One Diagram for MemSQL Rowstore and Columnstore Tables

MemSQL — SingleStore, under its original name — splits work between aggregator nodes that plan queries and leaf nodes that own data partitions. DbSchema connects to the aggregator over the MySQL-compatible interface and draws rowstore and columnstore tables together on one ER diagram.

Introspect the schema from the aggregator, the node that sees it all
Run the MemSQL dialect — SHARD KEY, CLUSTERED COLUMNSTORE — in the SQL editor
Browse table data with filters and paging, no shard topology knowledge needed
Capture a full schema baseline before a SingleStore migration

available for

ISO-27001 certified

Free Community Edition No signup required 15-day Architect trial included

Free Download

DbSchema drawing MemSQL rowstore and columnstore tables in one ER diagram

Trusted by teams at

Distributed Across the Cluster,
Assembled on One Canvas

A MemSQL deployment scatters its physical reality across aggregators and leaves, and its logical schema mixes two storage engines with different rules. Teams keeping such a cluster running — often while weighing the move to SingleStore — need one authoritative picture of what the schema is today.

DbSchema builds that picture from the aggregator's catalog: every table drawn, rowstore and columnstore distinguished, ready to compare, document, and carry forward.

Query, Diff & Document
a MemSQL Cluster

The schema work a distributed database still needs, from one desktop tool.

Run MemSQL dialect queries in the DbSchema SQL editor

The Dialect, Understood

Write and run MemSQL SQL — SHARD KEY, USING CLUSTERED COLUMNSTORE, stored procedures — with syntax highlighting and paginated results.

Compare MemSQL schemas across environments

Environments, Kept Honest

Compare MemSQL schemas across development, staging, and production, and generate MySQL-compatible migration scripts for the differences you approve.

Generate HTML documentation for a MemSQL deployment

A Record Before the Move

Export HTML documentation that distinguishes rowstore from columnstore tables — the baseline a SingleStore migration plan starts from.

Clusters Come and Go
The Model File Stays

DbSchema keeps the MemSQL schema in a model file independent of any node — designable offline, versionable in Git, and applied only after review.

Edit With the Cluster Down

Refine the schema without a live aggregator anywhere in reach.

Version the Structure

Keep the model in Git and give schema changes the same review trail as code.

Read generated SQL before it reaches the cluster

Scripts Before Actions

Each planned change appears as SQL to read before anything executes on the cluster.

Roll out MemSQL changes environment by environment

Roll Out on Your Terms

Apply approved changes environment by environment, at the pace the cluster allows.

MemSQL Connection: Protocol, Port, and the Aggregator Rule

MemSQL speaks the MySQL protocol: choose MySQL as the database type and DbSchema downloads MySQL Connector/J automatically. Connect to the aggregator node's host on port 3306 — the URL is jdbc:mysql://host:3306/dbname. Deployments on a custom port record it in the port setting of memsql.cnf.

Always target an aggregator rather than a leaf: DDL issued on a leaf stays on that leaf instead of propagating across the cluster. The aggregator's address is listed in the management console's cluster topology view. Driver notes are on the MemSQL JDBC driver page.