The Oracle Exalogic Elastic Cloud is not a new product. At least not in the usual sense of the description. Exalogic is a system engineered (built) from well known best of breed component products, each of which is well established, well known and generally matrue. What is new and unique about Exalogic is the integration of those components - the system as a whole, and the software "glue" that binds the system together and elevates the system beyond merely being the sum of it's parts. Exalogic is, in many respects, the productization of the integration and development work that our customers have been forced to do for years. Building a platform from a stack of components is not, and has never been, nearly as easy, quick or error free as many people believed, and maintaining such systems is a remarkably costly and complex long term commitment. Perhaps the main reason these costs and risks were not more of a focus of our attention before today (or at least didn't seem to be) is simply because we accepted them as a fact of life: this work was, almost by definition, what IT organizations did for a living. Only recently, with the advent of Cloud and a wave of vendor consoilidation, has it become clear that there are now (finally) real choices to be made. For organizations that want to concentrate their resources on their core business, we now have alternatives that allow vendors to shoulder an larger share of the burden of simply 'keeping the lights on'. For Oracle, this change has been one of our ambitions for several years.
Oracle Exalogic is key part of Oracle's long term strategy for datacenter and application platform evolution, which we began back in 2006 with the Oracle Database 10g release (the "g" stands for "Grid"). Oracle's observation, at that time, was that the majority of IT costs, risk and inflexibility are the result of "application silos". Most large organizations approached new application projects, such as deploying an ERP system or a data warehouse, as an engineering project that involved building a complete infrastructure from the ground up. While competence, facilities, vendors and the like would be re-used from project to project, the resulting applications were very deep multi-layered systems that were ultimately very sophisticated, complex, unique and proprietary. The result was that projects were long (and hard to plan), expensive to execute (typically requiring many outside expert consultants) and resulted in systems that were difficult to integrate and costly to own.
Many of Oracle's largest customers were already taking steps to reduce costs by encouraging standardization of components and architectural approaches as a means of improving supply chain efficiency and re-use of components and competence between projects. While this new focus on standardization was clearly the right first step, Oracle realized that even greater benefits could be realized if IT organizations went further and began consolidating applications onto common infrastructure. Not only would customers use the same type of server hardware or database software for multiple applications, they would actually use the very same physical servers and database or application server instances. This consolidation would amplify the benefits of standardization and represent the logical end state for any serious standardization initiative. To make this possible, however, Oracle realized that there were many technical challenges that would need to be overcome, and embraced the notion of "grids" as the model for this type of consolidation.
Many of the key management, scalability and reliability features of the 10g and 11g releases of Oracle's Fusion Middleware and Database products were the result of our work with customers that agreed with Oracle's approach and felt that Grid architecture was indeed the natural next step. Technologies like Real Application Clusters (RAC) and Oracle Coherence were direct results of this new focus on Grid architecture.
Even though Grid architecture represented a huge step forward and resulted in very significant savings and new efficiency, the rate at which IT organizations were expanding their use of technology and growing through acquisition continued to accelerate. In 2008 and 2009 it became obvious to Oracle that the complexities of consolidation could not be adequately addressed by an approach that involved only the management, Application and Database grid software alone. Oracle's customers were increasingly frustrated by the complexity of developing their own platforms, from disk to application, for applications which they did not develop themselves and the implementation of which did not represent any real opportunity for differentiation from competitors. Our customers began to ask why we were not doing more to relieve them of the need to invest in projects that were ultimately not their core business.
Our answer was to partner with HP to introduce the first turn-key engineered system for Data Warehousing, called the Exadata Database Machine. As the technology developed and the level of market acceptance for this approach grew, it became clear to Oracle that our future was in Engineered Systems. Today, we are entering a new era in IT, where enterprises demand the option of deploying applications on external infrastructure (public clouds), purchasing complete ready-to-run engineered systems or baskets of best-of-breed components. Decisions are made based on many factors, such as analysis of initial capital costs, security and business continuity requirements, and whether they feel that a given project represents an opportunity for differentiation and competitive advantage.
Oracle Exalogic is an ideal solution for customers that are seeking to balance all of their concerns and want to combine the flexibility of cloud-like management models with the cost efficiency of application consolidation on a turn-key hardware/software platform. With Oracle Exalogic and Oracle Exadata together, we are confident that there is no better environment for consolidation of Oracle applications, in particular.