Skip to content
Home › SQL Comparisons › Datadog vs SolarWinds DPA
Comparison · Database Monitoring

Datadog vs SolarWinds DPA

Datadog Database Monitoring is a SaaS add-on to the Datadog observability platform, billed per database host and linked to APM traces; SolarWinds Database Performance Analyzer (DPA) is a self-hosted, agentless, wait-time-focused database performance tool licensed per monitored instance. Choose Datadog if your teams already live in Datadog, DPA if you want a dedicated DBA tool that keeps data on your own servers.

Last verified October 2026. Versions checked: SolarWinds DPA 2026.2. Licensing and features change; check the official sources for the latest details.

Quick verdict

Short answer

Pick Datadog Database Monitoring if your organisation already uses Datadog for infrastructure and APM: query metrics, samples and explain plans sit next to the application traces that caused them, and nothing has to be hosted. Pick SolarWinds DPA if you are a DBA team that wants a dedicated, self-hosted tool built around wait-time analysis, with a long history window (summaries kept for five years by default) and broad engine coverage that includes Oracle, Db2, SAP ASE and SAP HANA. DPA keeps the monitoring data on your own repository database; Datadog sends it to Datadog's cloud.

How we know: This comparison is research-based: features, supported versions, licensing and prices were checked against Datadog's and SolarWinds' official documentation and pricing pages on 7 October 2026. We have not run either product, so nothing here is a measured result; overhead figures are the vendors' own statements.

Datadog Database Monitoring (DBM) is one product within Datadog's SaaS observability platform. The Datadog Agent connects to each database with a read-only user and sends query metrics, query samples, explain plans and database metrics to Datadog. Datadog documents support for self-hosted and managed cloud versions of Postgres, MySQL, MariaDB, Oracle, SQL Server, MongoDB, Amazon DocumentDB and ClickHouse. DBM can be correlated with Datadog APM so that a slow query links to the service and trace that issued it.

SolarWinds Database Performance Analyzer (DPA) is a self-hosted database performance tool. The DPA server runs on Windows Server or Linux, collects data without installing agents on the monitored servers, and stores it in a separate repository database. Its core idea is wait-time (response time) analysis: showing what each query spent its time waiting on. The current release is DPA 2026.2 (30 June 2026). SolarWinds has been a privately held company since Turn/River Capital completed its acquisition on 16 April 2025; the products continue under the SolarWinds brand. SolarWinds also sells SQL Sentry and a separate SaaS platform, SolarWinds Observability; see SQL Sentry vs SolarWinds DPA.

Side by side

AspectDatadogSolarWinds DPA
Deployment SaaS; the Datadog Agent runs on the database host or, for managed databases, on a separate host Self-hosted DPA server (Windows Server 2022/2025 or Linux) plus a repository database; agentless collection
Where data lives Datadog's cloud Your own repository (SQL Server, Azure SQL Database, MySQL or Oracle)
Engines Postgres, MySQL, MariaDB, Oracle, SQL Server, MongoDB, Amazon DocumentDB, ClickHouse SQL Server, Azure SQL, Oracle, PostgreSQL, MySQL, MariaDB, Db2, SAP ASE, SAP HANA (version lists per engine)
SQL Server versions SQL Server 2012 to 2025 (2025 needs Agent 7.79 or later) Self-managed SQL Server 2022 and 2025; RDS for SQL Server 2019 and 2022; Azure SQL Database and Managed Instance
Core analysis Normalized query metrics, query samples, explain plans, wait events, active connections, blocking Wait-time analysis per query, anomaly detection, tuning advisors, AI Query Assist
APM correlation Built in: trace IDs injected into SQL comments link queries to APM traces Not an APM product; DPA data can be shown in the SolarWinds Platform web console
History Query metrics kept 15 months, query samples 15 days Per-second detail for about 30 days, hourly summaries for 90 days, daily summaries for 5 years
Pricing basis Per database host per month, public list price Per monitored instance, subscription; no public price
Main trade-off Strong if you already use Datadog, but usage-based cost and data leaves your network Dedicated DBA depth and long retention, but you run and patch the server yourself

Key differences

Platform add-on versus dedicated database tool

Datadog DBM is designed to sit inside a wider observability estate. Datadog documents that connecting DBM with APM injects trace identifiers into the SQL sent to the database (as SQL comments), so you can see which services drive database load, filter database hosts by calling service and open the sampled APM trace behind a query sample. Datadog lists this for Postgres, MySQL, SQL Server, Oracle and MongoDB, with language support varying by engine (for SQL Server: Go, Java and .NET). If your developers already investigate incidents in Datadog, this is the main reason to choose it.

