Azure Synapse Analytics is Microsoft's cloud data warehouse and analytics service, joining T-SQL pools, Spark and pipelines on Azure with pay-as-you-go billing.
Azure Synapse Analytics puts enterprise data warehousing, big data processing and data integration in one workspace. Analysts and engineers reach warehouse tables, data lake files and operational databases from the same place. Microsoft builds and runs it as a service of the Azure cloud. Its warehouse engine keeps compute apart from storage, so compute can be resized or paused while the data stays put. The dedicated SQL pool is a distributed T-SQL data warehouse. Its materialized views pick up changes from base tables automatically, result set caching answers repeated queries, and workload groups give each kind of query its own resources. Every workspace also has a serverless SQL pool that reads Parquet, Delta Lake and CSV files in the lake through external tables, with no cluster to set up. Apache Spark pools run data engineering and machine learning jobs, and their Delta Lake tables can be queried as earlier snapshots. Data moves through pipelines and mapping data flows, which orchestrate, copy and transform it on Azure-hosted or self-hosted integration runtimes. Azure Synapse Link replicates operational SQL data for analytics in near real time, and built-in streaming lands data from cloud sources in SQL tables. Dynamic data masking and Row-Level Security control what each user sees inside the warehouse. The service suits data teams that already write T-SQL and keep their data on Azure, and that want SQL and Spark side by side. It is added to an existing Azure account like any other Azure service. For teams that decide to move on, Microsoft publishes a Synapse to Fabric migration guide and the Fabric Migration Assistant for Data Warehouse. Typical work is running a company's central warehouse, exploring raw lake files before they are modelled, and preparing data in Spark for machine learning. Reports are usually built in Power BI, and Synapse Link feeds those reports operational SQL data in near real time. Azure Cost Management budgets warn the team when Synapse spending passes a set amount. Microsoft Defender for SQL adds threat detection for organisations that need it. Synapse is at home in organisations that keep their lake in Azure Data Lake Storage and build their dashboards in Power BI.
Pricing
View plans-
Dedicated SQL pool
$876 per month
Pay-as-you-go monthly rate at DW100c, USD, East US 2; $551.92 a month on a 1-year reservation, $306.61 on a 3-year reservation
- Compute billed by performance level from DW100c to DW30000c
- DW1000c is $8,760 a month pay as you go, $5,519.17 on a 1-year reservation, $3,066.15 on 3 years
- DW30000c, the largest level, is $262,800 a month pay as you go
- Storage billed separately at $23 per TB per month, including 7 days of incremental snapshots
- Geo-redundant disaster recovery storage from $0.057 per GB per month
- Pausing stops compute charges; scaling is a manual change of level
-
Serverless SQL pool
Custom pricing
- $5 per TB of data processed
- 10 MB minimum per query, each query rounded up to the nearest 1 MB
- Data lake storage priced separately
-
Apache Spark pool
Custom pricing
- $0.143 per vCore-hour, memory optimized
- GPU accelerated pools in public preview at $0.15 per vCore-hour
- Usage rounded up to the nearest minute
-
Data integration
Custom pricing
- $1 per 1,000 orchestration activity runs, Azure hosted; $1.50 per 1,000 self-hosted
- Data movement $0.25 per DIU-hour Azure hosted, $0.10 an hour self-hosted
- Pipeline activity integration runtime $1 an hour Azure hosted, up to 50 concurrent activities
- Mapping data flows $0.257 per vCore-hour Basic, $0.325 per vCore-hour Standard
- First 1 million pipeline operations a month at no charge, then $0.25 per 50,000
- Copy activity egress out of an Azure datacenter adds network bandwidth charges
Features
- Included: Independent scaling of storage and compute
- Included: Serverless query execution
- Not included: Multi-cluster concurrency scaling
- Included: Time travel to prior table states
- Not included: Zero-copy cloning of databases
- Included: Auto-refreshing materialized views
- Not included: Native semi-structured columns
- Included: External tables on data lake files
- Included: Open table format interoperability
- Included: Streaming row ingestion into tables
- Not included: Live data sharing across accounts
- Included: Column masking policies
- Included: Row access policies
- Included: Automatic query result reuse
- Included: Workload management queues
- Included: Compute spend limits and alerts
- Not included: Cross-region warehouse replication
- Not included: In-warehouse Python execution
Additional Features
- Dedicated SQL pools for distributed T-SQL warehousing
- Serverless SQL pool endpoint in every workspace
- Apache Spark pools for data engineering and machine learning
- GPU accelerated Spark pools in public preview
- Pipelines for orchestration and copy activities
- Mapping data flows for visual transformation
- Azure-hosted and self-hosted integration runtimes
- Azure Synapse Link for SQL near real-time replication
Best for
- Data teams whose analytics already run on Azure
- SQL developers who write T-SQL and need a distributed warehouse
- Data engineers who want SQL and Spark in one workspace
- Teams querying Azure Data Lake Storage files without loading them first
- Organisations reporting in Power BI on operational SQL data
- Warehouses that sit idle for hours and can be paused
Use cases
- Running an enterprise data warehouse in T-SQL
- Querying Parquet, Delta Lake and CSV files in the data lake with SQL
- Orchestrating pipelines that copy and transform data
- Preparing data for machine learning in Spark pools
- Replicating operational SQL data for near real-time reporting
- Masking sensitive columns from non-privileged users
- Restricting table rows per user with Row-Level Security
- Giving loads and dashboards separate resources with workload groups
- Landing streaming data from cloud sources in SQL tables
- Serving repeated dashboard queries from a result cache
- Alerting on analytics spend with Cost Management budgets
Screenshots & Videos
Explore Azure Synapse Analytics in action
User Reviews
Compare with Similar Software
Click to compare side by side with Azure Synapse Analytics
