Memory Management

Lecture

Sadaqat Ali Khan Bangash

Institute of Information Technology,
University of Science & Technology Bannu
Memory Management Sadaqat Ali Khan Bangash 1

Memory Hierarchies
Like everything else in computers, Computer Memory is also organized into a hierarchy. This is as follows:
At 1st Level (closest to Processor) are the registers At 2nd Level, one or two Levels of Cache Memory At 3rd Level, Main Memory (also known as primary memory)

Memory Management

Sadaqat Ali Khan Bangash

Memory Hierarchies .
All above are considered part of Internal Memory (i.e. Internal to Computer System). Then the Hierarchy continues onto External Memory
At 4th Level, Secondary Storages (Hard disks, tapes etc) At 5th Level, other forms of memory comes such as Zip Disks, Floppy Drives, etc

Memory Management

Sadaqat Ali Khan Bangash

Internal Memory

Register, Cache, Main Memory

External Memory

Hard disk. CD-Rom,DVD, Internal Storage

Memory Management Sadaqat Ali Khan Bangash

Memory Hierarchies .
Registers
These temporary locations within the CPU are extremely fast, very small, very expensive & volatile.

Cache
Buffer memory Memory. regions between CPU & Main

Memory Management

Sadaqat Ali Khan Bangash

Memory Hierarchies .
Main Memory
Main Memory also called Primary Memory, Random Access Memory (RAM). Hundreds of Megabytes of Medium Speed & Volatile

Secondary Memory
Hundreds of Gigabytes of memory with Slow speed We may recall the factors of Cost, Size, and Speed of access that are the orders in which Memory components are divided into this hierarchy.

Memory Management

Sadaqat Ali Khan Bangash

Purpose of Memory Management

To ensure fair use of memory for processes To ensure secure access, i.e., space allocated for process is not allocated for another process, or a process does not have access to anything beyond its "process address space To ensure orderly memory access, i.e., when memory is to be allocated, when de-allocated, etc. etc. To ensure efficient use of memory. Since memory is expensive, it should be used efficiently
Sadaqat Ali Khan Bangash 7

Memory Management

Uni-programming environment

In a Uni-programming environment, main

memory is divided into two parts:

1. One part for the Operating System

2. One part for the Program currently being executed. (or the User Part)

Memory Management

Sadaqat Ali Khan Bangash

Multi-programming environment
In a multi-programming environment, the user part of the memory is further divided to accommodate multiple processes. This task of dividing and allocation is carried out dynamically by the Operating System and is known as memory management. Memory Manager has the following basic responsibilities: 1. Keep track of used & free memory spaces (Also known as Free Space Management) 2. When, where and how much memory to allocate or deallocate 3. Swapping processes in and out of main memory to secondary memory.

Memory Management

Sadaqat Ali Khan Bangash

Logical & Physical Addresses

Logical Address

An address generated by the CPU This normally refers to an instruction or data within the process address space
Physical Address An address for a main memory location where instruction

or data resides

Memory Management

Sadaqat Ali Khan Bangash

10

Logical & Physical Addresses

Since the Process is already loaded in memory, the logical addresses of the process need to be translated to its corresponding physical addresses. The CPU generates all logical Addresses for a particular process. A set of these logical addresses constitute the "Logical Address Space". Correspondingly, all the physical addresses that translate back to the logical addresses are known as the "Physical Address Space" for that process. This will be explained in detail in paging & segmentation. The logical address is also known as the Offset Address. So in simpler words

Memory Management

Sadaqat Ali Khan Bangash

11

Logical & Physical Addresses

Logical Address Space
The set of all logical addresses generated by a process

Physical Address Space

The set of all physical addresses corresponding to those logical addresses

This translation or mapping from Logical to Physical address is done by a piece of CPU hardware known as the Memory Management Unit.
Memory Management Sadaqat Ali Khan Bangash 12

MMU

Process

Logical Address

Physical Address

CPU
346 14346

Base or Relocation Register Value: 14000

14000

The MMU will check if the logical address is within the limit specified by the limit register. If yes, it will add the value of the relocation register. If no, it will give an addressing error or a trap.
Memory Management Sadaqat Ali Khan Bangash 13

Using Memory Efficiently

Dynamic Loading
With dynamic loading, a routine is not loaded into main memory until it is called. All these routines are kept on the disk in a relocatable format. The main program is loaded into memory & executed. The best example for explaining this would be MS-Word. Word had thousands of features. At any given time, a user will only be using a couple of these features. The routines for those additional features are kept on disk whereas only the main program for MS-Word is loaded. Dynamic Loading is also used in handling exceptions or errors. Routines for handling exceptions or errors are only loaded when they occur, otherwise they will remain on disk.

