The Lippis Report

Cloud computing has become of great interest to providers, business and IT leaders as the economic downturn forced review of business processes and IT’s automation of them. As business and IT leaders searched for efficiency, cloud computing came into focus as it promises a different and favorable IT economic and delivery model. There are multiple cloud visions and use cases, but one I hear most often in corporate IT organizations is that of thin and mobile clients accessing a mix of custom, consumer-based and cloud-based applications. In this scenario real or virtualized desktops and mobile clients present applications that are hosted in a cloud residing on a virtual machine isolated from another corporation’s applications. Economics, technology and business imperatives are driving this future into reality. In fact, IT organizations are increasingly losing control of their application portfolios as a new generation of IT savvy workers develop and/or find applications that help them get work done without the blessing or assistance of corporate IT. As cloud computing promises to radically expand access to applications and low cost application development, IT leaders fear that the portion of applications they control will increasingly shrink if they don’t get ahead of this curve. As such, IT organizations are focusing like a laser on cloud security, application control, portability and the critical potential role of the network.

Proprietary Clouds

The current state of cloud computing is that cloud providers are building proprietary cloud services and APIs that do not interact with other clouds. In addition business and IT leaders of Global 2000 firms are exploring building their own private clouds in an effort to hold on to application control, ensure security and offer elastic IT services. Without an open or standard way of connecting clouds or allowing clouds to interact, isolated clouds will be the norm and a huge lost opportunity will result. Just like islands of isolated networks were the norm during the ’80s and early ’90s, open networking via TCP/IP and the Internet unleashed global productivity and innovation that parallels the industrial revolution. The IT industry is presented with an opportunity to again increase global productivity and innovation by interconnecting clouds or building more advanced network infrastructure capable of handling these new demands.

The Need For Infrastructure 2.0

While the cloud adoption cycle is young and the need for interaction/communication between public and private clouds may not be front and center, it’s only a matter of time before it is. In addition to employees deploying their own applications on corporate networks, IT departments are struggling to meet business expectations. For example, IDC recently predicted that between 2008 and 2012 IT staff will grow at about 1.1 times the rate of business growth while servers will grow at 1.9 times, mobile internet users will grow at 3 times, non-traditional user devices will grow at 3.6 times, information will grow at 4.5 times and interactions per day will grow at 8.4 times. With IT services demand skyrocketing and IT staff budgets being held tight, a gap between business expectation and IT delivery is growing. To close the gap today’s static, manually managed networks need a new economic and delivery model and for most cloud computing is the answer. Today IT growth is accommodated, if funded, within the confines or perimeter of an enterprise network and IT structure while being governed by business and IT leaders, but what if it moved to the cloud?

Consider if IT engages cloud services to meet these business expectations and demands. There would be increasing economic and performance pressures on networks to address the emergence of increasing system/cloud connectivity. Clearly IT organizations are not going to rely upon a single cloud provider to meet their business needs, but a group of providers that may specialize in SaaS/PaaS/IaaS, etc. Herein lies the rub: clouds are going to have to interact and communicate securely among each other if they are going to reach their potential and live up to their hype. If the above IT growth and associated dynamics shift to public clouds over the next 5 to 20 years, which is probable, then there are fundamentally new networking services that need to be designed and incorporated into internet architecture to support Infrastructure 2.0 cooperation.

A New IT Industry Phase

Before I highlight some of the Infrastructure 2.0 services that may be needed, I offer an industry perspective of similar industry transitions and their associated scale. Consider 1984 and the break-up of the Bell systems. Prior to ’84 most large corporations used public voice services to meet their needs. But when bulk T1 transmission services were offered at attractive tariffs most large corporations starting building private voice networks that ushered in companies such as Network Equipment Technologies, Timeplex, Newbridge Networks, et al. To counter this exodus of revenue, service providers offered with success voice private networks (VPNs) to woo enterprises back. In the early ’90s enterprise data networks were proprietary and private, running over leased lines; then, as these proprietary networking protocols such as DECnet, SNA, etc., gave way to open IP networking a huge shift occurred in telecommunications away from traditional PTT/Telecom service provider offerings toward open internet services, changing a $600B worldwide telecommunication market. During this phase Cisco Systems, Wellfleet Communications, Proteon and many others grew at unprecedented rates as LANs, WANs, ethernet switching and routing became essential business tools. Computing clearly benefited from the transition from proprietary-to-open/industry standards as UNIX and WinTel systems linked over open networks drove record corporate productivity and created a new IT industry structure.