DPA is a tool for people who look after databases. It does not trace applications. Its screens are organised around database instances and the time queries spend waiting, and it integrates with the wider SolarWinds Platform rather than with application tracing. For the conceptual difference between the two approaches, see Database Monitoring vs APM.

Hosting, data location and the Agent

Datadog is SaaS only. For self-hosted databases the Agent usually runs on the database host; for managed services such as Azure SQL, Datadog states that the Agent must run on a separate host that can reach the database, because Azure does not give host access. The Agent connects with a read-only datadog login (for SQL Server: VIEW SERVER STATE, VIEW ANY DEFINITION and CONNECT ANY DATABASE). Datadog obfuscates query bind parameters before they are sent, but notes that schema elements and SQL comments are not obfuscated by default.

DPA runs entirely on infrastructure you manage. It is agentless: the DPA server polls each monitored instance, and SolarWinds advises that the repository should not be hosted on an instance you plan to monitor. This suits organisations that cannot send query text to a third-party cloud, but it means you provision, patch and back up the DPA server and repository. DPA 2026.2 dropped support for Windows Server 2019, 2016 and 2012 R2 as the DPA server OS.

What each collects, and the built-in baseline

Both tools build on what the engine already exposes. For SQL Server, that means dynamic management views such as sys.dm_exec_query_stats and sys.dm_os_wait_stats. This is the free baseline you can query yourself:

-- T-SQL (SQL Server): top waits since the last restart
SELECT TOP (10) wait_type, waiting_tasks_count, wait_time_ms
FROM sys.dm_os_wait_stats
ORDER BY wait_time_ms DESC;

Datadog collects metrics for the top 200 normalized queries per database host (parameters stripped so identical queries group together), plus query samples, explain plans, active connections, blocking queries and database metrics. Optional SQL Server collection covers SQL Server Agent job metrics, deadlocks, index usage, schemas and, through Extended Events, query completions and errors. DPA records wait time per query continuously and adds anomaly detection, table tuning advisors and AI Query Assist, which DPA 2026.2 extends to PostgreSQL and MySQL in addition to SQL Server and Oracle. To read wait statistics with more context, see our lessons on performance tuning and indexes.

Engine and version coverage

Coverage differs by engine, so check the exact version you run. Datadog lists SQL Server 2012 to 2025, Postgres 9.6 to 18 (self-hosted, RDS, Aurora, Cloud SQL, AlloyDB, Azure, Heroku and Supabase), MySQL 5.6 to 9 and MariaDB 10.5 to 13. It also covers MongoDB, Amazon DocumentDB and ClickHouse, which DPA does not.

DPA's list is narrower in versions but wider in enterprise engines: Oracle 19c to 26ai, Db2 11.5 and 12.1, SAP ASE 16 and, from 2026.2, SAP HANA. For SQL Server, DPA lists only 2022 and 2025 as self-managed versions, plus RDS for SQL Server 2019 and 2022, Cloud SQL for SQL Server 2022 and Azure SQL Database and Managed Instance. Teams still on SQL Server 2019 or older on premises should note that gap. PostgreSQL 14 to 18 and MySQL 8.4 to 9.2 are listed, along with Aurora, Azure Database for PostgreSQL Flexible Server and Cloud SQL.

History and retention

Datadog states that DBM query metrics are kept for 15 months and query samples for 15 days. DPA keeps per-second detail for about 30 days, hourly summaries for 90 days and daily summaries for five years, in your repository. For year-on-year capacity questions, DPA's default history is longer; for drilling into a recent incident alongside traces and logs, Datadog's samples and APM links are the stronger fit.

Pricing and licensing

Datadog. Listed on the vendor's pricing page in October 2026 (USD): Database Monitoring is $70 per database host per month billed annually, or $84 billed month-to-month or on demand. Datadog's billing documentation says each database host includes 200 normalized queries, with more billed by the number tracked. DBM is a separate product from Infrastructure Monitoring (from $15 per host per month billed annually for Pro) and APM, so a Datadog bill typically combines several products. Datadog offers a 14-day free trial with no credit card.

