Overview
What is Google Cloud SQL - SQL Server: Cloud SQL for SQL Server is Google Cloud's fully managed database service for Microsoft SQL Server, allowing project teams to run SQL Server workloads without managing the underlying infrastructure. Google Cloud automates backups, failover, replication, software patching, and capacity increases. According to the Cloud SQL documentation, instances scale to 96 processor cores, 624 GB of RAM, and 60 TB of storage, with an availability SLA of 99.95% on Enterprise and 99.99% on Enterprise Plus. Google's version support list, dated March 28, 2025, covers SQL Server 2017 through SQL Server 2025, with SQL Server 2022 as the default for new instances.
How to integrate Google Cloud SQL - SQL Server with Datagrid
Datagrid imports tables, views, and stored procedures from a Cloud SQL for SQL Server instance, but does not write changes back to Cloud SQL. Agents use the imported records to reconcile pricing, inventory, project, or client data across SQL Server, CRM, and ERP systems, generate summaries of detected mismatches, and route each exception to the appropriate team while project teams review and make final decisions. Follow this setup order: create a SQL Server login and a network path, choose the supported authentication method, then define the data sync scope and timing.
Create a SQL Server login and configure access
Setup needs a database user inside the instance and a network path to it.
Create a database user in your Cloud SQL instance for the connection. All SQL Server users on Cloud SQL need a password, per Google's Cloud SQL users docs.
Configure network access. SQL Server on Cloud SQL listens on port
1433. For public IP addresses, add the connecting source to the authorized networks; the instance firewall blocks all traffic by default. Private IP paths use Private Service Access or Private Service Connect.In Datagrid, use the instance connection details and the database credentials from step 1 to add the Google Cloud SQL-SQL Server integration.
Confirm which objects Datagrid imports from the instance.
Use this illustrative worksheet when preparing the Datagrid connection fields:
integration: Google Cloud SQL - SQL Server
instance: INSTANCE_CONNECTION_DETAILS
port: 1433
database_user: SQL_SERVER_USER
password: SQL_SERVER_PASSWORD
objects:
- databases
- tables
- views
- stored_procedures
- functions
- users
import_schedule: dailyReplace the placeholder values with the instance connection details, database user, password, and schedule for your environment. This worksheet organizes the required values; it is not an executable Datagrid configuration file.
Choose the authentication method
The connection uses a SQL Server login and password because Cloud SQL for SQL Server does not accept IAM database authentication. IAM authentication covers only instance and backup operations. Database operations use separate database credentials, per Google's IAM authentication docs. Google also offers customer-managed Active Directory as another database authentication option on the instance, with Managed AD setup and Microsoft Entra ID available.
Set the data sync schedule
The sync configuration establishes the imported objects, direction, and frequency. Datagrid syncs databases, tables, views, stored procedures, functions, and users. The import runs from Cloud SQL for SQL Server into Datagrid on a configurable schedule, with daily imports available. A window outside peak database usage keeps imports clear of production load.
Why use Google Cloud SQL - SQL Server with Datagrid
This integration gives project teams a consistent execution layer for database-backed reconciliation, reporting, and exception handling.
Current records for project team workflows: Each import refreshes SQL Server data in Datagrid for reconciliation and reporting.
Cross-system reconciliation: Project teams have agents compare imported SQL Server records against CRM and ERP data drawn from 100+ connected platforms.
Stable endpoint through maintenance: Project teams retain a managed database endpoint while Google automates patching, backups, and failover and minimizes planned maintenance downtime across its service tiers.
Condition-triggered workflows: Project teams have agents monitor imported data and trigger downstream actions when a condition, such as a stock threshold, is met.
Narrative reporting from raw tables: Project teams have agents turn query output into a weekly sales or inventory summary in prose, so reviewers receive a readable report without manual table formatting.
What you can build with Google Cloud SQL - SQL Server Datagrid integration
Project teams can turn structured database records into repeatable workflows across connected business systems.
CRM and ERP reconciliation: Project teams import customer and pricing tables into Datagrid, then point agents at connected CRM or ERP records. Agents flag duplicate customer records, outdated pricing, and inconsistent inventory counts, and each exception routes to the owning team.
Automated business intelligence reporting: Project teams configure an agent to read imported sales, inventory, or financial query results and push a formatted summary to a collaboration tool every week. Datagrid uses the same reporting pattern as with self-managed SQL Server, allowing imported records to feed trend analysis without anyone having to build pivot tables.
Compliance checking against database-stored rules: Project teams store inspection criteria or compliance requirements in a SQL Server table. Agents check incoming data and project files against those rules after each scheduled import and surface detected violations.
Post-migration data pipelines: Project teams migrate MS SQL Server workloads from on-premises servers to Cloud SQL using the Database Migration Service, gaining a managed instance and a ready-made agent pipeline. Datagrid handles extraction while project teams retain responsibility for exception handling and final decisions.
Resources and documentation
Cloud SQL docs: Instance creation, connections, high availability, and import/export.
Connection options: Public IP, Private Service Access, and Private Service Connect.
Auth Proxy quickstart: How to install the proxy client and use the instance connection name format.
Admin API reference: Instance, database, and connection management endpoints.
Configure CDC for Datastream: Change data capture setup for replication pipelines.
Frequently asked questions
Which authentication method does Cloud SQL for SQL Server use?
Configure the database credentials as described in the integration steps, as Cloud SQL for SQL Server does not support IAM database authentication. Cloud SQL for MySQL and PostgreSQL support IAM database authentication.
How do you connect Datagrid to the instance?
Google recommends a direct connection over private IP. For public IP, the connecting source must sit in authorized networks. The Cloud SQL Auth Proxy and Cloud SQL Language Connectors are the other public IP options. The proxy creates outgoing connections on port 3307 and needs egress TCP on port 443 to sqladmin.googleapis.com. The client-to-proxy hop is unencrypted, so run the proxy on the workload machine.
How do you schedule Cloud SQL imports?
Use the configurable import schedule defined in Datagrid and coordinate the import window with downstream reporting deadlines.
How do you run a CDC pipeline alongside Datagrid imports?
Datastream CDC is a separate Google-native replication path from Datagrid's imports. Google documents Cloud SQL for SQL Server as a Datastream source. It replicates changes to BigQuery, Cloud Storage, AlloyDB, or Spanner. Turn on CDC with the Cloud SQL-specific procedure EXEC msdb.dbo.gcloudsql_cdc_enable_db 'DATABASE_NAME'. Google documents the rest of the setup in its configuration guide. CDC skips these column types: SQL_VARIANT, HIERARCHYID, GEOMETRY, and GEOGRAPHY.
How do you route Datagrid outputs after import?
Agents route their outputs onward through Datagrid's other integrations. These integrations reach CRMs, accounting software, and project management tools.
Similar integrations
Google Cloud SQL - PostgreSQL: Sibling Cloud SQL service for heterogeneous database fleets, migrations, and unified management across Postgres and SQL Server instances.
Amazon RDS: Alternative managed relational service supporting cross-cloud migrations, replication, and multi-cloud disaster recovery with SQL Server workloads.
BigQuery: Analytics destination for CDC and federated queries that offloads analytical workloads from operational SQL Server instances.
Snowflake: Cloud data warehouse target for transformed operational SQL Server data used in analytics, BI, and machine-learning pipelines.
MS SQL Server: On-prem or self-managed SQL Server integration for hybrid architectures, migrations, and reconciliations with Cloud SQL for SQL Server.
Azure SQL Database: Companion cloud SQL option for multi-cloud strategies, replication testing, and migration planning between Azure and Google Cloud.