Top 10 technological trends in data processing and analytics in 2019 according to Gartner
Good evening. The translation of the following article was prepared specifically for students of the BI Analytic course. Enjoy your reading.
The focus on the Gartner Data & Analytics Summit on February 18-19 in Sydney was advanced analytics (Augmented Analytics) and artificial intelligence.
')
Expanded analytics , continuous intelligence, and explainable artificial intelligence are among the loudest trends in data processing and analytics technologies that will have a destructive potential in the next 3-5 years, according to Gartner, Inc.
Speaking at the Gartner Data & Analytics Summit in Sydney, Rita Sallam , vice president of research for Gartner, noted that data leaders and analysts should examine the potential impact of these trends on business and adjust business models and operations accordingly, otherwise In the event of a risk of losing their competitive advantage over those who paid enough attention to this.
“The history of analytics data processing continues to evolve, ranging from supporting internal decision making to continuous intelligence, information products, and hiring data specialists ,” said Rita Sallam. “It is very important to get a deeper understanding of the technological trends underlying the creation and development of this story, as well as to set certain priorities in relation to them, depending on the value for a particular business.”
According to Donald Feinberg , Vice President and Outstanding Analyst of Gartner, the main problem caused by digital failure (too much data) also opened up an unprecedented opportunity. The vast amount of data, coupled with the growing power of processing tools provided by cloud technologies, gives a clear understanding that you can now train and execute the algorithms on a large scale necessary to fully realize the potential of AI.
“The size, complexity, distributed nature of data, speed of operation, and continuous intelligence required for the digital business make it clear that hard and centralized architectures and tools can no longer cope,” says Feinberg. “The continued survival of any business will depend on a flexible data-oriented architecture that meets the ever-increasing rate of change.”
Gartner recommends that leaders in data processing and analytics discuss with the business representatives the main priorities of the company and consider how they can integrate the following trends into their work.
Extended analytics is the next wave of breakthroughs in the data processing and analytics market. It uses machine learning and artificial intelligence technologies to transform the development, consumption, and sharing of analytical content.
By 2020, advanced analytics will become the main engine of new purchases in analytics and BI, as well as Data Science, ML platforms and embedded analytics. Data processing leaders and analysts are required to plan the implementation of advanced analytics as the capabilities of the platform evolve.
The Advanced Data Management technology (Augmented Data Management) uses ML capabilities and AI mechanisms to create company information management categories, including data quality, metadata management, master data management, their integration, as well as self-tuning and self-tuning of database management systems (DBMS) . It automates many tasks and allows less qualified users to use the data themselves. In this way, highly qualified technical specialists can focus on more important tasks.
Advanced data management converts metadata from those used only for auditing, pedigree, and reporting, ultimately delivering them to dynamic systems. Metadata changes from passive to active and becomes the main engine for all AI / ML.
By the end of 2022, the number of tasks performed manually in the field of data management will decrease by 45% due to the introduction of machine learning and automated service level management.
By 2022, more than half of the new large business systems will use continuous intelligence, which in turn uses real-time contextual data to improve solutions.
Continuous intelligence is a design pattern in which real-time analytics is integrated into business operations, processing current and historical data to suggest actions in response to an event. It provides automation or decision support. Continuous intelligence uses several technologies, such as advanced analytics, event flow processing, optimization, business rule management, and machine learning.
“Continuous intelligence is a major innovation in data and analytics teams,” says Sallam. “This is a daunting task and a great opportunity for teams of analysts and BI specialists to help companies make smarter decisions in real time as early as 2019. It can be viewed as the final version of operational BI. ”
AI models are most often used to improve or completely replace a person in decision making. However, in some scenarios, companies must justify how these models arrive at specific solutions. To build user or stakeholder confidence, application architects should make these models more understandable and explainable.
Unfortunately, most advanced AI models are complex black boxes that cannot explain how they derived a specific recommendation or solution. Explainable AI in data science and ML platforms, for example, automatically generates an explanation of models in terms of accuracy, attributes, model statistics, and functions in natural language.