Memory Management

Sadaqat Ali Khan Bangash

14

Using Memory Efficiently

Advantages:
Less time needed to load a program Less memory space needed for program

Disadvantages:
More run-time activity. Lot of I/O requests are made which is time consuming.

Memory Management

Sadaqat Ali Khan Bangash

15

Dynamic Linking
Advantages: Less time needed to load a program Less memory space needed Less disk space needed to store binaries Updated libraries are used without recompiling a program Disadvantages: More run-time activity resulting in slower program execution
Memory Management Sadaqat Ali Khan Bangash 16

Overlays
Overlays allow a process to be larger than the amount of memory allocated to it. It works by keeping only those instructions in memory that are needed at any given time. I.e., lets suppose our Main Memory is 100 MB, will an executable of 110 MB will be able to run? The answer is yes if the underlying Operating Systems employs Virtual Memory techniques. This is the concept behind Virtual Memory. It is essential to understand Overlays because it gives a flavor of Virtual Memory which we would be cover later on.
Memory Management Sadaqat Ali Khan Bangash 17

Overlays
When other instructions & data are needed, they are loaded into the space occupied previously by instructions that are no longer needed. Overlays are implemented solely by the programmer. If the program does not have support for Virtual Memory or Overlays, then the Operating System cannot do anything about it.

Memory Management

Sadaqat Ali Khan Bangash

18

Swapping
A process can be swapped temporarily out of memory to a backing store, and then brought back into memory for continued execution. Backing store fast disk large enough to accommodate copies of all memory images for all users; must provide direct access to these memory images Roll out, roll in swapping variant used for priority-based scheduling algorithms; lower-priority process is swapped out so higher-priority process can be loaded and executed Major part of swap time is transfer time; total transfer time is directly proportional to the amount of memory swapped
Memory Management Sadaqat Ali Khan Bangash 19

Swapping
Main Memory Secondary Memory The area where swapped processes are kept in secondary memory is known as Swap Space
Memory Management Techniques
Multi-Programming with Fixed Number of Tasks (MFT) Multi-Programming with Variable Number of Tasks (MVT) Paging Segmentation Virtual Memory
Sadaqat Ali Khan Bangash 20

Memory Management

Swapping
Modified versions of swapping are found on many systems (i.e., UNIX, Linux, and Windows)
System maintains a ready queue of ready-to-run processes which have memory images on disk

Memory Management

Sadaqat Ali Khan Bangash

21

Contiguous Allocation
Main memory usually divided into two partitions:
Resident operating system, usually held in low memory with interrupt vector User processes then held in high memory

Relocation registers used to protect user processes from each other, and from changing operatingsystem code and data
Base register contains value of smallest physical address Limit register contains range of logical addresses each logical address must be less than the limit register MMU maps logical address dynamically
Memory Management Sadaqat Ali Khan Bangash 22

HW address protection with base and limit registers

Memory Management

Sadaqat Ali Khan Bangash

23

Contiguous Allocation (Cont.)

Multiple-partition allocation
Hole block of available memory; holes of various size are scattered throughout memory When a process arrives, it is allocated memory from a hole large enough to accommodate it Operating system maintains information about: a) allocated partitions b) free partitions (hole)
OS process 5 OS process 5 OS process 5 process 9 process 8 process 2 process 2 process 2 OS process 5 process 9

process 10
process 2

Memory Management

Sadaqat Ali Khan Bangash

24

Dynamic Storage-Allocation Problem

How to satisfy a request of size n from a list of free holes

First-fit: Allocate the first hole that is big enough Best-fit: Allocate the smallest hole that is big enough; must search entire list, unless ordered by size Produces the smallest leftover hole Worst-fit: Allocate the largest hole; must also search entire list Produces the largest leftover hole
First-fit and best-fit better than worst-fit in terms of speed and storage utilization

Memory Management

Sadaqat Ali Khan Bangash

25

Fragmentation
External Fragmentation total memory space exists to satisfy a request, but it is not contiguous Internal Fragmentation allocated memory may be slightly larger than requested memory; this size difference is memory internal to a partition, but not being used Reduce external fragmentation by compaction Shuffle memory contents to place all free memory together in one large block Compaction is possible only if relocation is dynamic, and is done at execution time I/O problem Latch job in memory while it is involved in I/O Do I/O only into OS buffers
Memory Management Sadaqat Ali Khan Bangash 26

End of Lecture
Questions

Memory Management

Sadaqat Ali Khan Bangash

27