Scaling SQL for Cloud Native – Yugabyte CEO discusses the desirability of a new database class
Yugabyte aims to give enterprise buyers the best of both worlds when it comes to choosing their next generation of databases. In the on-premises world, SQL relational databases dominated, but that thinking has since been challenged by a wave of NoSQL adoption in the cloud – led by MongoDB and DataStax (Cassandra).
The general message over the past five years has been that relational models are not suitable for scale in the cloud, while NoSQL databases are designed to distribute unstructured data for internet applications. However, what if you didn’t have to compromise on things like transaction consistency, while taking advantage of cloud characteristics like resiliency, scale, and geodistribution?
Yugabyte’s founding principle is that NoSQL databases will aim to become fully transactional, but also that a cloud-native SQL database capable of handling multiple data models without compromising transactions, performance and geodistribution could be superior.
Yugabyte was founded by a team from Facebook, after successfully growing the social media company’s platform from millions to billions of users in just a few years. Yugabyte has since opted to build on PostgresSQL, an open-source database that has been around for decades, and rebuild the “top half” to adapt to cloud trends.
The company “NewSQL” (a term often used to describe the market in which Yugabyte operates), has since received over $290 million in funding and has been valued at over $1.3 billion. Not only that, but he has client names that include Kroger, Plume, and Admiral.
diginomica had the chance to sit down with Yugabyte CEO Bill Cook and VP of Strategy and Solutions Suda Srinivasan last week to discuss how the company views its go-to-market strategy. in the market and opportunities for buyers.
Cook explains that the success so far is due to a number of cloud-based market dynamics. He said:
The database tier, for the most part, was not designed or built for this era – cloud native models – which means you can hook up one database service per micro service or application. And then obviously resilience and redundancy and geographic distribution, all those kinds of attributes as well, are important in today’s world.
[The way the Facebook team] basically went about it by saying “what the world doesn’t need, or want, is yet another database to deal with”. However, they require the cloud-native attributes of the underlying database. And so the idea was, on the SQL side, to take the top half of Postgres and give it a more robust foundation, for cloud-native resiliency and scaling attributes.
The idea here is for Yugabyte to solve database problems for cloud-native environments, while making it easier for developers who have extensive experience using SQL tools to adapt. Yugabyte is designed to be infrastructure agnostic – so the company is focused on Google Cloud, AWS and Microsoft – but is also delivered as a service, managed service or can be hosted in your own environment.
Srinivasan reinforced Cook’s point, but added that companies that had ventured into the NoSQL world to find an alternative to relational databases are now knocking on Yugabyte’s door. He said:
Many companies modernizing their databases have been using traditional relational databases for a very long time, but they are looking to move to something resilient, scalable, geodistributed, cloud native, ready for the future.
And then we have other companies that have had a severe pain in terms of scaling or resiliency with traditional SQL, have gone the NoSQL route, and then realize they have trade-offs.
For example, a major retailer now has their product catalog running on Yugabyte. They had it on Cassandra, and they found they weren’t getting secondary indexes, they couldn’t do very fast searches. Maintaining this material took a lot of effort.
Cassandra is not a system designed for large-scale operational simplicity, in many respects. And they were losing money because of ACID transactions and lack of transaction capabilities. So they were looking for something that would allow him to combine these two worlds together.
Understand the use case
Cook and Srinivasan said the need for a relational database capable of scaling in the cloud becomes increasingly critical as customers explore multiple use cases supporting IT workloads from edge, 5G and IoT. Doing this with traditional databases becomes expensive and complex, while NoSQL providers run into limitations, according to Cook. He said:
There are tradeoffs you make when upgrading to a Mongo or a Cassandra, aren’t there? You give up the basic database transactional features we talked about with the traditional version. On the transactional side, you run into limits on MongoDB or Cassandra. That’s fine for some apps, but core apps and transactional consistency…we’d say, why compromise either side of that equation if you don’t have to?
Srinivasan said the customers Yugabyte works with generally fall into two categories. He added:
We have the top 2000 global companies going through the database modernization process. They’ve spent the last 10 years on a digital transformation journey, and they’ve sort of focused on application modernization, microservices, and so on. And for them, they have a database infrastructure that is unlike the rest of their infrastructure stack. And they’re looking at data layer modernization, it’s kind of a holistic approach: 2,000 databases, next five years we’re going to modernize, where do you start as a workload?
Then you have another group of companies building microservices to meet immediate needs. So they’re basically going cloud in part of their organization and they’re looking for a database that will be a perfect fit for that world. And there, the pain point is usually more acute.
Finally, another key selling point for Yugabyte, notes Cook, is the ability to invest in talent where needed. He hopes Yugabyte can prove its worth at scale, while making it easier for SQL developers to transition to a new platform. Cook said:
The other nuance for me is if you think about people, I’ve spoken to many companies and people are their biggest problem. How can I find qualified people to help me move my business forward?
Database administration shouldn’t be a core skill I want to invest in, frankly. I just want a database that works and scales automatically in a cloud environment and as a service for my developers. I want as many developers to build functionality that I bring to my customers instead of worrying about the underlying infrastructure.
It’s still early days for Yugabyte. And we haven’t spoken to any clients yet, although the names he talks about publicly are impressive. Interestingly, Yugabyte started at the top of the enterprise market in terms of extended use cases for major global brands, whereas NoSQL vendors generally started out fostering the developer community and then grew. It’s too early to tell, but it’s just an interesting observation.
I can see the appeal of a company that has traditionally used relational databases now being offered an alternative to MongoDB or Cassandra, claiming that they can do everything they did before, but in the cloud . It’s a compelling proposition, but like I said, it’s just the beginning and we need to talk to a wide range of customers to better understand the points of proof.
My initial assessment is that the Mongo and Cassandra customers I speak to are generally quite happy, but aren’t necessarily abandoning relational databases and using NoSQL for targeted use cases. Whether or not they trade in the NoSQL tools for something like Yugabyte, I’m not sure. That being said, moving their old relational databases to something like Yugabyte might be compelling. I have a feeling that, like most things, businesses will end up with a diverse set of tools, targeted to their needs – but we’ll be watching this market closely to see how the dynamics play out.