For more than three decades, Logitech have been pioneers in the creation of keyboards, mice, trackballs, microphones, and other peripherals; but that was not enough. Video collaboration, gaming, digital home, and security camera technologies were becoming mainstream, and we could not miss the opportunity to jump on the bandwagon. We knew that in order to master the creation of such technology, hardware expertise would not suffice. We needed predictive analytics, real-time analytics, natural language processing, sentiment analysis, and machine learning. We needed these capabilities not only for digital home applications and gaming, but also to strengthen our demand planning, product pricing, inventory analysis, competitive product analysis, and other key efforts. Advanced analytics and machine learning became inevitable.
Journey to The Cloud – Often a Bumpy Ride
But to get there, we needed to start with modernizing our data services, which we used to deliver using on-premise systems. Each individual system (DRM, MDM, ERP, and POS) functioned as a data silo, so gathering insights was inefficient and unreliable, and our IT department wanted to move to the cloud.
We knew this would be challenging; when companies of our size move to the cloud, it can take a while to become completely cloud-based. For many months, or sometimes years, enterprise information systems need to work in a hybrid fashion – partially in the cloud and partially on-premise. In fact, some companies actually want to work in a hybrid environment forever, as they are reluctant to let their most sensitive information float in the public cloud.
Making on-premise and cloud systems work in harmony is often very challenging, and it requires numerous hours or days of downtime. While cloud technologies can reduce the siloed nature of on-premise system to a certain extent, we needed another solution to act as a data integration fabric to hold all of the enterprise information together. We also wanted this solution to integrate the data in real time, as our business is very dependent on time-sensitive data types such as streaming data, social media data, and transactional data.
Data Virtualization Makes the Journey Easy
After an extensive search process, we chose the Denodo Platform for data virtualization to be our data integration fabric. We envisioned a Nirvana state, in which all of our enterprise source systems and consuming applications would be connected through the data virtualization layer. That way we could spin up any new application or infrastructure component in our AWS cloud in a matter of hours instead of days or weeks. Currently our Smartphone, Gaming, Security Video, and many other business units use the Denodo Platform and AWS cloud platform to provision internal and external data to perform data mining and natural language processing. The solution architecture is at such a mature state that any new components can be added to the cloud platform as easily as Lego blocks, on an on-demand basis, to augment data analysis, and any component can be swapped out without disrupting day-to-day business operations.
This modern, agile, cloud based data platform has enabled our business users to consume data in a self-service manner. We have created a business glossary and catalogs so that business users can search for – and easily find – the information they need. Our cloud based analytics data is fed into applications and devices such as Tableau, Pentaho, Alexa, or our Natural Language Processing (NLP) Engine. Business users use the analytics to understand competitive product offerings and recommend product feature enhancements, execute supply-and-demand analysis based on various product line inventory levels, execute end user sentiment analysis, and many other critical activities. We are adding more and more intelligence to this data platform so end users can find information blazingly fast to make business decisions in real time.
Watch our story here:
Moving Towards a Better Future
Before we started our cloud modernization effort, our data provisioning took place in a reactive mode. IT was dependent on data extraction for sharing information with business users, which delayed the information, impacting business transformation, innovation, and revenue growth.
Now, our IT team is much more proactive in offering services to our business stakeholders and has embraced cloud technology as a model for achieving innovation through increased efficiency, reliability, and agility. IT has structured the enterprise data architecture for rapid innovation across business segments and enterprise operations, with proper governance, security, and access control in place, and in such a way that we can offer the best service levels to internal and external stakeholders. Our IT team has also enabled a broader audience, both internally and externally, to consume IT services via a self-service model, with greater efficiency and reliability. All of these achievements have been made possible by our modern, innovative solution architecture that embraces cloud technology, with data virtualization at the core.
I often am invited to speak about our enterprise data architecture, and I have seen immense interest from various organizations across a broad range of industries to learn from our solution and replicate our model. In the past two years, we have made significant progress toward our Nirvana state. I envision a great future for our IT team and our company in which we are more than a hardware company, we are a pioneer in digital innovation.