Today’s technology, particularly big data and analytics, is aiding the advance of AI in the enterprise. Big data helps to drive AI because it provides the information for computers to learn from.
There are several variants of AI, or the capability of a machine to imitate intelligent human behavior and some misunderstanding of what the various terms mean. Machine learning is a form of AI that enables computers to learn from data without explicit programming. Deep learning is a form of machine learning inspired by the structure of the human brain. It uses artificial neural networks to process data similar to the way neurons do in our brains. And, analytics is the scientific process of discovering and communicating the meaningful patterns which can be found in data.
All of these technologies are designed to enable enterprises to make informed business decisions, instead of simply relying on gut-feelings or error-prone human interpretations.
Examples of AI-enabled technologies that can benefit organizations include robots, autonomic software, self-managing systems, chatbots for service, and natural language processing systems (such as Siri and Alexa).
Super-intelligent machines are still many years in the future, but organizations need to understand the benefits that AI can deliver to their IT infrastructure today.
Database Management Systems
Today, database management systems are still at the heart of many modern applications, and relational systems remain the reliable, stalwart core of most. Their sound theoretical foundation helps protect and ensure ongoing access to data for multiple types of applications. When fast transaction processing is required, it is hard to beat a relational DBMS for most use cases.
A 2017 Unisphere Research report on trends in database administration, nearly 60% of the respondents have more than 100TB of structured data under management. To manage this data, companies are using a variety database systems.
Increasingly, many enterprises utilize more than one DBMS within their IT infrastructure (polyglot persistence) and there is an increasing awareness that for some modern, web-based applications, a relational database may not be the best choice. Social media data, streaming audio and video, and information from the Internet of Things (IoT) comprise different data content that requires more flexibility than a traditional RDBMS may be able to efficiently support.
As a result, NoSQL and NewSQL database technologies are being deployed for use cases that relational does not handle efficiently.
NoSQL describes a broad category of database systems that, in some cases, may have dramatically different capabilities and use cases. There are four types of NoSQL offerings: key/value (well-suited for quick lookups by key), document (ideal for storing JSON or XML documents), wide-column (for large records needing a flexible schema), and graph (for managing relationships).
According to the Unisphere study, while the data growth rate and number of databases instances has not changed dramatically in the last 3 years, the database infrastructure has gotten more complex. This complexity has been driven by the emergence of the cloud as a significant platform for database management and the rise of non-relational databases, including NoSQL databases.
Furthermore, the NewSQL category of relational DBMS—typified by a distributed, fault-tolerant architecture and in-memory storage and processing capabilities—has also gained attention. NewSQL systems usually do not have all of the bells and whistles of an RDBMS (such as Oracle or Db2), but neither do they have the overhead.
Virtualization in Multiple Forms
The use of virtualization has been around for decades and refers to the creation of a software-based, or virtual, representation of an application, server, storage device or network. With virtualization, the application, operating system and storage can be abstracted from the actual underlying hardware and software.