Storage vs Memory
Data location.
Storage vs Memory is a free Web3 & DApp Development Fundamentals lesson on CoddyKit — lesson 3 of 4. You can read the complete lesson below for free — then practise it hands-on in the browser with a built-in code editor and a 24/7 AI tutor. It is part of the Web3 & DApp Development Fundamentals learning path, one of 4 lessons in the course, and your progress syncs across the web and the CoddyKit app.
Why Data Location Matters
In Solidity, reference types must specify a data location. The three locations are storage, memory, and calldata.
Choosing the right one affects persistence, mutability, and most importantly gas cost.
Storage
Storage is the contract's permanent on-chain state. State variables live here, and writing to storage is the most expensive operation.
Data in storage persists between transactions.
<code>contract Bank {
uint256 public total; // lives in storage forever
function add(uint256 v) public {
total += v; // writes to storage (expensive)
}
}</code>Memory
Memory is a temporary scratchpad that exists only during a function call. It is much cheaper than storage and is cleared when the function returns.
Use memory for intermediate calculations and data you do not need to persist.
<code>function build() public pure returns (uint[] memory) {
uint[] memory temp = new uint[](3);
temp[0] = 1;
temp[1] = 2;
return temp; // discarded after call
}</code>Calldata
Calldata is a read-only, non-modifiable location holding function arguments for external calls. It is the cheapest option because nothing is copied.
Use it for function parameters you only need to read.
<code>function sum(uint[] calldata values) external pure returns (uint total) {
for (uint i = 0; i < values.length; i++) {
total += values[i];
}
}</code>Storage Pointer to State
Inside a function, a storage local variable is a pointer to existing state. Modifying it changes the contract's actual storage.
<code>struct User { uint balance; }
mapping(address => User) users;
function credit() public {
User storage u = users[msg.sender];
u.balance += 100; // updates real state
}</code>Memory Copy of State
If you instead load state into a memory variable, you get an independent copy. Changes to the copy do NOT update the contract's storage.
This is a common source of bugs for beginners.
<code>function noEffect() public view returns (uint) {
User memory u = users[msg.sender];
u.balance += 100; // only the copy changes
return u.balance; // storage is untouched
}</code>Gas Cost Comparison
The cost ranking, from cheapest to most expensive:
- calldata - read-only external args, no copy
- memory - temporary, copied into RAM
- storage - persistent on-chain, very costly to write
A fresh storage write (zero to non-zero) is one of the priciest EVM operations.
Prefer calldata for external
For external functions whose array or struct parameters are only read, use calldata instead of memory to avoid an unnecessary copy and save gas.
<code>// Cheaper: no copy is made
function process(bytes calldata data) external {
// read data directly
}</code>Where Defaults Apply
State variables are always storage. Function parameters and locals of reference type must declare a location explicitly. Value types and mapping members default to storage rules automatically.
Local mappings must always be storage - they cannot live in memory.
<code>// mapping locals must be storage
function f() public {
// mapping(...) memory m; // ERROR
}</code>Caching Storage Reads
A gas optimization: if you read the same storage variable multiple times, copy it into a memory local once. Each storage read (SLOAD) costs gas.
<code>function loop() public view returns (uint) {
uint cached = total; // one SLOAD
uint acc;
for (uint i = 0; i < 10; i++) {
acc += cached; // reads memory, cheap
}
return acc;
}</code>Returning Memory
Functions that build and return new arrays or structs use memory for the return value, since storage cannot be returned by value and calldata is input-only.
<code>function pair(uint a, uint b) public pure returns (uint[] memory) {
uint[] memory out = new uint[](2);
out[0] = a;
out[1] = b;
return out;
}</code>Quick Check
Test your understanding of data locations.
Recap
You learned how Solidity data locations work:
- storage - persistent, on-chain, most expensive to write
- memory - temporary scratchpad, copied, cleared after the call
- calldata - read-only external args, cheapest, no copy
Use storage references to mutate state, memory for temporary work, and calldata for read-only external inputs to save gas.
Frequently asked questions
Is the “Storage vs Memory” lesson free?
Yes — the full text of “Storage vs Memory” is free to read here on the web, and the Web3 & DApp Development Fundamentals course includes 4 lessons in total. To practise it interactively (a built-in code editor and a 24/7 AI tutor) and unlock the rest of the Web3 & DApp Development Fundamentals course, upgrade to CoddyKit PRO.
What will I learn in “Storage vs Memory”?
Data location. You practise Web3 & DApp Development Fundamentals with hands-on code you run directly in the browser, and a 24/7 AI tutor answers your questions as you work through the lesson.
Do I need any experience to start Web3 & DApp Development Fundamentals?
No prior experience is required. Web3 & DApp Development Fundamentals on CoddyKit is structured for beginners through advanced learners; this is — lesson 3 of 4, so you can start here or from the beginning and move at your own pace.
How long does the “Storage vs Memory” lesson take?
Most CoddyKit lessons take about 5–10 minutes. Each one is bite-sized and interactive, so you make steady progress and pick up exactly where you left off across the web and the app.
Can I write and run code in this Web3 & DApp Development Fundamentals lesson?
Yes. Every Web3 & DApp Development Fundamentals lesson includes a built-in code editor, so you write and run real code right in your browser and get instant AI feedback — no local setup required.
All lessons in this course
- Value Types
- Reference Types
- Storage vs Memory
- Constants and Immutables