Softwr

Databases · head to head

Apache Doris vs turbopuffer

Apache Doris logo

Apache Doris

Databases

MPP analytical database with a MySQL wire protocol and sub-second aggregation on wide tables

From
Free
Rated
-
Full reviewWebsite
turbopuffer logo

turbopuffer

Databases

Closed-source vector and full-text search service built directly on object storage, with cold queries measured in seconds rather than milliseconds.

From
$16/month
Rated
-
Full reviewWebsite

The short version

  • Only Apache Doris has a free tier, so it costs nothing to try first.
  • Each has a real cost: Apache Doris two competing commercial vendors, VeloDB and SelectDB, were founded by overlapping core contributors, which makes the long-term governance and roadmap of the project harder to predict than a single-sponsor project.; turbopuffer a cold namespace pays object storage latency on the first query, with a documented p90 around 1,214 ms on a million documents, so any interactive search box needs the data kept warm or the user waits about a second.
  • They diverge on capability: Apache Doris covers MySQL wire protocol, turbopuffer covers Object storage architecture.
  • Prices and features above were last checked on 31 August 2026.

Where they differ

Only the attributes on which Apache Doris and turbopuffer actually diverge.

Attributes where Apache Doris and turbopuffer differ
AttributeApache Doristurbopuffer
Starting priceFree$16/month
Pricing modelOpen source, no licence feesubscription
Free tierYesNo
PlatformsLinux, Docker, KubernetesWeb

Identical on both: user rating (Not yet rated), category (Databases).

What each one covers

Drawn from each product's published feature list. An absence here means we hold no record of it - not that the product lacks it.

Only in Apache Doris

  • MySQL wire protocol
  • Aggregate and unique key models
  • Materialised views
  • Multi-catalogue federation
  • Routine load from Kafka
  • Compute storage separation
  • Inverted indexes
  • Workload groups

Only in turbopuffer

  • Object storage architecture
  • Namespaces
  • Vector search
  • Full-text search
  • Attribute filtering
  • Documented limits
  • Configurable consistency
  • Durable writes

What people use each for

The jobs each tool is most often brought in to do.

Apache Doris

  • A team whose MySQL read replica can no longer serve reporting queries and wants an OLAP engine its existing drivers already speaknot turbopuffer
  • A real-time dashboard backend needing sub-second aggregation over billions of rows with hundreds of concurrent usersnot turbopuffer
  • An ad or ecommerce platform that needs updates and deletes on analytical tables, which append-only OLAP engines handle badlynot turbopuffer
  • A data team that wants one SQL endpoint over both internal tables and existing Hive or Iceberg tables in the lakenot turbopuffer

turbopuffer

  • A product with one search index per customer and thousands of customers, most of whose data is idle on any given daynot Apache Doris
  • Very large corpora where holding every vector in memory is the dominant cost and occasional cold-query latency is acceptablenot Apache Doris
  • Hybrid retrieval combining BM25 and vector search where running and synchronising two separate systems is the problem being solvednot Apache Doris
  • Retrieval for agent and assistant products where indexes are created and destroyed frequently and per-index overhead must be near zeronot Apache Doris

Where each one falls short

Documented limitations, not opinions. Every one is a constraint you would hit in normal use.

Apache Doris

  • Two competing commercial vendors, VeloDB and SelectDB, were founded by overlapping core contributors, which makes the long-term governance and roadmap of the project harder to predict than a single-sponsor project.
  • A large share of design discussion, issue reports and documentation detail originates in Chinese, so teams that do not read it get a thinner picture of known problems and workarounds.
  • Operating a cluster means managing frontend and backend node roles, tablet balancing and compaction tuning, and compaction backlogs under heavy upsert load are a recurring production complaint.
  • The MySQL protocol compatibility is at the wire level, not full MySQL semantics, so queries and functions still need porting and the familiarity can mislead.
  • Managed cloud availability outside China and major clouds is limited compared with ClickHouse or Snowflake, so many Western adopters end up self-hosting whether they wanted to or not.

turbopuffer

  • A cold namespace pays object storage latency on the first query, with a documented p90 around 1,214 ms on a million documents, so any interactive search box needs the data kept warm or the user waits about a second.
  • Queries are eventually consistent by default, and after roughly 128 MiB of outstanding writes new data is invisible until indexed, which the vendor puts at tens of seconds for small namespaces and tens of minutes for large ones, so a bulk re-index is not immediately queryable.
  • It is closed source with no community edition, so single-tenant or bring-your-own-cloud deployment is a commercial negotiation rather than a deployment choice, and there is no path to running it yourself if the relationship ends.
  • Per-namespace ceilings, roughly 10,000 writes per second, 32 MB/s and 500 million documents per shard, mean a single enormous index has to be sharded across namespaces by your application rather than by the service.
  • It is a search engine, not a database: there are no joins, no cross-document transactions and no SQL, so it sits beside a primary datastore and keeping the two in step is work that belongs to you.

