You have come this far in the search for a solution for performing analytics on the mainframe data. You are now ready to consider stopping the ETL process in favor of doing analytics on the mainframe, directly on the operational data. So, it is time to define those elusive goals, objectives and use cases.
We will explore the iBM Open Data Analytics for z/OS product’s proof points for implementing this new product into your enterprise.
Defining the value of the IBM Open Data Analytics for z/OS POT
Goals
Sometime the goals of the POT can elude you, even when there are known objectives. Here are some examples of goals that you could consider.
- The Proof of Technology will determine applicability of IBM Open Data Analytics for z/OS; or, does this product “fit” with the overall enterprise analytics strategy?
- The Proof of Technology will provide insights to the benefit of using IBM Open Data Analytics for z/OS by the data scientists; or, do the data scientists find more value using this tool than another methodology?
- The Proof of Technology will prove the business case justification for IBM Open Data Analytics for z/OS; or, can a quantifiable business be proven, thereby justifying the costs to implementation?
The key is to define the goal that will drive the go/no go decision.
Objectives
The objectives will support meeting the defined goal. If the goal has an affinity to the data scientists’ needs, so should the objectives. When defining the objectives, consider these categories.
- Technology
- Operations
- Usability
- Education
Technology
These objectives focus on proving technical features of the product.
Some might include:
- Prove query processing is offloaded to the zIIP.
- Prove a query can successfully federate to SQL Server data base.
- Prove VSAM and Db2 for z/OS data can be joined.
Operation
Here the infrastructure is evaluated by the enterprise architects and operations teams.
Examples might be:
- Products can be installed and maintained by the current systems team.
- Understand how to handle the change management life cycle on the queries.
- Validate the monitoring of production queries.
Usability
The usability can address use of the tool from any team member, not only the business side.
Examples might include:
- Data Service Studio is an easy tool for the data analysts to create and test new queries.
- Prove that the data scientists can execute a query through the Jupyter notebook without a steep learning curve (assume the data scientist is already familiar with Jupyter notebooks).
- Show that the xyz query, that accesses the mainframe operational data for sold products, presents resulting query response more timely and accurate than running that query against the data lake (or data warehouse).
Education
These objectives address the learning curve and roll out plans to the team members.
Use Cases
One use case can successfully address many of the objectives, therefore only a small number of use cases might be needed.
Each use case might present the specifications, such as:
- The query, whether new or replicated from current environment
- Tables, files and location of these tables and files (not all data will reside on the mainframe.)
- Amount of data from the tables and files (it’s best to use a test data with a few number of rows.)
- The tools used, such as Data Service Studio, to create the replicated query and Jupyter notebook to test the query
- The resultant data expected, i.e. the number of rows
- Team members’ roles or assignments for the use case’s tasks
- Required authorizations (system and team members)
Ultimately the use cases should address all objectives in all categories.
Next Steps
Through these four BLOGS, you now know the Proof of Technology project plan; you understand the resource team to be included; and now, you are ready to get started with developing the business case, goals, objectives and use cases. Let’s begin the IBM Open Data Analytics for z/OS Proof of Technology.
Reach out to Marianne Eggett at marianne.eggett@mainline.com or your local Mainline Account Executive to learn how IBM Open Data Analytics for z/OS can help your IT infrastructure.
Please contact your Mainline Account Executive directly, or click here to contact us with any questions.
Additional BLOGs:
1) IBM Open Data Analytics for z/OS for Mainframe Data Access – An Evolution of Mainframe Apache Spark
2) The Financial Business Case for IBM Open Data Analytic for z/OS
3) IBM Open Data Analytics for z/OS Proof of Technology – What to know to begin your project
4) IBM Open Data Analytics for z/OS Goals, Objectives and Use Case
5) The Evolving Mainframe – Approaching the Analytics Community