Dataphin Empowers Enterprise Data Governance

Over the past decade, China's big data industry has experienced a revolutionary transformation, growing at an astonishing paceA mere ten years ago, industries like finance and telecommunications relied almost entirely on international vendors for building data warehouses, leading to project costs that could easily run into the tens of millions, if not hundreds of millions of yuan.

This reliance fostered a perception that data construction and governance were both prohibitively expensive and complex, effectively stalling many businesses from pursuing necessary data initiativesDespite these financial constraints, the demand for robust data infrastructure persisted.

As technology rapidly evolves, new challenges have emerged regarding the scalability of enterprise data architecturesAccording to Gartner, by 2028, half of the data analytics and AI platforms established in China prior to 2023 may become obsolete due to disconnection from evolving ecosystems.

The domain of data construction and governance is in urgent need of reform.

Recently, during the Data × AI forum at the Yunqi Conference, Peng Xinyu, the CEO of Lingyang and Vice President of Alibaba Group, emphasized that companies must embrace the AI era by reconstructing their business and data scenarios

Advertisements

This indicates that a wave of reengineering is beginning within the realm of data governance.

The launch of Dataphin, Lingyang's intelligent data construction and governance product, directly addresses the critical challenges enterprises face during digital transformation: dealing with disorganized and poor-quality dataBy standardizing data processing workflows and unifying data standards, Dataphin offers comprehensive data asset management solutions to help businesses establish reliable and consistent data foundations, paving the way for successful digital transformations.

Originating from Alibaba’s extensive operational experiences as a "superfactory," Dataphin has evolved into a platform capable of managing data governance for companies across various sectors and scales.

To tackle issues such as high costs and limited scalability in data governance, Dataphin has undergone significant upgrades

Advertisements

The newly enhanced flexible architecture leaves room for the evolution of enterprise data governance, introducing an agile version for smaller businesses with moderate data needs, as well as a DataAgent powered by large models, which facilitates better utilization of data assets.

The timing for effective data governance is crucial, particularly as affordable solutions open up new options for businesses.

Wang Sai, Vice President of Lingyang, observed a strong demand for data governance and asset development among small and medium-sized enterprises (SMEs). Even though their data volumes may not rival larger corporations, these SMEs face unique complexities and require light governance solutions for their diverse data sets.

However, the path to effective data governance for SMEs is fraught with challenges

Advertisements

There is often a lack of talent specializing in big data, limited budgets for investment, and inadequate understanding of data asset management and governance.

Addressing these pain points, Lingyang has taken its internal data governance expertise from Alibaba and refined it for external clients, resulting in the agile version of Dataphin.

The agile version of Dataphin features a lighter product architecture, allowing enterprises to start their data governance projects with just three hardware devices and an investment of around 200,000 to 300,000 yuanThis significant reduction in cost compared to previous comprehensive versions of Dataphin provides SMEs with a valuable new option.

As companies embark on their data governance initiatives, they often grapple with a pressing question: as business evolves and data volumes swell, can their existing systems accommodate future needs? Will they need to reassess their data architecture?

For instance, a data manager at a retail company noted the “growing pains” of a widespread operational footprint

• Challenges of Global Economic Recovery
• US CPI Data is Here!
• Gold Prices Continue to Fluctuate
• Synology Technology Empowers Data Value Realization
• A New Contender in the AI Cloud Computing Arena

As their company expanded, the complexity of data demands and processing escalatedThey realized their approach to data governance had to shift: “Previously, we focused on what data was being generated, which business processes could be digitized, and compliance issues.” Looking ahead five years, they recognized that traditional data warehousing strategies would no longer meet the requirements for storing, managing, and utilizing data.

Many enterprises share this concern regarding data governance—how to design a data architecture that not only meets current requirements but also anticipates future complexityResponding to this industry-wide concern, Lingyang has innovatively upgraded Dataphin’s product architecture to feature high scalability and sustainable evolution.

Utilizing the agile version during the initial phase allows companies to smoothly upgrade as they grow, expanding the computational engine beneath without disruption, ultimately transforming into a comprehensive platform for data construction, governance, and operation

Moreover, for firms facing stringent compliance requirements that restrict cloud functionalities, Dataphin supports multi-cloud and multi-engine deployments, accommodating over 50 data sources for versatile configurations that cater to various complex scenarios.

