Reviewed by Aditya Kumar · Last reviewed 2026-03-24
**Why Self-Hosted IR (SHIR)**: ADF runs in Azure; on-prem data sources (SQL Server, file shares) are not directly reachable. SHIR runs on-prem, registers with ADF, executes data movement locally. **Architecture**: Install SHIR on Windows/Linux VM; requires outbound HTTPS (port...
This hard-level Cloud/Tools question appears frequently in data engineering interviews at companies like Presidio. While less common, it tests deeper understanding that distinguishes strong candidates. Mastering the underlying concepts (sql, window) will help you answer variations of this question confidently.
This is a senior-level question that tests architectural thinking. Lead with the high-level design, then drill into specifics. Discuss trade-offs explicitly - there is rarely one correct answer. Show awareness of scale, fault tolerance, and operational complexity.
Why Self-Hosted IR (SHIR): ADF runs in Azure; on-prem data sources (SQL Server, file shares) are not directly reachable. SHIR runs on-prem, registers with ADF, executes data movement locally. Architecture: Install SHIR on Windows/Linux VM; requires outbound HTTPS (port 443) to Azure. Register with auth key from ADF. SHIR pulls jobs from ADF control plane; executes copy/transform on-prem. Use ExpressRoute or VPN for secure, predictable connectivity. Scalability: Add nodes to an IR for high availability and parallelism; multiple activities can run concurrently. Cost: You pay for the VM; ADF does not charge for SHIR execution time (only for data movement in some cases). Right-size VM—4 vCPU often sufficient for moderate throughput. Best practice: Dedicated VM; avoid sharing with other workloads. Use managed identity where possible for Azure-side auth. Monitor IR heartbeats; alert on node failures.
Pro-Move: 'We run SHIR on a 4-node cluster behind a load balancer for HA—when one node failed during a 2TB copy, the job resumed on another node.' Red Flag: Running SHIR on a shared app server—resource contention and security risk.
Some links below are affiliate links. If you buy through them we may earn a small commission at no extra cost to you — it helps keep DataEngPrep free.
According to DataEngPrep.tech, this is one of the most frequently asked Cloud/Tools interview questions, reported at 1 company. DataEngPrep.tech maintains an editor-reviewed database of 1,863 data engineering interview questions across 7 categories.