Graph analytics (Graph analytics) is a set of analytical methods that allow you to explore the relationship between objects of interest, such as organizations, people and transactions.
The use of graphic processing and graphic DBMS will increase by 100% every year until 2022, which will speed up the preparation of data and provide more complex and adaptive data science.
Graphical data warehouses can effectively model, explore, and query data with complex relationships between data warehouses, but the need for specialized skills to work with them is their main constraint today.
Over the next few years, graphical analytics will steadily grow, as there is a need to ask complex questions to complex data, which is not always practical or at least feasible at a scale in which SQL queries can be used.
Data fabric provides easy access to and sharing of data in a distributed data environment. It is a unified and consistent framework for data management, which provides unimpeded access to the data and the possibility of its architectural processing in any other repository.
Until 2022, custom data fabric projects will be deployed mainly as a static infrastructure, forcing organizations to invest in a new wave of costs of complete reorganization to provide more dynamic approaches to the data mesh (data mesh).
By 2020, 50 percent of analytical queries will be generated using natural language processing (NLP) or voice processing, or automatically generated. The need to analyze complex data combinations and make analytics available to everyone in an organization will lead to its wider use, which will allow analytics tools to be as easy as a search interface or a conversation with a virtual assistant.
Gartner predicts that by 2022, 75% of new end-user solutions that use AI and ML methods will be built on commercial solutions, rather than on open source platforms.
Commercial vendors embed connectors into an open source ecosystem, thereby providing the enterprise functions necessary to scale and democratize AI and ML, such as project and model management, reuse, transparency, lineage of data, and consistency and integration with other platforms, which is so lacking in open platforms.
The main value of the blockchain and the distributed registry (distributed ledger technologies) is to provide decentralized trust in the network of untrusted participants. There is a significant potential for analytics use options, especially those involving participants' relationships and interactions.
However, it will be several years before four or five major blockchain technologies begin to dominate. Until this time comes, end-users of technologies will have to adjust to the technologies and standards of the blockchain, which are dictated by the prevailing customers or networks. This includes integration with existing data and analytics infrastructure. Integration costs can exceed any potential benefit. Blockchain is a data source, not a database, and does not replace existing data management technologies.
New technologies with the use of persistent memory (persistent-memory technologies) will help reduce the cost and complexity of implementing architectures with support for in-memory calculations (IMC). Permanent memory is a new level of memory between DRAM and NAND flash memory, which can serve as an economical storage device for high-performance loads. It has some potential that can be used to improve application performance, availability, load time, clustering methods, and security methods, while keeping costs under control. It will also help organizations reduce the complexity of their application programs and data architectures by reducing the need for data duplication.
“The volume of data is growing rapidly, and the relevance of transforming ordinary data into valuable data in real time is growing with it,” said Feinberg. “New server workloads require not just faster processor performance, but also more memory and faster data storage.”
More information about using data and analytics to gain competitive advantage can be found in the Gartner Data & Analytics Insight Hub .
Gartner Data & Analytics Summits in 2019 will be held March 4-6 in London , March 18-21 in Orlando , May 29-30 in Sao Paulo , June 10-11 in Dubai , September 11-12 in Mexico City , October 19-20 Frankfurt . Follow the news and updates on Twitter via the #GartnerDA hashtag .
Gartner, Inc. is a leading global scientific consulting company and member of the S & P 500. We supply business leaders with the necessary data, tips and tools to achieve their goals today and create successful organizations tomorrow.
Our unrivaled combination of expert, practical data research helps clients make the right decisions on the most important issues. We are a reliable consultant and objective resource for more than 15,000 organizations in more than 100 countries - for all core functions, in any industry and for companies of any size.
To learn more about how we help decision makers build the future of their business, visit gartner.com .
That's all. Write comments and see you on the course!
The focus on the Gartner Data & Analytics Summit on February 18-19 in Sydney was advanced analytics (Augmented Analytics) and artificial intelligence.