“Whether small, medium, or large, we operate under a single deployment framework, enabling seamless upgrades for enterprises,” Wang Sai notedGiven the long-term nature of data governance, enterprises are free to select products based on their data sizes and governance needsDataphin creates an environment that avoids the pressures to achieve immediate results while ensuring continuous, stepwise advancement.

Additionally, enterprises often face the dilemma of balancing customization with cost efficiencyLarge organizations typically seek private deployment solutions to meet specific operational needs, which inherently incurs higher costs; conversely, standardized cloud products are more affordable but lose the benefit of tailored configurations.

To address this conundrum, Dataphin offers flexible deployment options: fully managed, semi-managed, and independently deployed solutions to cater to diverse needs.

Organizations can subscribe to a public cloud multi-tenant SaaS version much like renting an apartment, paying as needed; they can opt for a semi-managed approach, quickly accessing an independent cloud setup; or they can fully manage their own infrastructure, akin to constructing their own building.

Among these options, the cloud-based semi-managed model is the current favorite among mainstream clients

This setup provides controlled environments while reaping the benefits of public cloud flexibility—essentially allowing users to rent a "villa" on the cloud, available instantly, with services akin to an independent setup.

For example, large corporate groups may have varying data processing requirements across different business modules or subsidiariesFinancial and membership data may need local computations, while less sensitive data can be processed in the cloud, interlinking with cloud-based operationsSuch needs align well with the semi-managed model, which caters to both bespoke customization and cost considerations.

In summary, Dataphin, built on Alibaba Group's years of systematic data governance experience, offers scalable and upgradeable solutions across diverse environments for businesses of all sizes.

The data governance sector is entering a new phase of on-demand procurement and seamless upgrades.

In the AI era, how can enterprises capitalize swiftly on their data assets?

Wang Sai observed a recurring pattern among data governance projects: if a team lacks a data asset management perspective, particularly when composed solely of IT professionals, the chances of project success diminish significantly.

When data remains confined to databases, it becomes merely a cost and a burden

As Wang remarked, “Our fundamental belief is that once data is constructed, it must be actively utilized.” Thus, Lingyang places great emphasis on both governing and operationalizing data assets; aggregating data alone is insufficient—you must also enable effective use of that data.

However, a significant barrier exists between enterprises and their ability to leverage data effectivelyBusiness personnel often find understanding business concepts distinct from grasping data, resulting in a lack of "data literacy." More often than not, they rely on data experts to retrieve data, leading to excessive communication and time costsMoreover, the data teams face their own pressures as they handle numerous inquiries regarding data locations, meanings, and applications, facing the challenge of efficiently pinpointing the necessary data among vast datasets.

This illustrates a greater challenge: companies' demands for data go beyond simply retrieving results from a chatbot; it necessitates an understanding of how to locate and utilize internal data effectively based on business requirements.

For instance, Wang frequently encounters enterprise users asking how to improve conversion rates for business opportunities—requesting data but missing the underlying business context

The solution, he insists, involves a deep understanding of the operational processes—including who is involved and which processes matter, thereby allowing for valuable insights rather than just data-driven results.

In response to these needs, Lingyang has introduced the industry’s first data asset intelligence agent—Dataphin·DataAgentLeveraging advanced models, enterprises can customize their unique agents, allowing business personnel to easily navigate from problems to ideas, data, and applications through self-service operation.

“In the future, everyone will have their own dedicated data assistant,” Wang stated, emphasizing that this digital tool transcends ordinary data retrieval, bridging business contexts to deliver response-based data insights.

Consider the scenario of a leading dairy brand that has developed an extensive data platform over the years, encompassing over 20 data domains, thousands of metrics, and hundreds of user tags

Despite this, some business needs still require data engineering support, leaving business personnel frustrated with slow responses and engineers overwhelmed by demandsWith Dataphin's intelligent DataAgent, they can swiftly locate necessary assets for rapid development, improving efficiency exponentially.

During this process, the three core capabilities of Dataphin·DataAgent come to lightFirstly, it can quickly facilitate the construction of a vector database for a knowledge repository based on comprehensive data assetsSecondly, it offers numerous auxiliary tools and interactive elements to enhance the design and orchestration of the intelligent agentLastly, the data agent can be released with one click, making data consumption simpler for enterprises.

From lowering the entry barriers for SMEs in initiating data governance, to facilitating upgrades in enterprise data system architecture, to balancing customization with cost-effectiveness in deployment, and reducing the complexities of data consumption through the DataAgent, Lingyang's Dataphin is restructuring the paradigms of corporate data governance and redefining the value of enterprise data services.