A few years ago, I worked as an evangelist for a midsize software company, and sometimes had to ask my boss to fund a project that hadn’t been anticipated in the quarterly plan. As long as my reasons for the change were sound, he was usually accommodating, telling me, “Steve, money is fungible.”
He simply meant that funding could be moved from one department to the next, or a project here could be put on hold to help a more urgent project there. At a simpler level, “fungible money” means a dollar could be divided into four quarters, ten dimes, etc. Dollars can’t be produced out of thin air, but they can be re-allocated.
Of course, today we also have Non-Fungible Tokens (NFTs), which are linked to assets on a blockchain with unique identification codes and metadata that distinguish them from each other. These are assets, popularly images and video, that can’t easily be replicated nor transferred without permission. Non-fungible is more expensive, both for electronic art and database usage.
Vertica’s “fungible licensing” gives customers a way to choose how much of their total analytic resources to fund via one cloud vendor, how much for another, and if needed how much to fund on-premises in the data center. You purchase a license for the total size you need in terabytes, or in nodes – and how you spread that licensing across your available resources is totally up to you.
We believe this flexible approach to your overall Vertica licensing is how it should be when it comes to using data analytics software. Our competitors tend to disagree, forcing you to buy a separate license for different deployments. Often there’s a separate installable for separate projects. That makes no sense to us in a world where architectures are fluid for many companies.
Take a look at how two of the big players in the data analytics space handle licensing, and see how the non-fungible costs can really mount up quickly based on those pricing models.
Snowflake offers two licensing options – on-demand and pre-purchased capacity. Unlike Vertica, which offers you a license for the number of nodes you need, Snowflake requires you to buy fixed-size warehouses by node size in geometric progression (2, 4, 8, 16, 32, 64, 128). That means if you don’t get the desired performance running a 32-node size warehouse, the next immediate available option is the 64 node one, which will nearly double your cost.
Of course, performance improves linearly as the size of a warehouse increases. But do you need to DOUBLE your current capacity simply to gain the incremental performance you’re looking for? Vertica doesn’t believe you should be forced to do that. If you need exactly 11 nodes to complete your analysis, we can price that for your exactly, and that can save you money.
Remember, also, all of the fees you could encounter. If you have a lot of tough analytics, you might want to leverage Amazon’s finest, fastest compute nodes. Snowflake compute node options tend to be generic and fairly limited, so there is no way to pick top-of-the-line nodes. Separate charges might apply for storage, egress fees, machine learning and predictive analytics, geospatial, and so on.
The Vertica license is fungible, so why not noodle with different node types for your best price/value?
Your data will be fungible on Google BigQuery, too. Google BigQuery offers three different pricing models: on-demand, reserved, and flex pricing. If you need a data warehouse, you probably should not be using on-demand unless you do not need to scan a ton of data for each query. You should be using reserved slots with flex slots to reduce the costs of workload variations.
However, with Google BigQuery, pricing gets complicated. You may at any time be charged for 1) data processed; 2) number of flex slots; 3) storage with a separate hot and cold storage fee; 4) streaming data into the database; 5) number of queries; and 6) data egress. Unless you have a boring analytical workload, it’s nearly impossible to predict what you’ll spend month-to-month.
Reducing the licensing hassle for big data projects
Vertica pricing is more straightforward. You can license Vertica for as many terabytes or nodes as you need, then run Vertica on-premises and on multiple clouds – or in any combination – without additional licensing. You don’t pay separately for dev, test, and production. It’s all covered in one Vertica license.
Fungible licensing simplifies how you allocate, and re-allocate, your overall data analytics capability as requirements evolve. Need to shift a workload from the cloud to on-premises? No problem. Need to move some of your storage from Azure to AWS? It’s simple.
This is just one way Vertica is taking the administrative hassles out of data analytics projects, so you can spend less time on licensing arrangements, and more time delivering data-driven insights.