Building a scalable knowledge base for Lenovo

Lenovo is one of the world's leading technology companies, producing innovative PCs, servers and mobile internet devices. With more than 63,000 employees, it serves customers in more than 180 markets and operates 46 world-class labs.

The Lenovo Infrastructure Solutions Group (ISG) is the server side of Lenovo and has established itself as the world's leading provider of Supercomputers, according to Top500.org. They also hold the top spot in terms of reliability for x86 systems and are a leader in performance benchmarks.

Through industry leading technology and global HPC (High Performance Computing), ISG seeks to drive forward breakthroughs in computing systems and make it possible to overcome obstacles in simulation. Their solutions help customers of all sizes build a smarter way forward with analytics and AI by accelerating and optimizing their data, delivering faster, more cost-efficient insights, and improving AI-based decision making. The efficiency and innovation provided have been highlighted through the ISC Vendor Showdown and via multiple HPCWire Awards.

The Lenovo team was searching for a tool to migrate and accommodate the contents of their technical support community and knowledge base. We were aware from day one that it wouldn't be an easy project, mainly because of the short time frame for the implementation, and complex needs. After pitching our solution against other candidates, we were awarded the project and started working. From signing the contract to production day we had around two months and a half.

One of the important factors that contributed to XWiki being chosen was our team demonstrating the ability to migrate the data from Lenovo's team instance to XWiki, through a quick proof of concept. Together with the Lenovo team, we had very efficient meetings that led to the timely implementation of the project.

Lenovo chose to use XWiki 11.10.10 On-Premise (on their server), via Docker. The estimation for the migration was of 35 days, but our team managed to finish it in 29 days, keeping all the documents content, attachment, history, and comments. For this project, we have introduced 5 innovations, namely custom field authentification, reversed comments order, ratings, unique identifiers, and an external platform import.

All the pages, 6725 on the main wiki, were imported including their history (10 latest authors), likes (were converted into ratings provided by our Rating module), and files attached. The order of the comments that were ordered chronologically was reversed. The existing "communities" (independent documentation spaces in HCL Connections) were transferred to sub-wikis in XWiki and we provided integration with 3rd party analytics platforms (Matomo and Adobe Analytics) for a better overview of the pageviews.

For the 7000 users imported, we simplified the login process with a custom extension. They can now log in with the email address and not a username that can be easily forgotten. At the same time, all their existing rights on the pages were kept, and passwords reset. As one of their main needs was to be compliant with data privacy regulations, such as GDPR, the users' data is stored as unique identifiers.