실행 계획 노드 해석
순차 스캔, 인덱스 스캔, 조인 유형, 정렬과 같은 일반적인 실행 계획 노드를 해석하는 방법을 학습합니다.
실행 계획 노드 해석은(는) CoddyKit의 무료 PostgreSQL Performance & Query Optimization 강의입니다. 이것은 4개 중 2번째 강의입니다. 아래에서 전체 강의를 무료로 읽을 수 있으며, 내장 코드 에디터와 24/7 AI 튜터와 함께 브라우저에서 직접 실습할 수 있습니다. 이 강의는 PostgreSQL Performance & Query Optimization 학습 경로의 일부이며, 진행 상황이 웹과 CoddyKit 앱에 동기화됩니다. PostgreSQL Performance & Query Optimization 강의에는 총 4개의 강의가 포함되어 있습니다.
이 강의의 일부는 아직 번역되지 않았으며 영어로 표시됩니다.
Decoding Query Plans
Welcome back! In the previous lesson, you learned how to use EXPLAIN to view a query's execution plan. Now, let's dive into interpreting the different 'nodes' within these plans.
Each node represents a specific operation PostgreSQL performs. Understanding them is key to identifying performance bottlenecks.
What are Plan Nodes?
Think of a query plan as a tree, where each branch and leaf is a 'node'. These nodes tell you:
- What operation is being done (e.g., scanning, sorting, joining).
- How it's being done (e.g., sequentially, using an index).
- Cost estimates: How much time and resources PostgreSQL *expects* the operation to take.
We'll look at the most common and important node types.
Sequential Scan: The Full Read
A Sequential Scan (often called a 'Seq Scan') means PostgreSQL reads every single row in a table from start to finish to find the data it needs.
- When it happens: For small tables, or when querying a large portion of a table with no suitable index.
- Performance impact: Can be slow for large tables, especially if only a few rows are needed.
It's like looking through every page of a book to find one sentence.
Seq Scan Example
Let's see a sequential scan in action. We'll create a simple table and then query it without an index.
CREATE TABLE products (
product_id SERIAL PRIMARY KEY,
name VARCHAR(100),
price DECIMAL(10, 2)
);
INSERT INTO products (name, price) VALUES
('Laptop', 1200.00),
('Mouse', 25.00),
('Keyboard', 75.00),
('Monitor', 300.00);
EXPLAIN SELECT * FROM products WHERE price > 100;Index Scan: Targeted Search
An Index Scan is much more efficient. PostgreSQL uses an index to quickly locate the specific rows it needs, much like using an index in a book.
- When it happens: When a query uses a
WHEREclause on an indexed column, and the index is selective enough. - Performance impact: Generally much faster than a sequential scan for selective queries on large tables.
It allows PostgreSQL to jump directly to the relevant data pages.
Index Scan Example
Now, let's add an index to our products table and observe the change in the query plan.
CREATE INDEX idx_products_price ON products (price);
EXPLAIN SELECT * FROM products WHERE price > 100;Sort Node: Ordering Data
The Sort node appears when PostgreSQL needs to order data, typically for an ORDER BY or GROUP BY clause, and there isn't an index that can provide the data in the required order.
- When it happens: Explicit
ORDER BY, or implicitly for some operations likeGROUP BYor unique constraints. - Performance impact: Sorting can be CPU and I/O intensive, especially for large datasets.
If the sort happens 'on disk' (meaning it can't fit in memory), it becomes even slower.
Sort Node Example
Here's an example where PostgreSQL has to sort the results because no index exists for the ordering column.
EXPLAIN SELECT name, price FROM products ORDER BY name DESC;Join Nodes: Combining Tables
When you join two or more tables, PostgreSQL uses specific Join Nodes to combine the data. There are three primary types:
- Nested Loop Join: Often good for small inner tables or when an index is available.
- Hash Join: Efficient for larger tables where no useful index is present on the join key.
- Merge Join: Requires both inputs to be sorted on the join key, then merges them.
The choice depends on table sizes, available indexes, and data distribution.
Nested Loop Join Example
Let's create another table and then join it with products to see a Nested Loop Join. This often happens when one side of the join is small.
CREATE TABLE orders (
order_id SERIAL PRIMARY KEY,
product_id INT,
quantity INT
);
INSERT INTO orders (product_id, quantity) VALUES
(1, 1),
(2, 2),
(1, 3);
EXPLAIN SELECT p.name, o.quantity
FROM products p JOIN orders o ON p.product_id = o.product_id
WHERE o.order_id = 2;Identify the Scan Type
Consider the following query and its execution plan snippet. What kind of scan is most likely being performed on the customers table?
EXPLAIN SELECT * FROM customers WHERE age > 30;
Partial Plan Output (assume no index on age):
-> Seq Scan on customers (cost=0.00..10.50 rows=3 width=...)Recap: Decoding Plan Nodes
You've taken a big step in understanding PostgreSQL performance by learning to interpret key plan nodes!
- Sequential Scan: Full table read, can be slow for large tables.
- Index Scan: Uses an index for targeted row access, faster for selective queries.
- Sort: Occurs when data needs ordering and no suitable index exists.
- Join Nodes: (Nested Loop, Hash, Merge) combine data from multiple tables, chosen based on data size and indexes.
In the next lesson, we'll put this knowledge to use to identify actual performance bottlenecks!
자주 묻는 질문
“실행 계획 노드 해석” 강의는 무료인가요?
네 — “실행 계획 노드 해석” 전체 내용을 이 웹사이트에서 무료로 읽을 수 있습니다. 인터랙티브하게 실습하려면(내장 코드 에디터와 24/7 AI 튜터), CoddyKit PRO로 업그레이드하면 PostgreSQL Performance & Query Optimization 강의 전체를 잠금 해제할 수 있습니다. PostgreSQL Performance & Query Optimization 강의에는 총 4개의 강의가 포함되어 있습니다.
“실행 계획 노드 해석”에서 뭘 배우나요?
순차 스캔, 인덱스 스캔, 조인 유형, 정렬과 같은 일반적인 실행 계획 노드를 해석하는 방법을 학습합니다. 브라우저에서 직접 실행하는 실습 코드로 PostgreSQL Performance & Query Optimization을(를) 배우며, 24/7 AI 튜터가 강의를 진행하면서 질문에 답변해줍니다.
PostgreSQL Performance & Query Optimization을(를) 시작하는 데 경험이 필요한가요?
사전 경험은 필요하지 않습니다. CoddyKit의 PostgreSQL Performance & Query Optimization은(는) 초급자부터 고급 학습자까지를 위해 구성되어 있으므로, 여기서 시작하거나 처음부터 시작할 수 있으며 자신의 속도대로 진행할 수 있습니다. 이것은 4개 중 2번째 강의입니다.
“실행 계획 노드 해석” 강의는 얼마나 걸리나요?
대부분의 CoddyKit 강의는 약 5~10분이 소요됩니다. 각 강의는 간결하고 인터랙티브하여 꾸준한 진행이 가능하며, 웹과 앱에서 중단한 부분부터 바로 시작할 수 있습니다.
이 PostgreSQL Performance & Query Optimization 강의에서 코드를 작성하고 실행할 수 있나요?
네. 모든 PostgreSQL Performance & Query Optimization 강의에는 내장 코드 에디터가 포함되어 있으므로, 브라우저에서 바로 실제 코드를 작성하고 실행한 후 즉시 AI 피드백을 받을 수 있습니다 — 로컬 설정이 필요 없습니다.