Why should you care?

Enterprise ShopNest at scale cannot share one monolithic DbContext — teams deploy Catalog API and Order API independently with separate migration pipelines.

See it live — copy this example

Paste into a .NET project with EF Core packages, then run with LocalDB/SQL Server (dotnet ef / dotnet run).

// Catalog service
public class CatalogDbContext : DbContext
{
    public DbSet<Product> Products => Set<Product>();
    public DbSet<Category> Categories => Set<Category>();
}

// Order service
public class OrderDbContext : DbContext
{
    public DbSet<Order> Orders => Set<Order>();
    public DbSet<OrderItem> OrderItems => Set<OrderItem>();
    // ProductId is external reference — no FK to Catalog DB
}

// Order service validates product via Catalog HTTP API before insert
var product = await _catalogApi.GetProductAsync(line.ProductId);
order.Items.Add(new OrderItem { ProductId = product.Id, UnitPrice = product.Price, Quantity = line.Qty });

What happened?

• CatalogDbContext owns Product schema.
• OrderDbContext stores ProductId snapshot without cross-database FK.
• Price copied at order time for autonomy between services.

Practice next

1. Create Catalog and Order solutions with separate Infrastructure projects.
2. Run independent dotnet ef migrations add per service database.
3. Replace cross-context navigation with HTTP/gRPC calls and event bus.
4. Publish ProductUpdated integration event; Order service updates read model cache.
5. Add InventoryDbContext with StockItem and consume OrderPlaced event to reserve stock.

Remember

Capstone: multiple EF contexts per service. Reference data via API not cross-DB FK. Snapshot price and SKU on OrderItem at purchase. Independent migrations per bounded context.

ShopNest enterprise microservices

Catalog, Order, and Inventory teams ship separate EF migrations weekly; checkout saga coordinates via messages.

Outcome: Flipkart-scale ShopNest operates as deployable services sharing no single database while staying consistent at business level.