DS acquires NuoDB for cloud-native distributed SQL

Nov 30, 2020 | Hot Topics

I hope you had a lovely and peaceful Thanksgiving if you’re in the US, and a bit of hush to hear yourself think if you’re somewhere else.

Last week Dassault Systèmes announced that it was acquiring the 84% of the equity of NuoDB that it did not already own. NuoDB makes a cloud-native distributed SQL database ” with on-demand scalability, continuous availability and transactional consistency, [that] is built for mission-critical applications”. NuoDB was founded in 2010 and its relationship with DS stretches back to 2013, so the companies know one another well. There’s no indication in the announcement of why an acquisition was necessary, or why now is the time. Price and other terms were also not mentioned, but DS did say that it expects the transaction to close in December and that NuoDB had revenue of about $6 million in 2019.

I don’t think I knew about DS’ minority stake in NuoDB so did a bit of research. First, SQL. SQL stands for Structured Query Language and is a computer language designed to manage and retrieve data in a relational database. Relational database? A database that stores (and, of course, provides access to) bits of data that are related to one another via a unique ID. The columns of the table hold attributes of the data, and each record usually has a value for each attribute, making it easy to establish the relationships among data points.

An example? Today is Cyber Monday, so let’s look at shopping. You probably get emails from retailers with whom you’ve done business in the past. Your name, email address, and billing information, each in a column in a database and the whole record identified by your customer number. Another database (separate for security reasons) keys your customer number to orders you’ve placed — dates, item numbers, quantities, etc. Today, the merchant wants to promote something adjacent to what you last bought and so uses SQL to interrogate the databases for item numbers and winds that back to customers’ email addresses. The end-result: you get a targeted email that points you to something you’re likely to be interested in.

The CAD/PLM world is similar. If you use PDM, for example, CAD data and other documents are stored in a central vault –that’s usually a relational database– to manage all the files and references, their connections to one another, permissions, workflows, and so on. Complex CAD products may actually create a database per part, to manage assemblies. Unless you’re using a very simplistic CAD tool, you’re probably working with a relational database. Yes, even AutoCAD can be used with one.

Relational databases have been around in one form or another since the 1980s, usually in huge data centers, behind heavily protected firewalls. Safe, but ultimately hard to work with. Moving this all to a cloud and SaaS environment (while maintaining that level of security) means more availability, elastic processing, and all of the other cloud-y attributes.

NuoDB claims that it enables clients to “move mission critical applications from legacy databases to modern architectures, so … organization can capitalize on the competitive advantages gained by cloud adoption—and meet … customers’ current and future needs.”

For its part, as a NuoDB customer DS offered this quote: “NuoDB is at the forefront of a wave of SQL database innovation, providing customers such as ourselves with the flexible, scalable database foundation we need to keep pace with the demands of today’s business environment.” It’s unclear from that endorsement exactly what DS was using NuoDB for — as a component technology that it sold on to customers, or in its own day-to-day operations.

I’ve been told that relational databases are really hard to move from on-prem architectures to the cloud — that’s why SAP had to reimagine its architecture to take its ERP solutions to the cloud with S/4HANA, for example, and why many CAD and PLM products are only partially in the cloud today. To be fully cloud-native means a rethink of the basic nuts and bolts of how the CAD or PLM (or CAE or or or) stores and accesses data.

Our PLMish world will need to adopt new technologies that enable cloud-based data management, including where the underlying database is stored and how it is accessed. In addition to DS and NuoDB, PTC is leveraging the technology it acquired with Onshape into the Atlas platform that will someday bring Creo and Windchill to the cloud. Siemens PLM introduced Teamcenter X, SaaS applications that run on top of the existing Teamcenter platform. Autodesk has Fusion 360, which has many SaaS and cloud elements but still uses Vault as its data management layer. [That’s only partly correct — see below for Autodesk’s correction.]

The question remains, of course: when will traditional manufacturing customers be ready to jump to the cloud? One of the few silver linings in 2020’s life-under-Covid realty is the discovery that we all work better when we can easily access the data we need, no matter where we and it are. That should move even the more conservative enterprises in our universe to at least consider cloud — and now the vendors must race to capture that demand.

Update: someone with an obviously fake email address commented as follows:

Autodesk has Fusion 360 together with a complete native SaaS data management collaboration layer powering it all, and not its on-prem Vault solution.

Please, whoever sent this, use your real email. You can use a fake name, since that is visible to readers but I cannot approve comments from faux addresses because of WordPress vulnerabilities.

Readers, I’ve spoken with Fusion 360 users who use the built-in tools for collaboration, and use Vault for more comprehensive data management. If you know otherwise, please chime in with a (verifiable) comment!

Update #2: The very helpful Jason Love of Autodesk sent in the following correction:

In the case of Fusion 360, all [of] a project’s data is in the cloud (including PLM data that’s used by Fusion Lifecycle). There is some integration between Fusion and Vault, but Vault is really Inventor’s PDM solution. The Fusion 360 Manage Extension (coming in January) will increase Fusion’s PDM / PLM functional sophistication.

So, more accurate today would be to say: Autodesk has Fusion 360, which entirely stores data in the cloud and is almost entirely SaaS, but which can also tie into data stored in Vault.