By Charles H. Silver
Sept. 15, 2011
With more than 37 percent of U.S. companies running a significant portion of business applications in the Cloud (Advanced Micro Devices survey, May 2011), clearly the adoption of Cloud computing will soon become ubiquitous. This marked increase in the use of the Internet for accessing computing resources will necessitate an evolution in the Cloud computing network.
For one thing, Cloud computing will move from operating on the relatively simple networks of today, to vastly more-complex networks needed to integrate applications and data distributed across the Internet.
Beyond this, there will be a need to analyze vast amounts of data - enterprise data, market data and social media data - most of which is corporate information that is not contained in a pre-defined database, i.e., is “unstructured.” Integrating all this data will necessitate replacing archaic, 40-year-old data-management systems with more advanced programs capable of managing data of all kinds - structured or unstructured - distributed anywhere across heterogeneous global networks.
Indeed, in a recent survey, 62 percent of respondents said it is inevitable that unstructured information will exceed the volume of traditional relational data within the next decade. (Unisphere Research, a division of Information Today Inc.) Further, 35 percent say unstructured information has already surpassed or will surpass traditional relational data as soon as over the next 36 months.
Spreading across an increasingly diverse array of data centers, computing will inevitably become much more distributed than it currently is, bringing with it new data-management, architectural and performance challenges.
Significant changes will mean enterprises need to prepare to meet a number of issues. These include mounting costs, increasing government regulation, stringent e-discovery requirements and the increasing complexity of their IT infrastructures.-¨
A key concern will become whether Cloud databases can overcome scaling and performance issues that have plagued conventional databases for years. To gain access to data in the Cloud as things stand now, data-management technologies require that all the data be stored in a centralized database in a single location. Beyond that, there is a severe limitation in the ability for conventional data-management technologies to manage unstructured data.
A database management system that can query data across multiple distributed databases located in multiple geographically dispersed data centers, including Cloud data centers, is what we define as a Cloud database management system.
Conventional systems could not fulfill the role of a Cloud database management system. This includes, in particular, relational database management systems as they are currently engineered. They have centralized architectures, most of which were designed decades ago, that prevent them from being effectively distributed across data centers. To satisfy the most important characteristics of a Cloud database management system, a distributed peer-to-peer architecture is needed. This will be part of the evolution.
Businesses need data-management technology that can effectively access data in any format, distributed anywhere across global computing networks. Eliminating the need to upload or download large volumes of data across the Internet will be a requirement for the Cloud computing networks of the future.
Charles Silver is CEO of Algebraix Data Corp., which provides patented, mathematically based data-management technology across the entire spectrum of computer data-management applications. He has more than 25 years of experience as a successful entrepreneur.