After some years working as a database administrator, architect and consultant, I decided to make a change on my career. I decided to make a step forward and ask, how can I help business with data? How can I help business with this technology, and why this one and not the other? So, I started looking at the available openings around data world, and I arrived to the “Data Architect or Big Data Architect”(DA) as a Senior Technical role with strong communication skills specially with stakeholders and IT. After that, I began to read some papers and documents meanwhile I was looking for a role in any industry. After the perfect role arrived, I was ready but a very good question to begin was: What exactly a Data Architect does on his/her daily basis? This question has really lots of different answers, so I will summarize here what I have read on all sites and books around:
- DA need a data strategy first.
- DA must work together and collaborate with business areas and IT. Its like a middleware there, to capture all strategic business needs and translate this into actual IT landscape. Investigating from both sides and aligning data environments to organizational strategies.
- It’s very important to identify vuable data from business stakeholders or key users so to be sure of having impact on these areas.
- DA should elaborate a master plan or roadmap to follow in terms of data needs to achieve the desired organizational goals. This plan must attend not only new requirements but also should fit on existing systems, integrating with existing DBMS.
- Data Governance is a priority for DA in order to assure data quality and relevance. Assigning responsibilities over data.
- Build data systems and architectures to change, in order to be able to adopt new technologies for instance. Flexible for evolution.
- Data systems should be dynamic, integrating on premise with cloud adoption in a hybrid model and ingesting diverse data.
- The architecture should be able to process and ingest real time data but also query to historical.
- Security foundations must be from beginning a concern in this architecture. Access should be mapped to roles.
- MDM Strategy. MDM Repo to have a single point of access to master data and a single point of truth.
- Bring data access as a service for users and applications in order to integrate easily with new requirements or tools.
- Self Service access to data, bring models access so to be ready for users to connect them preferred BI tools.
- Always keep aware of technology trends in the industry. Keep an eye on storage saving technologies.
- Evaluate current data architecture state, beginning from scratch it’s not always possible.
- Plan end to end model, from data repositories, data flows, integrations and access.
- Preserve a repository for all data architecture related artifacts so to be easily access by people inside organization to be query.
- Data modelling is a very important skill. Each phase of data modelling should be achieved, from business model to technology or physical model.
- Achieve a holistic view of data (central repository like a data lake or hub).
- Continue education and act as data evangelist inside company.
It is true that most of the companies nowadays want to be or generate data driven businesses. But only a few have success doing it, this role scope is extense one and every days seems a challenge so be prepared to success with data.
There is an organization called DAMA (Data Management), that release a very complete guide/book called DMBOK. Please take a look, its really recommended for fresh starters. DAMA also offers a certification for Data professionals called CDMP (Certified Data Management Professional) with several levels that depends on your experience. I will start writing on each Data Management topic a review.
Hope you Enjoy it.
Data journey continues.