摘要：
Last year we started down the new road of uncovering enterprise analytics with machine learning. This year will see an acceleration. Here are some trends to watch.
The amount of data and analytics that are being made available in enterprises, and the enterprise’s ability to utilize them is growing at unprecedented levels. It’s to the point where the core fiber of the modern enterprise is changing, which will lead to a changing of the guard in those we hold up as exemplary.
Fueling the imperative are several new approaches and platforms. It is indeed an exciting time to be working with enterprise data and analytics. Enterprise data in 2022 is going to be an exciting journey. Here are some trends to keep an eye on as the year unfolds:
Edge Gains
Embedded databases at the edge of the architecture have become a popular use of database technologies. Now that enterprises have become pseudo-software factories churning out applications, building mobile applications, and supporting IoT, enterprises have jumped into embedded databases in a big way.
Enterprises using IoT can use embedded databases at the edge to copy aggregated sensor data to a back-end database when online. This brings the value of data directly to operations. At the same time, data from all the devices is being managed in the back-end database to develop analytics to advance the business.
Artificial intelligence chips are taking center stage in these environments. AI chips refers to a new generation of microprocessors that are specifically designed to process artificial intelligence tasks faster and use less power. They are particularly good at dealing with artificial neural networks and are designed to do the machine learning model training and inference at the edge.
We’ll also see the need for higher performance from edge computing hardware since better sensors and larger AI models now enable a host of new applications. There is a growing requisite to infer more data and then make decisions without sending data to the cloud. Also distributed sites can be linked together with an enterprise computing environment to create a unified computing environment.
Demand for intelligent edge applications is rising rapidly and with widely available development tools and with semiconductor companies launching new machine learning (ML) features, adoption of edge ML applications will become a major trend. Also expect graph databases to emerge on the edge this year.
Wide Adoption of Containerized Environments
Enterprises are definitely interested in containerized environments. The issue that’s been holding back containerization is that stateful applications that need persistent storage have been leaning on legacy infrastructure in production. More Kubernetes-ready distributed-RDBMS platforms have addressed the stateful persistence challenges in their latest releases. This will expand the Kubernetes envelope this year.
The solution simulates a single logical database while guaranteeing transactions and enabling scalable deployment across regions and clusters without federation.
The number of high-quality products for Kubernetes has been growing. Advances in security and orchestration will also boost Kubernetes.
AI, Based on Data, Moves Hard into Design
We’re not all in the NFT art, whiskey, music, or paintings businesses, but we can look to them as examples of what AI-based design is capable of and create bridges into our designs in the enterprise. Keep in mind these nascent works of art are as bad as AI-based design will ever be. It only goes up from here, just like AI-based design in the enterprise.
Ignoring AI in enterprise design or relegating it due to a notion that design is a fully human activity is done at peril this year. This design extends to the technology and software we develop in the enterprise.
Google said that a chip that would take humans months to design can be dreamed up by its new AI in less than six hours. The AI has already been used to develop the latest iteration of Google’s tensor processing unit chips.
AI can explain code in English and suggest improvements. It is starting to write code. For example, The Defense Advanced Project Agency’s Probabilistic Programming for Advancing Machine Learning (PPAML) program is developing new technologies that improve machine learning for questions.
So, will AI be writing most of enterprise code out there? Eventually yes and it all starts this year and it may start in earnest with AutoML, a machine learning model designed to select algorithms and create other machine learning models using reinforcement learning.
Additional enterprise data trends to watch are data observability, data catalogs, AI-based applications, data fabrics, streaming analytics, and synthetic data.
留下你的回應
以訪客張貼回應