The Data Cloud
1. The Data Cloud
Before we finish this module on Builder Workloads, I want to come back to an idea that we talked about in the very first video of the course, but I suspect might resonate more now than at the beginning. Here it is: “Snowflake is the Data Cloud.” If I had a mic, I’d drop it for effect. I guess I do have a mic, but it’s attached to… it would be hard to drop, and I think some people would get angry at me. Anyway, here’s my interpretation of what that means. I grew up in an era where the companies I worked for were usually already deep into their cloud migrations – If we had big servers sitting around in a closet, I was not aware of them. But many people more experienced than me remember times when everyone stored their data in on-prem servers that they owned, and the data was physically siloed – in different closets, in different regions, etc. But even those of us who grew up in the cloud era have experienced a slightly more nuanced version of data silos – You have one system that doesn’t talk to another. Maybe your customer relationship management tool (your CRM) doesn’t easily talk to your product usage data, and you’d really, really like it to; maybe you actually have multiple CRM instances, and it’s hard to get one to talk to another; maybe your company is so big that different teams have their own tooling, and possibly even use different cloud providers. These aren’t silos of the old sort, necessarily – all of these systems could be in the cloud – but they’re still silos, because the data still needs to be moved and copied between different systems for teams to have access to it. So at Snowflake, we think of the Data Cloud as having two elements: One, Snowflake’s platform – the technology itself required to bring data together and support the different development that runs against it. And two, the content - this can be data sets, machine learning models, or applications similar to what you built earlier. These come together to create a global network where orgs and users can connect to the relevant content, through a single platform. And the vision that we’re working towards is at a minimum one that eliminates the need for silos. One where any employee at a company with the right permissions can access and work with the content they need – again data, but also apps and models – when they need it. Without the complexity of moving it between silos or limiting the type of development they can do. This is a powerful vision, and it’s made possible by the fact that Snowflake makes available all the tooling we’ve covered in this course – and more. But there’s a more ambitious version of this vision, and this is the one I’m particularly excited about: Snowflake is building a world where organizations not only have zero silos within the org, but where orgs, through Snowflake, have bridges to other orgs. So company A can share data and apps with company B through the Snowflake Marketplace. No ingestion necessary – When data is replicated across clouds and regions, you can access it in a zero-copy way, almost as if it were a dataset in your own Snowflake instance. Plus those same access permissions and governance policies get carried through. You need weather data, but don’t want to spend weeks scraping it, cleaning it, and getting it into a shape where you can use it? Just join your tables to a weather dataset available in the same cloud and region where you’re already working. You’re a geospatial engineer and want access to a specialized tool? Find a geospatial app in the marketplace. Builders are essential to this vision. Who is going to make the apps? Who is going to create the pipelines to keep these shared datasets up-to-date and accurate? Builders. Some marketing terms are just spin, and don’t have much substance behind them. I was skeptical of the term Data Cloud for a while, because I worried it might be like that. But I think it’s not – If you have all the tools you need to work with your data all in one place, *and* you can provide and use data and apps from other orgs, that’s a big deal. Add in a single governance model that cuts across everything, and you have something that you can use whether you’re experimenting or deploying to production. So Snowflake is the Data Cloud, and Builders are essential for making the most ambitious version of this vision happen.2. Let's practice!
Create Your Free Account
or
By continuing, you accept our Terms of Use, our Privacy Policy and that your data is stored in the USA.