')
Expanded analytics , continuous intelligence, and explainable artificial intelligence are among the loudest trends in data processing and analytics technologies that will have a destructive potential in the next 3-5 years, according to Gartner, Inc.
Speaking at the Gartner Data & Analytics Summit in Sydney, Rita Sallam , vice president of research for Gartner, noted that data leaders and analysts should examine the potential impact of these trends on business and adjust business models and operations accordingly, otherwise In the event of a risk of losing their competitive advantage over those who paid enough attention to this.
“The history of analytics data processing continues to evolve, ranging from supporting internal decision making to continuous intelligence, information products, and hiring data specialists ,” said Rita Sallam. “It is very important to get a deeper understanding of the technological trends underlying the creation and development of this story, as well as to set certain priorities in relation to them, depending on the value for a particular business.”
According to Donald Feinberg , Vice President and Outstanding Analyst of Gartner, the main problem caused by digital failure (too much data) also opened up an unprecedented opportunity. The vast amount of data, coupled with the growing power of processing tools provided by cloud technologies, gives a clear understanding that you can now train and execute the algorithms on a large scale necessary to fully realize the potential of AI.
“The size, complexity, distributed nature of data, speed of operation, and continuous intelligence required for the digital business make it clear that hard and centralized architectures and tools can no longer cope,” says Feinberg. “The continued survival of any business will depend on a flexible data-oriented architecture that meets the ever-increasing rate of change.”
Gartner recommends that leaders in data processing and analytics discuss with the business representatives the main priorities of the company and consider how they can integrate the following trends into their work.
Trend number 1. Advanced Analytics
Extended analytics is the next wave of breakthroughs in the data processing and analytics market. It uses machine learning and artificial intelligence technologies to transform the development, consumption, and sharing of analytical content.
By 2020, advanced analytics will become the main engine of new purchases in analytics and BI, as well as Data Science, ML platforms and embedded analytics. Data processing leaders and analysts are required to plan the implementation of advanced analytics as the capabilities of the platform evolve.
Trend number 2. Advanced data management
The Advanced Data Management technology (Augmented Data Management) uses ML capabilities and AI mechanisms to create company information management categories, including data quality, metadata management, master data management, their integration, as well as self-tuning and self-tuning of database management systems (DBMS) . It automates many tasks and allows less qualified users to use the data themselves. In this way, highly qualified technical specialists can focus on more important tasks.
Advanced data management converts metadata from those used only for auditing, pedigree, and reporting, ultimately delivering them to dynamic systems. Metadata changes from passive to active and becomes the main engine for all AI / ML.
By the end of 2022, the number of tasks performed manually in the field of data management will decrease by 45% due to the introduction of machine learning and automated service level management.
Trend number 3. Continuous intelligence
By 2022, more than half of the new large business systems will use continuous intelligence, which in turn uses real-time contextual data to improve solutions.
Continuous intelligence is a design pattern in which real-time analytics is integrated into business operations, processing current and historical data to suggest actions in response to an event. It provides automation or decision support. Continuous intelligence uses several technologies, such as advanced analytics, event flow processing, optimization, business rule management, and machine learning.
“Continuous intelligence is a major innovation in data and analytics teams,” says Sallam. “This is a daunting task and a great opportunity for teams of analysts and BI specialists to help companies make smarter decisions in real time as early as 2019. It can be viewed as the final version of operational BI. ”
Trend number 4. Explainable AI
AI models are most often used to improve or completely replace a person in decision making. However, in some scenarios, companies must justify how these models arrive at specific solutions. To build user or stakeholder confidence, application architects should make these models more understandable and explainable.
Unfortunately, most advanced AI models are complex black boxes that cannot explain how they derived a specific recommendation or solution. Explainable AI in data science and ML platforms, for example, automatically generates an explanation of models in terms of accuracy, attributes, model statistics, and functions in natural language.
Trend number 5. Graphics
Graph analytics (Graph analytics) is a set of analytical methods that allow you to explore the relationship between objects of interest, such as organizations, people and transactions.