Infrastructure 2.0 computing could represent the next phase of IT and be as disruptive as the shift from mainframe computing with SNA to mini-computers with peer-to-peer networking, to personal computing with client-server, to the Internet. With each IT industry phase or transition multi-billion dollar markets were created and the world economy grew thanks to productivity improvements. This shift promises a vast reduction in IT operational expenses along with productivity gains enabled by waves of innovation delivered via increasingly automated networks.

The Open Cloud of Clouds

During each of the above transitions IT leaders questioned vendor lock-in and the security of utilizing public IT services. More often than not, IT leaders were concerned if not frightened about the co-mingling of their bits/applications/compute/storage, etc., with other corporations. Each time the industry responded with a suite of security technology to comfort IT leaders and standards that offered a path away from vendor lock-in. Cloud computing and in particular a suite of open internet services for Infrastructure 2.0 communications could usher in the largest and most sweeping IT transformation yet. Dare I say it; open internet services for the “cloud of clouds” could have a larger economic and societal contribution than even the Internet. The Internet provides connectivity with a few popular applications, i.e., the web, email and increasingly voice and video. Infrastructure 2.0 has the potential to hollow out or outsource IT applications from corporations.

With an open approach to interconnecting clouds, cloud-computing services could scale, applications/objects/data, etc., could be portable, IT organizations could be offered control through visibility and a suite of management/troubleshooting tools could be exposed so that IT infrastructure placement is location/hosting independent. Think of it this way. If an open approach to Infrastructure 2.0 connections were available then a market place made up of cloud computing providers selling a wide range of services to IT departments and consumers could be created. IT leaders could shop the cloud providers, picking and choosing which ones to provide or host applications without being locked in and fearing loss of control plus security vulnerabilities. For example, change management could be policy-based or on demand, versus today’s model of manual configurations, spreadsheets and layers of processes created to compensate for human error potentials.
So what is needed for clouds to communicate with each other so that IT leaders can exploit the cloud for a larger and increasing set of their IT application portfolio while offering a competitive cloud environment for consumers? What follows are a few ideas that could move Infrastructure 2.0 forward.

Open Clouds: Cloud deployments are proprietary and thus there is little to no interoperability among them. To create a pro-competitive environment, clouds and the interconnection between them should be open and interoperable.

Greater Trust: Business and IT leaders plus consumers need to be able to trust cloud services. Ronald Reagan used to say, “trust but verify”; however, clouds need to be verified then trusted. To meet that end, embedded security services versus security appliances could be most effective at creating trust within and between clouds.

Cloud Common Registration System: To find objects, i.e., a VMs, applications, storage, data, etc., within and between clouds a registration system would be very helpful.

Portability: In order to move applications, VMs, data, objects, etc., between clouds portability and persistence is highly desirable.

Management and Visibility: To ensure troubleshooting, diagnostics, repair, modification and control of applications a set of Infrastructure 2.0 management and visibility services could ease the concerns of IT leaders and application developers over their loss of control.

A Means for Billing: A mechanism for providers to bill each other for services may be needed for connections between clouds.

The above is simply a sample of desirable Infrastructure 2.0 attributes. Much work needs to be done to develop an open framework, blue print or architecture. It’s not clear how long it will take to develop an evolved industry network architecture, but as more and more cloud services are consumed market demand will heat up for open approaches to portability, control, scale, security and the ability to mix and match cloud services to construct application portfolios. Infrastructure 2.0 is a game changer that promises to cast a new IT competitive landscape by offering a new standards-based approach to IT delivery.