z Systems Solutions Consultant
We have all been there at least once, another Proof of Technology where we are fighting for resources, struggling with the learning curve, and being asked by management, “How’s it going?”. We wonder, “Where are the experts that can help us do this smarter and quicker, and bring us a successful POT?”
As a project manager, I influence these POTs with a planning perspective. Yet, no plan works without having the right people to make it work. Let’s explore the plan, and the people involved, in the IBM Open Data Analytics for z/OS Proof of Technology.
The Plan for IBM Open Data Analytics for z/OS POT
There have been many approaches to enabling mainframe data for analytics throughout the years. The IBM Open Data Analytics for z/OS product offers a new approach by bringing Apache Spark analytics capabilities to the mainframe data, thereby using operational data instead of replicated data. In addition, it offers an easy data-mapping tool for data analysts, masking the data access complexities from the users, while running a significant portion of its execution on mainframe specialty engines, called zIIPs. This sounds great, but how does it help the company’s business. To find out, let’s look at executing a Proof of Technology at your company.
Discovery: In this phase, the product is introduced to the company’s IT and Lines of Business teams. The team is made up of z/OS Systems Administrators, Business Analysts, Data Analysts, Data Administrators and Data Scientists. The goal is to learn enough about the product to decide on a Proof of Technology. IBM or Mainline Information Systems can provide a presentation of the product’s capabilities and discuss how other companies have gained benefits from implementing this product. This is followed by a brainstorming session to explore potential use cases for the company.
Next, the company’s team builds a strawman of quantifiable use cases. A Business Case justification may be required. For more detail on building a business case for IBM Open Data Analytics for z/OS, see reference 2 below where this topic was explored.
Solution Design: In this phase, the use cases for Apache Spark analytics are developed. The Proof of Technology’s goals and objectives are documented. The strawman from the discovery phase is shared with the IBM, Rocket Software and Mainline team. IBM and Rocket bring their product expertise to ensure the value of the use cases. This team advises the Company’s IT team on further ideas for the Apache Spark product exploitation. To finalize the use cases, there is a detailed walk through between the customer team, IBM and Rocket experts, and Mainline. Once the use cases are finalized, the source data for the testing is identified, i.e. tables, sequential files, VSAM files, etc.
Concurrently, the company’s mainframe infrastructure readiness is explored. IBM, Rocket and Mainline offer expertise, making sure the mainframe has the capacity required to execute a Proof of Technology. For the POT, IBM offers a 90-day loaner contract for the specialty engines, zIIPs, and memory. The required loaner hardware capacity is determined by many factors, including scope of use cases and current mainframe capacity. Many mainframes have dormant installed capacity that can easily be activated by IBM. A loaner contract is initiated between IBM and the company for the hardware and the software, IBM Open Data Analytics for z/OS product. Mainline facilitates this phase.
Both a project and resource plan are required at this point. The names associated with z/OS systems administrators, business analysts, data analysts, data administrators and data scientists, along with IBM, Rocket and Mainline team members, are identified. Mainline provides the project management for the POT. The customer may include their project manager who coordinates the company’s resources during the testing phase.
The software is delivered by SHOPz, and the installation process begins. IBM offers lab services, at a nominal cost, to support the systems programmers with the installation and customization. The installation of IBM Open Data Analytics for z/OS includes Apache Spark, Python and core Anaconda libraries, the Operational Data Integration Layer, JobServer, the Mainframe Data Service Studio (UI for the Operational Data Layer), as well as Jupyter Notebook capabilities.
In-Depth Workshop: Using the newly-installed system, a one-day in-depth workshop is conducted for the full IT team. This day’s agenda is filled with demonstrations, skills transfer, and starting the set up and execution of the first use case. At the end of this day, the company’s IT team is off and running on creating and executing the use cases.
Testing: Weekly status meetings are recommended. These meetings are used to track testing progress, and as a forum for the product’s Q&A and problem resolution. Depending on how skilled the company’s IT team is in the Apache Spark’s analytics, this will dictate the speed with which they get through the use cases.
POT Results: A final walk through of the testing results with the company’s team, IBM, Rocket Software and Mainline is recommended. The POT’s success depends on validating the business justification and ensuring the testing addressed the POT’s goals and objectives.
In the next BLOG about IBM Open Data Analytics for z/OS, we will explore the development of the POT’s objectives and use cases.
For further information, reach out to Marianne Eggett at firstname.lastname@example.org 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.