The use of graphic processing and graphic DBMS will increase by 100% every year until 2022, which will speed up the preparation of data and provide more complex and adaptive data science.
Graphical data warehouses can effectively model, explore, and query data with complex relationships between data warehouses, but the need for specialized skills to work with them is their main constraint today.
Over the next few years, graphical analytics will steadily grow, as there is a need to ask complex questions to complex data, which is not always practical or at least feasible at a scale in which SQL queries can be used.
Trend number 6. Data fabric
Data fabric provides easy access to and sharing of data in a distributed data environment. It is a unified and consistent framework for data management, which provides unimpeded access to the data and the possibility of its architectural processing in any other repository.
Until 2022, custom data fabric projects will be deployed mainly as a static infrastructure, forcing organizations to invest in a new wave of costs of complete reorganization to provide more dynamic approaches to the data mesh (data mesh).
Trend number 7. NLP / Talking Analytics
By 2020, 50 percent of analytical queries will be generated using natural language processing (NLP) or voice processing, or automatically generated. The need to analyze complex data combinations and make analytics available to everyone in an organization will lead to its wider use, which will allow analytics tools to be as easy as a search interface or a conversation with a virtual assistant.
Trend # 8 Commercial AI and ML
Gartner predicts that by 2022, 75% of new end-user solutions that use AI and ML methods will be built on commercial solutions, rather than on open source platforms.
Commercial vendors embed connectors into an open source ecosystem, thereby providing the enterprise functions necessary to scale and democratize AI and ML, such as project and model management, reuse, transparency, lineage of data, and consistency and integration with other platforms, which is so lacking in open platforms.
Trend # 9: Blockchain
The main value of the blockchain and the distributed registry (distributed ledger technologies) is to provide decentralized trust in the network of untrusted participants. There is a significant potential for analytics use options, especially those involving participants' relationships and interactions.
However, it will be several years before four or five major blockchain technologies begin to dominate. Until this time comes, end-users of technologies will have to adjust to the technologies and standards of the blockchain, which are dictated by the prevailing customers or networks. This includes integration with existing data and analytics infrastructure. Integration costs can exceed any potential benefit. Blockchain is a data source, not a database, and does not replace existing data management technologies.
Trend number 10. Persistent storage servers
New technologies with the use of persistent memory (persistent-memory technologies) will help reduce the cost and complexity of implementing architectures with support for in-memory calculations (IMC). Permanent memory is a new level of memory between DRAM and NAND flash memory, which can serve as an economical storage device for high-performance loads. It has some potential that can be used to improve application performance, availability, load time, clustering methods, and security methods, while keeping costs under control. It will also help organizations reduce the complexity of their application programs and data architectures by reducing the need for data duplication.
“The volume of data is growing rapidly, and the relevance of transforming ordinary data into valuable data in real time is growing with it,” said Feinberg. “New server workloads require not just faster processor performance, but also more memory and faster data storage.”
More information about using data and analytics to gain competitive advantage can be found in the Gartner Data & Analytics Insight Hub .
Gartner Data & Analytics Summit
Gartner Data & Analytics Summits in 2019 will be held March 4-6 in London , March 18-21 in Orlando , May 29-30 in Sao Paulo , June 10-11 in Dubai , September 11-12 in Mexico City , October 19-20 Frankfurt . Follow the news and updates on Twitter via the #GartnerDA hashtag .
About Gartner
Gartner, Inc. is a leading global scientific consulting company and member of the S & P 500. We supply business leaders with the necessary data, tips and tools to achieve their goals today and create successful organizations tomorrow.
Our unrivaled combination of expert, practical data research helps clients make the right decisions on the most important issues. We are a reliable consultant and objective resource for more than 15,000 organizations in more than 100 countries - for all core functions, in any industry and for companies of any size.
To learn more about how we help decision makers build the future of their business, visit gartner.com .
That's all. Write comments and see you on the course!
Source: https://habr.com/ru/post/457450/
All Articles