Scribd
Upload
11 views

Memory Organization

Uploaded by

Krishna Shrivastava
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
11 views

Memory Organization

Uploaded by

Krishna Shrivastava
Copyright
© © All Rights Reserved
Available Formats
Download as PPTX, PDF, TXT or read online on Scribd
You are on page 1/ 9

Memory

Organization:

Main Memory, Associative Memory, Cache Memory - Organization and


Mappings
Memory
Hierarchy
Memory hierarchy is a way of organizing computer
memory in a system. It consists of different levels of
memory, each with different characteristics and access
times. The memory hierarchy is designed to optimize the
performance and efficiency of the system by placing
frequently accessed data in faster and more expensive
memory levels, while less frequently accessed data is
stored in slower and cheaper memory levels. The
different levels of the memory hierarchy include:

Register – Cache -- Main Memory -- Secondary


Storage
Main Memory

 Main memory, often referred to as RAM (Random Access Memory),


plays a crucial role in computing by providing temporary storage for
data and instructions that the CPU (Central Processing Unit) needs to
access quickly.
 RAM is volatile memory, meaning that it loses its contents when the
power is turned off. Its primary function is to hold data that is actively
being used by the CPU, allowing for fast retrieval and manipulation.
Unlike storage devices such as hard drives or SSDs, which store data
permanently, RAM is much faster but has limited capacity and is more
expensive per unit of storage.
 The CPU continuously reads and writes data to and from RAM during
program execution, making it an essential component in overall system
performance.
Associative Memory

 Associative memory is a type of computer memory that allows for


data retrieval based on content rather than specific memory
addresses. It enables parallel search and retrieval of information,
making it useful for tasks such as caching, database management,
and pattern recognition. Associative memory is implemented using
content-addressable memory (CAM) or hash tables, facilitating
efficient data access by comparing query content with stored keys or
tags. It's valuable for searching large databases, quick data retrieval,
and implementing associative arrays in programming languages.
Cache
Memory
 Cache memory is a high-speed type of memory located between the
CPU and the main memory (RAM) in a computer system. Its primary
purpose is to store frequently accessed data and instructions,
allowing the CPU to access them quickly without needing to fetch
them from the slower main memory every time they are needed.
 Cache memory operates on the principle of a hierarchy, with multiple
levels of cache (e.g., L1, L2, L3) arranged in increasing size and
decreasing speed. The closer the cache is to the CPU, the faster it
operates but with lower capacity.
Advantage of Cache
Memory
 The key benefit of cache memory is that it significantly improves
system performance by reducing the time it takes for the CPU to
access data. This is achieved through the principle of locality,
which states that programs tend to access the same data and
instructions repeatedly or nearby in memory. By keeping a copy
of this frequently accessed data in the cache, the CPU can
retrieve it much faster, leading to shorter execution times and
overall faster performance for the system.
Cache Memory Organization
Cache memory organization and mappings refer to how data is stored and accessed
within the cache memory system. There are several techniques used to organize and
map data in cache memory:
1. Direct-mapped Cache:
1. In direct-mapped cache, each block of main memory can be mapped to only one specific
cache location.
2. The mapping is determined by dividing the main memory into blocks and assigning each
block to a unique cache location based on a modulo function.
3. While this method is simple and requires less hardware, it may lead to conflicts where
multiple memory blocks map to the same cache location (known as cache collisions).
 Fully-Associative Cache:
 In fully-associative cache, each block of main memory can be
mapped to any cache location, without any restrictions.
 This allows for maximum flexibility in mapping data and reduces
the likelihood of cache collisions.
 However, it requires additional hardware, such as a content-
addressable memory (CAM), to perform the associative mapping.

 Set-Associative Cache:
 Set-associative cache combines elements of direct-mapped and
fully-associative caches.
 It divides the cache into sets, with each set containing multiple
cache lines.
 Each block of main memory can be mapped to any cache line
within a specific set.
 This approach reduces the likelihood of cache collisions compared
to direct-mapped cache while still maintaining simplicity and
efficiency.
Cache Mappings:

 Cache mappings determine how data is located within the cache


memory.
 Common cache mapping techniques include write-through, write-
back, and write-allocate.
 Write-through cache writes data simultaneously to both the cache
and main memory, ensuring consistency but reducing performance.
 Write-back cache writes data to the cache first and only updates
main memory when the cache block is evicted, improving
performance but potentially leading to data inconsistency.
 Write-allocate cache fetches data from main memory into the
cache upon a cache miss before performing a write operation,
optimizing performance by reducing memory accesses.
 Overall, cache memory organization and mappings play a crucial role in determining the efficiency and
performance of the cache system, balancing factors such as speed, complexity, and resource utilization.

You might also like