Planetscale Alternative: TiDB Serverless, Your Next Free DBaaS

Why TiDB Serverless is the right long-term choice if you’re migrating from your former provider’s free plan.

Change is constant in the world of developer toolkits. It’s beneficial because it enables us to deliver more value, faster. New technologies, products, and open-source projects propel this dynamic progress.

But it’s a different story when someone else’s decisions force change upon us. With the news of PlanetScale closing their free Hobby plan, it’s a good time to consider three things:

What are your options for migrating to a free, serverless, MySQL-compatible database offering?
Is now a good time to think about how to tackle new data problems, such as vector search for AI use cases?
Can you rely on free infrastructure?

We can start by looking at your options now that you need to find an alternative to PlanetScale.

1. Checklist for choosing your new DBaaS

Let’s be honest. You aren’t short of choice when it comes to hosted database tools. But comparing them can be tricky because each technology and provider makes different assumptions about data model, allowances, performance, integrations, and more.

Choosing a serverless database provider comes down to four big questions:

How hard will the migration be?
What’s it like to work with over the long term?
Can it perform when you need it to?
Can you rely on the vendor to provide a good user experience and to keep pace with new features, such as vector search?

Let’s unpack each of them.

1.1 How hard will the migration be?

No one wants to deal with an unplanned database migration. By definition, you haven’t set the time aside or put much thought into how it should go. So the first consideration in choosing an alternative to PlanetScale has to be how much additional work you’ll need to do.

So, the first four items on our migration checklist are:

✅ Is the data model the same? PlanetScale is built using Vitess, which adds scalability to MySQL. Many alternatives use PostgreSQL or entirely different data models, such as MongoDB’s document model. Choosing one of those alternatives could require costly schema migration and changes to your application code.

✅ Is the SDK or driver you’re using compatible with the new DBaaS? Similarly, can you continue to use the same library in your application or will you need to integrate with a new SDK?

✅ Are there tools available to help migration? Replicating a live database to a new provider isn’t to be taken lightly. How can you minimize downtime while making the move?

✅ Does it offer the same functionality? For example, PlanetScale supports non-blocking schema changes. If you’re used to making schema changes without downtime, you’ll need to migrate to a DBaaS with that capability.

1.2 What’s it like to work with over the long term?

Once you’ve made the move, can you be sure that you’ll get the developer experience and customer service that you need?

✅ Is it future-proof? Choosing a DBaaS isn’t just about the features it has now but also its long-term roadmap. Does the vendor’s vision align with what your application needs? For example, does the vendor have a plan for vector search?

✅ Does it integrate with the tools you need? Check for ready-made integrations with the tooling that’s important for your application. And if they aren’t available, how easy is it to write your own?

✅ What happens if something goes wrong? Free plans tend to offer minimal support but you need to know what’s available should you need help.

✅ Is there a developer community? Will you be able to rely on the experience of other developers who already work with the DBaaS? What is the ecosystem like?

1.3 Can it perform when you need it to?

The promise of serverless is that you don’t need to worry about scaling, resilience, or the rest of the operational side of running a database tool in production. While most free plans are limited in size, traffic, and other factors, you need to know that the DBaaS you choose can meet your application’s demands.

✅ Does it scale seamlessly? With some offerings, you need to provision capacity manually. Others auto-scale. True serverless DBaaS providers shield you altogether from the underlying infrastructure. Although the free offering might limit your ability to scale, you need to know that the tool can support your growth even if that means paying later on.

✅ Does scaling require compromise? Scaling relational databases can mean sacrificing consistency, availability, or partition tolerance. What trade-offs will you need to make?

✅ Is it resilient? If one shard becomes unavailable, can the DBaaS fall back automatically to a replica?

1.4 Can you rely on the vendor?

It’s not just about the product but also the company behind it. You’re searching for a new database provider because the one you’ve been working with has made changes. So, can you look forward to a long-term relationship with the new vendor?

✅ Does a free plan make economic sense? Cloud vendors often remove free plans because their own costs are unsustainable. When building TiDB Serverless, we designed a multi-tenant architecture to give us maximum resource efficiency.

✅ Are the paid plans reasonably priced? If you need to upgrade, is there a big leap from free to their first paid plan? Or can you make a smooth transition? For example, PlanetScale’s first paid plan is $39, whereas with TiDB Serverless you can start from $0.01 per month.

✅ Do they offer truly serverless billing? One controversial move by PlanetScale was to transition away from serverless billing to CPU, memory, and storage pricing. Serverless billing saves you from having to consider the underlying infrastructure.

Try TiDB Serverless

Start Instantly

2. Is free too good to be true?

The world-changing, economy-driving internet we have today exists largely thanks to “free”. With a few exceptions, the foundations of the internet are open-source software. The programming languages, frameworks, operating systems, and storage tools we use are almost all free to use. And those that aren’t entirely free often offer generous developer plans.

In the cloud era, providers use free plans to capture developer interest and establish a loyal user base. This strategy not only wins developer mindshare but also represents a calculated bet: some of these free users will achieve success and eventually transition to paid plans.

But free plans are viable only if the product makes efficient use of the underlying cloud resources. The cloud native architecture of TiDB Serverless is built to draw the most value from each cloud service on top of which it sits, without compromising scalability, resilience, or response times. That’s why we can offer a TiDB Serverless free plan and then charge only for the additional resources you use.

3. How does TiDB Serverless shape up against PlanetScale?

TiDB Serverless is a fully managed database as a service platform that ticks every one of the checklist items we covered just now. Specifically, with TiDB Serverless you get:

MySQL compatibility: Migrating from PlanetScale is super straightforward. Unlike Postgres-based alternatives, TiDB Serverless offers a drop-in replacement for PlanetScale’s Hobby plan.
Hassle-free automatic scaling: When demand on your application changes, TiDB Serverless responds in seconds.
No-ops approach: As a fully managed service, TiDB Serverless lets you focus on building and operating your application rather than worrying about the details of your database management system.
Ability to run mixed workloads: Whether it’s for transactional or analytical workloads, you can use a single database to do both.
Commitment to AI application development: You don’t need a separate database for vector search. Join TiDB’s waitlist to build AI applications with the SQL you already know.
Generous free tier: No credit card required to get started with up to five databases that each can store up 5 GiB. If you need more, pricing starts at only $0.01 per month.