“Using their programmeming skills, a data architect determines how to store and access a company's data. They develop database solutions, install database systems, maintain database performance, make updates, provide support, and conduct training. Other duties include implementing data warehousing, data modelling, and creating metadata registries.” 

"Data warehouse concept characteristics are subject-oriented, for instance, sales, distributions, and marketing. It's made-up of integrated data from several sources such as the mainframe, flat files, and relational databases. Time variance where data has an element of time and once inserted it is static. Finally, data warehouse is non-volatile because it saves old data even new data is input."

“The most widely used database warehouse architecture comprises three tiers: the bottom tier which is a relational database system; the middle tier, which is an abstract view of the database from an OLAP server; and the top tier which is a front-end client layer. I once restructured a database by compiling recurring customer subscribers in the bottom tier, constructing the middle tier using sales data for various reporting periods, with the top tier containing user information operations. This method breaks large amounts of data down into easily accessible files.”

“I am detail-oriented and focused. I have hands-on experience with SAS, SQL, data modelling, the ETL process, Tableau, and developing data strategies. I hold a master's degree in computer science and have ten years of experience in this field.” 

“There are three case manipulation functions. LOWER which takes a string as an argument and returns it in lowercase. UPPER which returns the argument string converted into uppercase. Finally, INITCAP returns the argument string with an uppercase first letter and the remaining letters in lowercase.”

" I first meet with the end-users to develop a requirements document of the functional and business requisites. We define the design requirements, describing how to build the solution. After we put together a project schedule with each phase defined by a deliverable, we write test plans for each phase. The project is complete when the execution of the test plan validates each documented requirement, on budget, and on time."

“In my experience, the main problem is that integration creates data in an incompatible format. I resolved this problem by writing and implementing a script before uploading the external data into the data warehouse tables. I then ran tests verifying that the external data format was a match to our system.”

“In my experience with Hadoop and MySQL, I've never faced major problems. However, I know that open-source software utilities or databases have their disadvantages. There's no customer support provided, and since developers mostly ignore user interface, you may not have the resources needed to get started.”  

“Most of the time, the biggest challenge with data security is coming to a consensus with my team on what poses security risks. One time my team wanted to change the process for uploading data. To help them understand the associated risks, I reviewed what the probable financial loss should their new idea result in a security breach for the company. As a result, they modified their plans and strengthened security measures. To ensure security, I consider the specifics of each architecture design, then I use the specifics of each design to strategize against specific threats and procedures.”

“I like working with MySQL because it's a specialised tool and easy to integrate with various types of businesses.”

"Data modelling has two schemas, the snowflake and the star, and they're what give a logical description to data objects. The star schema is the simplest and has a fact table in the centre surrounded by lookup tables which have a direct reference to the centre table. The snowflake system is a lot more complex. Each first-level lookup table has its own lookup table, and it's dispersed throughout the entire system. The star takes up more disk space than the snowflake. It also has a high redundancy and duplication rate, while the snowflake schema data is normalized and has a low level of data redundancy."

“The virtual data warehouse is where you find a completed view of the company's completed data, has no historical data, is a logical data model, and includes metadata. It's an information system used in analytical decision-making systems, and translates raw data into meaningful data while providing suggestions when used for decision-making.”

“Sometimes I have to reject requests for deliverables because of data architecture limitations. Because they are unfamiliar with how our data systems work, this rejection can cause some frustration. I've learned to resolve the tension with collabouration. We discuss their needs, I learn about their work, and we brainstorm solutions we can test and adjust.”

“First is the conceptual model which shows the basic high-level design of physical data. The second is logical, where entity names and relationships, attributes, foreign and primary keys show up. The third is the physical data model that shows a detailed design with more information and how it's implemented in the database. I have hands-on experience with PowerDesigner and Oracle SQL Developer Data Modeller.” 

“One of the more commons mistakes is building a huge data model, and that will inherently have faults. Another is a lack of purpose or an unclear definition of the software deliverable. People also sometimes denormalize a database which creates redundant data.”