Database Patterns

Database Patterns Skill

ORM Selection

Schema Design

Naming Conventions

-- Tables: plural, snake_case
CREATE TABLE user_profiles (...);

-- Columns: snake_case
user_id, created_at, is_active

-- Indexes: idx_{table}_{columns}
CREATE INDEX idx_users_email ON users(email);

-- Foreign keys: fk_{table}_{ref_table}
CONSTRAINT fk_orders_users FOREIGN KEY (user_id) REFERENCES users(id)

Common Patterns

Soft Delete

ALTER TABLE users ADD COLUMN deleted_at TIMESTAMP NULL;

-- Query active records
SELECT * FROM users WHERE deleted_at IS NULL;

Audit Columns

created_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP,
updated_at TIMESTAMP DEFAULT CURRENT_TIMESTAMP ON UPDATE CURRENT_TIMESTAMP,
created_by UUID REFERENCES users(id),
updated_by UUID REFERENCES users(id)

UUID vs Serial

Index Strategies

When to Index

• Foreign keys (always)
• Columns in WHERE clauses
• Columns in ORDER BY
• Columns in JOIN conditions

Index Types

Composite Index Order

-- Good: matches query pattern
CREATE INDEX idx_orders_user_date ON orders(user_id, created_at);
SELECT * FROM orders WHERE user_id = 1 AND created_at > '2024-01-01';

-- Index used for:
-- WHERE user_id = 1
-- WHERE user_id = 1 AND created_at > ...

-- Index NOT used for:
-- WHERE created_at > '2024-01-01' (missing leading column)

Query Optimization

Explain Analyze

EXPLAIN ANALYZE
SELECT * FROM users WHERE email = 'test@example.com';

Common Issues

N+1 Prevention

# Bad: N+1
for user in users:
    print(user.orders)  # Query per user

# Good: Eager loading
users = User.query.options(joinedload(User.orders)).all()

Migration Best Practices

Safe Migrations

-- Add column (safe)
ALTER TABLE users ADD COLUMN phone VARCHAR(20);

-- Add NOT NULL column (safe pattern)
ALTER TABLE users ADD COLUMN phone VARCHAR(20);
UPDATE users SET phone = '' WHERE phone IS NULL;
ALTER TABLE users ALTER COLUMN phone SET NOT NULL;

-- Rename column (use application-level)
-- 1. Add new column
-- 2. Copy data
-- 3. Update application
-- 4. Remove old column

Migration Checklist

• Tested on production-like data
• Rollback script ready
• No long locks on large tables
• Indexes created concurrently
• Application handles both states

Connection Pooling

PgBouncer Settings

[pgbouncer]
pool_mode = transaction
max_client_conn = 1000
default_pool_size = 20

Application Settings

Vector Database (Qdrant) Patterns

Client Setup

from qdrant_client import QdrantClient
from qdrant_client.models import Distance, VectorParams, PointStruct

# Sync client
client = QdrantClient(host="localhost", port=6333)

# Async client
from qdrant_client import AsyncQdrantClient
async_client = AsyncQdrantClient(host="localhost", port=6333)

Collection Management

# Create collection (single vector)
client.create_collection(
    collection_name="documents",
    vectors_config=VectorParams(size=384, distance=Distance.COSINE)
)

# Create collection (multi-vector)
from qdrant_client.models import VectorParams

client.create_collection(
    collection_name="multimodal",
    vectors_config={
        "text": VectorParams(size=384, distance=Distance.COSINE),
        "image": VectorParams(size=512, distance=Distance.EUCLID),
    }
)

Upserting Vectors

# Single upsert
client.upsert(
    collection_name="documents",
    points=[
        PointStruct(
            id=1,
            vector=[0.1, 0.2, 0.3, ...],  # 384-dim vector
            payload={"title": "Doc 1", "category": "tech"}
        )
    ]
)

# Batch upsert
points = [
    PointStruct(id=i, vector=vectors[i], payload=payloads[i])
    for i in range(len(vectors))
]
client.upsert(collection_name="documents", points=points, batch_size=100)

Searching Vectors

from qdrant_client.models import Filter, FieldCondition, MatchValue

# Basic search
results = client.search(
    collection_name="documents",
    query_vector=[0.1, 0.2, ...],
    limit=10
)

# Search with filter
results = client.search(
    collection_name="documents",
    query_vector=[0.1, 0.2, ...],
    query_filter=Filter(
        must=[
            FieldCondition(key="category", match=MatchValue(value="tech"))
        ]
    ),
    limit=10,
    with_payload=True,
    score_threshold=0.7
)

# Search with range filter
from qdrant_client.models import Range

results = client.search(
    collection_name="documents",
    query_vector=query_vector,
    query_filter=Filter(
        must=[
            FieldCondition(key="price", range=Range(gte=10, lte=100))
        ]
    ),
    limit=10
)

Payload Indexing

# Create payload index for faster filtering
client.create_payload_index(
    collection_name="documents",
    field_name="category",
    field_schema="keyword"  # or "integer", "float", "bool"
)

Best Practices

Distance Metrics

Common Rationalizations

Rules

• MUST profile queries with EXPLAIN (ANALYZE, BUFFERS) before adding an index — indexes chosen by intuition miss the real hot path half the time
• MUST design the schema around the dominant access pattern, not the logical entity graph — storage follows queries, not the other way round
• NEVER write to production with raw SQL when a migration file fits — ad-hoc changes break rollback and audit
• NEVER add a SELECT * in a loop — N+1 is the most common performance regression in code review
• CRITICAL: every foreign key has an index on the referencing column. Postgres does not create one automatically, and ON DELETE CASCADE without the index causes full-table scans on delete.
• MANDATORY: numeric IDs use bigint (or bigserial) in new tables unless there is a stated reason to cap at 2^31. Integer overflow on a growing table is a late, painful surprise.

Gotchas

• EXPLAIN without ANALYZE shows the planner's estimate, not the actual execution. A query plan that "looks good" with EXPLAIN can still be slow in practice — always use ANALYZE for real diagnosis.
• ORM-generated queries often look efficient in one row but emit N+1 at scale. prisma, sequelize, activerecord all have "eager loading" switches that must be explicit — the default is lazy and bites under load.
• Postgres transactions hold row locks until commit or rollback. A long-running transaction that reads rows another writer needs blocks progress silently. Investigate pg_stat_activity for state=idle in transaction when writes stall.
• Index-only scans require both the query columns AND the filter to be in the index (or in the visibility map for heap tuples). Adding a single column to WHERE can demote an index-only scan to an index scan with a 10× slowdown.
• MySQL implicit collation on JOIN across tables with different utf8mb4 collations forces a row-by-row collation conversion — a 100× slowdown that shows as a full scan in the plan. Align collations during schema design.

When NOT to Load

• For schema evolution (zero-downtime, expand-contract, backfill) — use /migration-patterns
• For running migrations as a task — use /migrate
• For query-plan profiling and the four golden signals — use /performance-profiling
• For vector/embedding-specific schema — this skill covers the mechanics; use /rag-patterns for retrieval design
• For observability of DB metrics (slow query log, connection pool saturation) — use /observability-patterns