No matter whether you think that mainframes are legacy technology or not, the fact remains that mainframes are not going away any time soon. Today more than 70 percent of enterprise data resides on a mainframe and 71 percent of all Fortune 500 companies have their core businesses located on a mainframe. Currently, 10,000 mainframes are being actively used.

Mainframes are still popular as business computing systems for obvious reasons: They are efficient in handling large numbers of processes and computing vast amounts of data. If your businesses require scaling up (processing large quantities of credit or debit card payments, for example), mainframes are still a preferred option.

While mainframes offer unique capabilities that meet some enterprises’ critical needs, they also present unique challenges in terms of data governance and management.

One of these challenges is the lack of traceability and understanding of the mainframe environment. This is because, as systems of records, the mainframe stores tremendous amounts of data. Furthermore, many of the mainframe applications are written in 60+ year-old programming languages such as COBOL and PL1. It is impossible to have visibility into these complex data assets and codes with manual efforts and unscalable tools.

Another challenge is the lack of integration of the mainframe environment with the rest of the IT systems. Whether it is due to their complexity or size, the mainframe environment tends to be an island of its own. Enterprises need to change this status quo in order to establish end-to-end transparency of their IT landscape.

Orion’s Enterprise Information Intelligence Graph (EIIG), a self-defined data fabric, can help overcome these challenges.

Complete coverage of all mainframe data assets

To start off, EIIG offers dedicated scanners for all types of data assets in the mainframe system, including JCL, COBOL, and PL1. When vendors claim that they support the mainframe, the questions one should always ask are: do you support all these programming languages? What percentage of the codes can you parse? How much is done manually? With EIIG, metadata ingestion from all these technologies and data sources are done automatically in near real time. EIIG enables organizations to discover what data they have, how it flows, and how it is used in their mainframe environment.

Stitching the most comprehensive knowledge graph of your data landscape

Since EIIG supports more than 60 technologies and data sources, as part of this intelligence knowledge graph, the mainframe is no longer an island of its own. Metadata from the mainframe is now weaved into a centralized data fabric that offers true end-to-end traceability and transparency, dynamic data lineage of all IT assets, and a robust data catalog with embedded metadata analytics.

With EIIG, mainframe customers can gain actionable insights more accurately and quickly. Fact-based decision making allows them to achieve desired business outcomes including:

Modernization/Migration done smart

Though mainframes are not going away soon, the call for their modernization becomes urgent for organizations to reduce cost and deal with the dwindling pool of mainframe talents. One of the common initiatives is to migrate some of the workloads to cloud-based data warehouses such as Snowflake or Databricks for more agile analytics and computing usage.

EIIG helps accelerate migration readiness assessment and the actual implementation by providing the following capabilities:

• Thorough understanding of the mainframe environment: EIIG offers a full inventory of your mainframe data assets including codes and reports within hours to make sure every report, dashboard, metric, dimension is accounted for, the source tables are identified, and the transformation logic is documented.
• Impact analysis and change management: With EIIG, you can understand the impacts of any change within minutes. It can provide snapshot versions of the data landscape between iterations, enabling running comparisons and doing impact analysis for root cause analysis.
• Solving the data bloat problem: Automatically identify information being stored with no use to business and redundant data assets such as duplicate reports, codes, and tables to dramatically reduce the scope of migration.
• Identifying reusable data and setting up migration priorities: Empowered by AI/ML, EIIG helps determine which data assets can be reused in the new environment and set up migration priorities and approaches according to the popularity, criticality, sensitivity, and relevance to regulatory compliance.

With EIIG, mainframe customers can be reassured that their modernization project will be deployed on time and within budget.

Compliance ensured

With EIIG, customers have full traceability and understanding of their mainframe data assets to comply with regulations such as GDPR, BCBS-239, HIPPA, and CCPA. EIIG can also tag all sensitive information such as PII and automatically map compliance-related terms such as GDPR glossaries with technical assets.

One of EIIG customers leverages the platform to show full traceability of millions of data points as it travels through their various systems including the mainframe, Oracle, MS SQL, Greenplum, Java, IBM Datastage, Qilk and Tableau. EIIG automated data discovery for data lineage, term-to-asset mapping, and better workflows around data management. It was able to perform at the required scale and handle the large volume of transactions per day, as no other vendor could.

Here is another mainframe specific example: a large bank used EIIG to achieve detailed auditability of all the data flowing through their mainframe system. EIIG was able to scan 70 million lines of PL1 programs running data transformations to extract metadata for analysis. As a result, the client prevented another compliance violation fine of a huge amount.

Self-service and collaboration facilitated

With metadata analytics-imbued data catalog in EIIG, data users can easily discover and access the data they need. EIIG also propagates trust and understanding of data quality so that business ready data is at the fingertips of the data consumers. An EIIG client now can build more efficient risk management model to enhance their AML program because the data they use for the model is more accurate and trustworthy thanks to EIIG.

Since EIIG breaks down data silos, now different members of a team and different teams can work together in this centralized environment with common understanding of the data they are working with, and transparency needed to encourage and expedite collaboration.

Cost reduction guaranteed

With EIIG, cost saving is immediate and tangible. In the above-mentioned example, the bank client saved a lot of time and resources, because they no longer needed to use hard-to-find PL1 programmers and spend years just trying to decipher 70 million lines of code.

In the data migration scenario, cost saving is realized in several ways. What EIIG enables is a metadata driven “migration factory” approach. EIIG gives a visual presentation of mainframe data usage without any manual effort. This understanding is one of the fundamental steps in assessing migration readiness. Also, as pointed out, by getting rid of data bloat (some mainframes applications are notorious for large amounts of reductant code) and identifying data assets to reuse, savings are showing even before the implementation starts.

During the actual migration, EIIG’s AI/ML powered automation helps reduce cost and alleviate resource constraint. For example, Javascript is widely used in cloud-native systems such as Snowflake. With EIIG, business users can see in a knowledge graph the logic that’s been coded in Javascript and can provide feedback to the developers of Javascript modules, without the need to understand Javascript. The elimination of the need for hiring or training more Javacript developers saves time and money.

Impact analysis is another tool to ensure a smooth migration. Automated and accurate impact analysis gives a central point of control and removes fear of changes. EIIG identifies the dependencies between datasets and executables that are impacted by a proposed change. The ability to make small, reliable, and frequent ensures fewer surprises and wasted efforts during the migration.

In summary, Orion’s Enterprise Information Intelligence Graph, a self-defined data fabric, is beneficial to mainframe customers in helping them unlock the value of their data, comply with regulations, modernize, and reduce costs.

About the author: Niu Bai, Ph.D. is the Head of Global Business Development and Partnerships at Orion Governance, Inc.