Performance tuning
The most common reason people call. A slow application is almost always a database that is waiting on something specific, and the wait statistics say what. I find it, fix it and measure the difference, usually without new hardware.
- Wait statistics and execution plan analysis, live and from Query Store
- Indexing strategy: missing, duplicate and unused indexes, statistics, fill factor
- tempdb, memory, parallelism (MAXDOP, cost threshold) and I/O configuration
- Query rewrites, parameter sniffing, plan forcing, stored procedure tuning
- Blocking and deadlock analysis, isolation levels such as RCSI
- A written report with before-and-after figures for every change
Typical engagement 1 to 5 daysStarts with a wait-stats screenshot, or the health check
SQL Server Health Check
Three days, fixed price, up to ten instances. Configuration, backups, security, performance and availability checked against what I have seen go wrong in more than thirty years, written up in priority order with an effort estimate per finding. The best first step for an estate nobody has looked at for a while.
- € 2.950 fixed, excluding VAT, credited against a follow-up engagement
- Read-only access, no disruption to production
- Report plus a one-hour walkthrough with your team
Typical engagement 3 daysEverything about the health check →
High availability & disaster recovery
Being able to fail over is a claim until it has been tested. I design and build the right topology for your recovery objectives, then run the failover with you and write down what happened.
- Always On availability groups, including distributed AGs and read-only routing
- Failover cluster instances, log shipping, transactional replication
- Backup strategy with tested restores; RPO and RTO agreed and then measured
- Documented failover runbook and an annual test
- Azure: zone-redundant deployments, geo-replication, auto-failover groups
Typical engagement 3 to 15 daysStarts with your RPO and RTO, honestly stated
Upgrades & migrations, including Azure
Out-of-support versions, a data-centre exit or a move to Azure. Every migration gets compatibility testing on a copy, a sizing from real workload data, a rehearsed cut-over and a rollback that has actually been tried.
- SQL Server 2008 R2 through 2025, in-place or side-by-side
- Compatibility-level testing with Query Store and the Database Experimentation Assistant
- On-premises to Azure SQL Database, Azure SQL Managed Instance or SQL Server on Azure VMs
- Sizing and a cost estimate from your actual workload, before you commit
- Cut-over plan with minimal downtime; log shipping or distributed AG where minutes matter
- Licensing sanity check: Standard versus Enterprise, Azure Hybrid Benefit, cores
Typical engagement 5 to 30 daysStarts with an inventory of instances, versions and dependencies
Security & compliance
When the auditor, the DPO or a customer asks where the data is and who can read it, the answer should be a report, not a shrug. I harden SQL Server to the level your obligations require and document it so the question is easy next time.
- Transparent Data Encryption, Always Encrypted, backup encryption, TLS
- Dynamic Data Masking, row-level security, least-privilege roles
- SQL Server Audit and Extended Events for GDPR, ISO 27001, HIPAA and SOX evidence
- Vulnerability assessment, patch level, surface-area and sysadmin review
- Certificates, service accounts, Kerberos, linked-server hygiene
Typical engagement 2 to 10 daysStarts with the question the auditor asked
Architecture & database design
A design that is right for the first year and wrong for the fifth is expensive. I design data models, partitioning and integration for the growth you expect, and give a second opinion on designs you are about to buy.
- Data models and normalisation for OLTP; star schemas for reporting
- Partitioning, compression and archiving for tables that keep growing
- Replication and near real-time data mirroring between sites or to Azure
- Integration with .NET applications, APIs, Power BI and Microsoft Fabric
- A second opinion on a vendor's database design before you sign
Typical engagement 2 to 20 daysStarts with the volumes and the growth you expect
Remote DBA & 24x7x365 monitoring
For organisations without a DBA, or with one who is busy. The routine done properly by a senior engineer: backups, patching, capacity, jobs and index maintenance, with a monthly report you can read in five minutes. Around-the-clock monitoring with on-call DBAs is delivered through DatabaseOnline.
- Backups, restore tests, patching, capacity and job management
- Monthly health and capacity report, quarterly review with your IT lead
- Retainer from a few hours per month to a dedicated senior DBA
- 24x7x365 monitoring in Silver, Gold and Platinum packages via DatabaseOnline
Typical engagement ongoing, monthlyStarts with the health check