services, and making the distinction between public and private clouds: Cloud architecture: services and data reside in shared, dynamically scalable resource pools, based on virtualisation technologies and/or scale-out application environments. Cloud service: service is delivered by the public Internet to consumers or enterprises. Sitting upon a cloud architecture, service scales without user intervention and is typically billed by usage. Private cloud: cloud architecture is deployed behind an organisation’s firewall, delivering IT-as-a-service for internal use. Cloud computing can be seen as a re-envisioning of other distributed and utility models that have been around for the last decade, including utility computing, on-demand services, grid computing and software-as-a-service. What separates the cloud from traditional web services is an underlying dynamic and elastic architecture that enables IT to pay only for what they need and scale rapidly. Why has the cloud emerged now? Cloud computing is what is referred to by Gartner as an ‘emerging phenomenon’ - in other words a phenomenon that emerges when conditions are right and multiple factors are in play - in this case, service orientation, virtualisation and standardisation of computing through the Internet. Combine this with global economic turmoil and cloud services present an opportunity to reduce some IT expenses. How is cloud computing used today? The majority of cloud architecture deployments today support public clouds designed to deliver a service over the Internet, such as Google Search, Microsoft Hotmail or Salesforce.com. Large service providers, together with typical early adopters such as financial services, high performance computing and pharmaceutical companies are also deploying cloud architectures to support private clouds behind the firewall. Still in its infancy, this usage is expected to grow out of enterprise virtualisation technologies already being deployed. Public cloud services are primarily targeted at consumer services, such as Internet search, personal email, social networking and other Web 2.0 deployments. Small businesses and start-ups are also attracted to the concept of cloud-delivered services to reduce upfront IT investments. Some larger enterprises are expected to deploy public cloud services for applications deemed not business critical - such as web-delivered software to manage employee travel scheduling or HR (human resource) support suites. For many enterprises, IT infrastructure is closely linked to their core competence and outsourcing this computing capability could pose significant risks to their business. Security, reliability, performance, and compliance issues are top of mind among CIOs as they evaluate next generation architectures. For example, a biotechnology company may leverage its own servers for advanced drug simulations, as the company may be wary of running such sensitive modelling outside its own walls. To take advantage of the flexibility and efficiency of cloud architectures, they may develop an internal, private cloud, where they have greater control and security over the computing resource. Who is delivering cloud services? There are three high level categories of cloud services: software-as-a-service (SaaS) - software deployed as a hosted service and accessed over the Internet; platform-as-a-service (PaaS) - platforms that can be used to deploy applications provided by customers or partners of the PaaS provider; and infrastructure-as-a-service (IaaS) - computing infrastructure, such as servers, storage and network, delivered as a cloud service, typically through virtualisation. Infrastructure-as-a-service can provide access to a server, part of a server or storage, for example, Rackspace’s Cloud Servers. An example of a platform as a service is Amazon’s Elastic Compute Cloud (EC2). Essentially Amazon allows developers to use its virtualised, remote servers for application development and hosting on a pay-as-you-go model. Other examples of platform as a service include Google App Engine. An example of software-as-a-service is Salesforce.com, which has been providing customer relationship management (CRM) solutions using the SaaS model for nearly ten years and is thought to be one of the pioneers of ‘as-a-service’ computing delivery. Giant, IBM and Microsoft, have now made their own investments to establish cloud services, and SAP and Oracle say the next-generation of their business applications will be cloud-enabled and available as SaaS. Conclusion The cloud computing phenomenon is real, but should be seen in context. Most IT spending will continue to be on traditional server and software deployments: IDC estimates that nine per cent of IT spend will be on cloud services by 2012. Cloud architectures do provide an attractive opportunity for increased efficiency, flexibility and time to market. Whether companies are building large cloud service infrastructure or private clouds for internal enterprise use - there are common requirements of ultra energy efficiency, virtualisation capabilities and performance. Fortunately, there are products and technologies uniquely positioned to address the needs of cloud infrastructure, both now and as the industry develops this innovative computing model.