SolarWinds DPA. No public price. DPA is licensed per monitored database instance on subscription; since DPA 2024.4 the licence types are Database Self-Hosted (DBSH) for any supported database type and Database Self-Hosted DBaaS (DBSHDS) for targets such as Azure SQL Database. Perpetual licences were sold for earlier versions and are legacy. A 14-day trial is available, and DPA is also sold through AWS Marketplace with hourly charges.

Pricing checked on the vendors' official pages on 7 October 2026. Prices change; confirm before buying.

Where each one leads

Datadog strengths

  • Query samples and explain plans linked to APM traces and services
  • Nothing to host: SaaS with a public per-host list price
  • Wide engine list, including MongoDB, Amazon DocumentDB and ClickHouse
  • SQL Server 2012 to 2025 supported, plus managed Postgres and MySQL services
  • Query metrics retained for 15 months

SolarWinds DPA strengths

  • Self-hosted and agentless; monitoring data stays in your own repository
  • Built around wait-time analysis per query, with tuning advisors
  • Covers Oracle, Db2, SAP ASE and SAP HANA alongside SQL Server, PostgreSQL and MySQL
  • Daily summaries kept for five years by default

Limitations

Datadog limitations

  • Usage-based bill across several products (DBM, infrastructure, APM) that needs managing
  • SaaS only; query text and metadata are sent to Datadog
  • Metrics cover the top 200 normalized queries per host unless you pay for more
  • Query samples are kept for 15 days only

SolarWinds DPA limitations

  • No public price; you need a quote
  • You run, patch and back up the DPA server and repository
  • Self-managed SQL Server support is listed for 2022 and 2025 only
  • No application tracing; correlation with code needs another tool

When to choose each

Choose Datadog if

  • Your organisation already uses Datadog for infrastructure, logs or APM
  • Developers and SREs, not only DBAs, investigate database slowdowns
  • Most databases are managed cloud services and you want no monitoring servers
  • You run MongoDB, DocumentDB or ClickHouse alongside relational engines

Choose SolarWinds DPA if

  • Monitoring data must stay on your own infrastructure
  • Your DBA team wants wait-time analysis and long history in one dedicated tool
  • You run Oracle, Db2, SAP ASE or SAP HANA
  • You prefer per-instance subscription licensing over usage-based billing

When neither is right

Final recommendation

Bottom line

These products answer different questions. Datadog DBM is the better fit when database monitoring is one part of a shared observability practice: it links queries to traces, needs no servers of your own and has a published per-host price, at the cost of sending data to Datadog and a bill that grows with hosts and products. SolarWinds DPA fits a DBA team that wants a dedicated, self-hosted, wait-time-centred tool with long history and coverage of enterprise engines such as Oracle, Db2 and SAP HANA, accepting that you host it, request a quote and check its narrower SQL Server version list. See also the best database monitoring tools and the best SolarWinds DPA alternatives.

Frequently asked questions

Is SolarWinds DPA agentless?

Yes. SolarWinds describes DPA as agentless: the DPA server collects data remotely from each monitored instance and stores it in a separate repository database.

Does Datadog Database Monitoring need an agent?

Yes. The Datadog Agent collects DBM data. For self-hosted databases it usually runs on the database host; for managed services such as Azure SQL Database it runs on a separate host that can connect to the database.

Which SQL Server versions does each support?

Datadog lists SQL Server 2012 to 2025 (2025 needs Agent 7.79 or later). DPA 2026.2 lists self-managed SQL Server 2022 and 2025, Amazon RDS for SQL Server 2019 and 2022, Cloud SQL for SQL Server 2022, and Azure SQL Database and Managed Instance.

Who owns SolarWinds now?

Turn/River Capital completed its acquisition of SolarWinds on 16 April 2025, making it a private company. DPA and SQL Sentry continue to be sold under the SolarWinds brand.

How much overhead do they add?

We have not measured either. Datadog's setup documentation states that DBM typically represents less than 1 percent of query execution time and less than 1 percent of CPU for most workloads; SolarWinds' DPA page also claims less than 1 percent. Both are vendor statements.

Sources

Checked October 2026.

How we research comparisons: our editorial method.

More comparisons

Browse all SQL comparisons or the tools directory.