2022 WWDVC thoughts

This year’s World Wide Data Vault Consortium (WWDVC) conference– held in Vermont – provided the first opportunity for the movers and shakers in the Data Vault world to get together face-to-face in three years.

Of course since the last event the world has changed considerably, in no small matter because of the COVID-19 pandemic which forced the 2020 event to be cancelled just weeks before it was due to be held as world travel shut down.

After the 2021 WWDVC was held virtually, the organisers were able to offer a hybrid format for 2022 and it proved to be – in Datavault CEO Neil Strange’s experience and opinion – the strongest in terms of content across the board over the seven years he has been attending.

Not only was there a full schedule of workshops, presentations and keynote addresses on all five days at the Stoweflake Resort, in Stowe, but they were also relayed around the world online.

Billed as the conference for Business Intelligence and Analytics professionals, the 8th WWDVC was certainly interesting to catch up with many old faces and to finally have the opportunity to network and make many more new acquaintances and contacts.

So, what were the key emerging trends that the 2022 WWDVC unearthed in the Green Mountain State, and what features and factors will drive the constant race to develop and define the most effective data platform that meets businesses’’ demands and needs in the 21st Century?

Certainly, the debate over whether a Data Mesh approach will prove to be the latest “Silver Bullet”.

Most interestingly, what emerged over the five days in Vermont was that rather than being seen as a rival or replacement for Dan Linstedt’s Data Vault 2.0 methodology, there were plenty of examples given where large-scale organisations and businesses have already created a working platform using the Data Mesh architecture that incorporates a Data Vault. And the evidence so far confirms that rather than being rivals, they can be used to complement each other. Not only can they work side by side, but the Data Vault can be utilised to increase the power of the data analytics provided by the domain-based nature of a Data Mesh. Preventing a scenario where data is locked within silos, adding nothing to the business in terms of insight or profit.

The message emerging from those practitioners and data architects who have produced effective working models is that Data Mesh and Data Vault are complementary. An excellent case study from Roche Pharmaceuticals was presented explained how they have implemented Data Mesh at scale with Data Vault as a key underpinning method.

Zhamak Dehghani’s new book Data Mesh: Delivering Data-Driven Value at Scale is currently a hot topic of conversation.

Jacek Majchrzak, who works in the drug discovery sector, is the co-author of ‘Data Mesh in Action’ and promotes domain-driven designpresentedat the July meeting of the Data Vault User Group. The group will be discussing the use of Data Mesh further over the coming months.

Another interesting development in the use of Dan’s Data Vault design– was how it can be used to boost data science applications in different organisations and businesses.

Here at Datavault, a strong point in our promotion of the Data Vault 2.0 principles is how its correct implementation and use can improve an enterprise’s data science operations by helping to convert data science insights about the business and its customers or users into analytics that can be used by the wider organisation. More recently experiments to push Data Vault’s boundaries even further have found that the structures of the vault itself can be used to drive metadata to feed the data scientists’ work.

So, for instance rather than just examining a table of customers which data scientists can use to gain insights, within the Data Vault, that table could have links to X or Y and an analysis of information about the information – in other words – an analysis of the structure can provide more insights into your data, than previously discovered by using that metadata.

The growth in big data sets partially driven by the increasing digital transformation of businesses, the explosion in the Internet of Things (IoT) and the migration of many data warehouses in the Cloud – is also leading to a new surge in the volumes of data.

Furthermore there is a growing demand for businesses wanting to trade and share increasingly large sets of data. With the cost of, and complexity in, moving or replicating such data sets around the world, the ability to share them virtually, means firms can now concentrate on selling access and use of such data, without paying expensive computing and storage costs on top. Companies can now just concentrate on storing data that is completely unique to the business, while other information can be acquired and used, only when it is needed.

In reality, what the new commercial opportunity is also revealing is that such data sets have to offer industrial robustness – both in terms of being point-in-time and version-controlled – while in practice many are not.

The future picture is looking like a collection of federated sets of Data Vaults, sometimes even from different suppliers connecting to one another. Which brings us neatly back to Data Mesh which stresses that a business or organisation’s data should be federated, with strict data governance and lineage, owned and managed by the business users, on a strictly “need-to-share” only basis.

In conclusion, the 2022 WWDVC was easily the best in our opinion, and if you have not managed to attend one in the past, keep an eye out for confirmation of the dates in May 2023, and get it in your diary.