This note covers my own study of what mapping of cache is
Cache Mapping Types
In computer architecture, cache mapping is a crucial concept that determines how data from the main memory is stored and accessed in the cache memory. Understanding cache mapping is essential for optimizing system performance. Here’s a detailed explanation of the different types of cache mapping techniques, their attributes, and their significance.
Attributes of Cache Mapping
You will need to understand the following attributes to grasp the concept of cache mapping:
- Tag:
- The tag is a part of the memory address used to identify which memory block is stored in a cache line. It helps in determining if the data in the cache is the same as the requested data.
- Index:
- The index specifies the cache line where a memory block can be placed. It is used to locate the specific cache line within the cache.
- Offset (Block Offset):
- The offset indicates the specific word or byte within a block. It is used to access data within the block once it is loaded into the cache.
- Set:
- In set-associative caches, the cache is divided into sets, each containing a fixed number of lines (ways). The set index determines which set a memory block maps to.
- Way:
- In set-associative caches, each set contains multiple lines (ways). The way determines the specific line within a set where the block is stored.
Types of Cache Mapping
1. Direct Mapped Cache
- Description: Each memory block is mapped to exactly one cache line.
- Structure: The address is divided into tag, index, and offset fields.
- Why It Matters: Simple and fast due to straightforward mapping, but can suffer from high collision rates.
2. Fully Associative Cache
- Description: Any memory block can be loaded into any cache line.
- Structure: The address is divided into tag and offset fields; there is no index field.
- Why It Matters: Flexible and reduces collisions, but more complex and slower due to the need to search all cache lines.
3. Set Associative Cache
- Description: Combines aspects of direct-mapped and fully associative caches. The cache is divided into sets, and each set contains a fixed number of lines (ways).
- Structure: The address is divided into tag, set index, and offset fields.
- Why It Matters: Balances simplicity and flexibility, reducing collisions and improving performance compared to direct-mapped cache.
Advantages and Disadvantages
| Mapping Type | Advantages | Disadvantages |
|---|---|---|
| Direct Mapped Cache | - Simple and fast. - Easy to implement. | - High chance of collisions. - Limited flexibility. |
| Fully Associative Cache | - Flexible placement of blocks. - Reduces collisions. | - Complex and slower. - Requires searching all cache lines. |
| Set Associative Cache | - Balances simplicity and flexibility. - Reduces collisions compared to direct-mapped. - Improved performance. | - More complex than direct-mapped. - Still has some collision potential. |
Summary of Cache Mapping Types
| Attribute | Direct Mapped Cache | Fully Associative Cache | Set Associative Cache |
|---|---|---|---|
| Mapping | One-to-one | Any-to-any | Set-based |
| Address Fields | Tag, Index, Offset | Tag, Offset | Tag, Set Index, Offset |
| Collision Rate | High | Low | Moderate |
| Complexity | Low | High | Moderate |
| Performance | Fast | Slower | Balanced |
| To checkout the problems, visit Computer Organization and Architecture Lecture 13. |
References
Information
- date: 2025.03.05
- time: 09:32