Pricing, plan by plan

Apache Doris

Free
  • Apache DorisFree
    • Apache 2.0 licence with no usage restrictions
    • All engine features included
    • Community support via mailing list and Slack

turbopuffer

$16/month
  • Launch$16/month
    • All database features
    • Multi-tenancy deployment
    • SOC2 & GDPR-ready DPA
  • Scale$256/month
    • Everything in Launch
    • HIPAA-ready BAA
    • Single Sign-On (SSO)
  • Enterprise$4096/month
    • Everything in Scale
    • Single-tenancy & BYOC deployment options
    • Private networking

Which should you pick?

Choose Apache Doris if

  • You need mysql wire protocol.
  • You want to start without paying.
  • You work on Linux, Docker, Kubernetes.
  • You also want aggregate and unique key models.
Apache Doris review

Choose turbopuffer if

  • You need object storage architecture.
  • You also want namespaces.
turbopuffer review

Questions people ask

Is Apache Doris or turbopuffer better?
Neither clearly leads. Apache Doris starts at Free and turbopuffer at $16/month, and user ratings are close enough to be indistinguishable. Choose on capability and platform support.
Which is cheaper, Apache Doris or turbopuffer?
Apache Doris has a free tier; the other does not. Paid plans start at Free for Apache Doris and $16/month for turbopuffer.
Does Apache Doris or turbopuffer run on more platforms?
Apache Doris runs on Linux, Docker, Kubernetes. turbopuffer runs on Web.
Can I use Apache Doris for free?
Yes. Apache Doris has a free tier, so you can try it without paying. turbopuffer starts at $16/month.
What is Apache Doris best used for?
Apache Doris is most often used for a team whose mysql read replica can no longer serve reporting queries and wants an olap engine its existing drivers already speak, a real-time dashboard backend needing sub-second aggregation over billions of rows with hundreds of concurrent users, an ad or ecommerce platform that needs updates and deletes on analytical tables, which append-only olap engines handle badly, a data team that wants one sql endpoint over both internal tables and existing hive or iceberg tables in the lake. Of those, a team whose mysql read replica can no longer serve reporting queries and wants an olap engine its existing drivers already speak and a real-time dashboard backend needing sub-second aggregation over billions of rows with hundreds of concurrent users are not what turbopuffer is typically brought in for.
What can Apache Doris do that turbopuffer cannot?
Apache Doris covers MySQL wire protocol, Aggregate and unique key models, Materialised views, Multi-catalogue federation. turbopuffer covers Object storage architecture, Namespaces, Vector search, Full-text search.

Answered from the vendors’ own pages

Apache Doris: Is Apache Doris really free?

Yes, it is Apache 2.0 with no usage restrictions. The commercial products are managed services from VeloDB and SelectDB.

turbopuffer: Can I self-host turbopuffer?

There is no open source or community edition. Single-tenant and bring-your-own-cloud deployments exist as commercial arrangements, but there is no way to run it independently of the vendor.

Apache Doris: Can I use my MySQL tools with it?

Yes, it implements the MySQL wire protocol, so clients and BI connectors attach without a new driver, though SQL semantics differ.

turbopuffer: How fast is it really?

Warm queries perform comparably to in-memory search engines. Cold queries, where data is not cached, have a documented p90 around 1,214 ms on a million documents. Write p90 is around 248 ms for a 512 KB upsert because writes go straight to object storage.

Apache Doris: How does it compare with ClickHouse?

Doris handles updates and high concurrency more comfortably; ClickHouse is generally faster on raw single-query scan throughput and has far wider Western support.

turbopuffer: Is it consistent?

Eventually consistent by default, with the vendor reporting that over 99.8% of queries return consistent data. Strong consistency can be requested per query at a latency cost. Large write bursts have a longer visibility delay while indexing catches up.

Apache Doris: Who maintains it?

The Apache Software Foundation project, with most committers employed by VeloDB or SelectDB.

turbopuffer: What is it best at?

Large numbers of namespaces where most are idle. The architecture makes cold data cheap to keep, which is exactly the shape of a multi-tenant product with a long tail of inactive customers.

turbopuffer: What are the hard limits?

Up to 128 billion documents and 256 TB per namespace, 500 million documents per shard, 64 MiB per document, 10,752 dense vector dimensions, roughly 10,000 writes per second per namespace and a maximum result set of 10,000.

Share

Related pages

More on Apache Doris

More on turbopuffer

Keep looking

Other head to heads