2010 ~ 2021 Gartner 数据管理技术成熟度曲线

原创 eygle 2022-01-06

2733

Gartner 数据管理技术成熟度曲线。

2021年度

2020年度

2019年度

2018年度

dbPaaS是任何数据库管理系统(DBMS)或数据存储，设计为可伸缩、弹性、多租户的订阅服务，具有一定的自我管理能力，并由云服务提供商(CSP)或CSP基础设施上的第三方软件供应商销售和支持。

Innovation Triggers in 2018

DataOps is a collaborative data management practice focused on improving the communication, integration and automation of data flows between data managers and consumers across an organization. Much like DevOps, DataOps is not a rigid dogma, but a principles-based practice influencing how data can be provided and updated to meet the need of the organization’s data consumers.

“DataOps is a new practice without any standards or frameworks,” said Nick Heudecker, research vice president at Gartner. “Currently, a growing number of technology providers have started using the term when talking about their offerings and we are also seeing data and analytics teams asking about the concept. The hype is present and DataOps will quickly move up on the Hype Cycle.”

Private cloud dbPaaS offerings merge the isolation of private cloud database platforms with the self-service and scalability of the public cloud. They recently started to appear in vendors’ portfolios and provide a cloud experience in an on-premises data center. Gartner analysts said private cloud dbPaaS can play the role of a transition technology as organizations develop their long-term cloud strategy.

“Private cloud dbPaaS is an option for organizations that are unable or not ready to move to public cloud offerings, due to security, regulatory or other concerns,” said Adam Ronthal, research director at Gartner. “Often, these organizations use existing on-premises infrastructure for dbPaaS, which as a result will shorten mainstream business adoption.”

Rudimentary machine learning ML has been used in data management products since the 1970s. Today, with the increased availability of ML and artificial intelligence (AI) libraries, vendors use modern varieties of ML for many self-management operations within data management software. These solutions not only tune and optimize the use of the products themselves, but suggest new designs, schemes and queries.

“We placed ML-enabled data management in the pre-peak section of the Hype Cycle because many of the modern use cases are in infancy,” Feinberg said. “However, the technology will move fast. Many of the products using ML in data management are today only available on cloud platforms and likely to be trained with massive amounts of data. The improvements resulting from these training sessions will spread to on-premises software and there will be a surge in ML-enabled data management adoption during the next few years.”