How a data-first approach can fast-track a Google Cloud migration

By: Vinu Russell Viswasadhas, Rajesh Ramachandran, Cat Perry

The potential of data has never felt as boundless as it does today.

The call for automated excellence, unparalleled insights and unmatched experiences echoes from the boardroom to the point of sale.

At the same time, the significant amount of underutilized data locked in mainframe systems can stand in the way of data-driven ambitions. After all, if you can’t reach the magic beans, how can you expect anything to sprout?

Enter a data-first approach.

Data-first is a strategy that allows companies to swiftly and securely tap into the compute power of Google Cloud while maintaining their core mainframe environments. It’s an attractive option for teams looking to explore emerging technologies, like AI, but are limited by their current computing frameworks. Data-first aims to make the essential data available for these technologies without the complexity—and cost—associated with more comprehensive modernization efforts. Think of it as data liberation as a service.  

Let’s look at how adopting a data-first approach can fast-track a team’s Google Cloud migration.

1. Assess the environment and define objectives

The goal of assessing your environment is to better understand existing data assets and how they are used through assessment tooling. Through this, your team will be able to make more informed decisions on which data you’ll need to copy over first, leading to a more efficient and impactful cloud transformation.

Identifying these data assets will depend on your objectives—and, by extension, your initial use cases. For most teams, this will require identifying, within a data democratization session, the business needs and pain points you want to target up top: the low-hanging fruit that promises the most significant and immediate impact.

For example, if you’re a retail company, your objectives might be personalization, data intelligence, operations, customer engagement and risk and fraud analysis. Suppose your team is in the manufacturing industry. In that case, you might focus on inventory management, fleet management and smart supply chain, while healthcare teams might instead choose to zero in on diagnostics, insurance and patient monitoring.

2. Find the right tools for the job

Identifying a suite of tools that will support your team on this journey is critical. The tools should deliver:

• Analytics and insights: Find an advanced analytics platform that enables your team to derive comprehensive insights into mainframe data.
• Data integration: Identify tools that handle high-volume transactions and allow the usage of data from multiple mainframe data sources in Google Cloud.
• Data security: Opt for services that prioritize data security and align with all of your data security and compliance needs.
• Cost benefits: Choose tools to provide a better return on investment while lowering the risk of mainframe data project failures.
• Data processing: Explore managed services for streamlining data processing pipelines to enhance real-time data handling and analysis.
• Data storage: Seek scalable cloud storage solutions to store and offload mainframe files and data.
• Managed databases: Consider managed regional database services to ensure efficient data management.

When selecting these tools, consider both their immediate and future applications. In the short term, they should help bridge the gap between your mainframe and Google Cloud by ensuring a seamless transition while also maintaining the integrity and accessibility of your data. One approach here might be to seek tools designed to integrate existing mainframe workflows and translate mainframe data formats directly into cloud-optimized formats.

In the long term, meanwhile, these same tools should support innovation and business insights by enabling your team to leverage your newly liberated data for applications such as analytics and machine learning.

3. Move the data  

Before executing the project, it’s essential to conduct a standardized health check to ensure only high-quality data makes the cut. Such a health check might eventually form part of a broader governance framework, keeping additional checks and balances—on, for example, security, performance, reliability, scalability and compliance requirements—at the forefront during this process.

Once this is complete, it’s time to bridge the gap between your mainframe and Google Cloud.

Your data is now successfully liberated—saving your team time, effort and cost by side-stepping the complexities and risks inherent to a full-fledged migration.

4. Explore the art of the possible

With your data now securely in Google Cloud and ready for use, your team can start iterating on the use cases you’ve identified.

Take the example of the retail company from earlier: With their data consolidated, they can begin to explore how predictive analytics, powered by machine learning, could bolster their operations during peak seasons. Or how data analysis in a Google Cloud-based data warehouse may uncover significant correlations and patterns during high-traffic periods.

This is the “art of the possible.”

Vinu Russell Viswasadhas is an Associate Director at Kyndyl; Rajesh Ramachandran is Global Mainframe Modernization and Migration Practice Lead at Google; Cat Perry is a Technical Solutions Consultant at Google.