[go: up one dir, main page]
Scribd
Upload
78 views2 pages

Memory and Resource Management

The document discusses memory and resource management in modern operating systems, focusing on techniques like virtual memory, demand paging, and page replacement policies. It also covers huge pages, NUMA-aware memory management, and resource allocation in cloud-native environments. Additionally, it includes an activity to simulate demand paging and evaluate page replacement algorithms such as FIFO and LRU.

Uploaded by

priya selvakumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
78 views2 pages

Memory and Resource Management

The document discusses memory and resource management in modern operating systems, focusing on techniques like virtual memory, demand paging, and page replacement policies. It also covers huge pages, NUMA-aware memory management, and resource allocation in cloud-native environments. Additionally, it includes an activity to simulate demand paging and evaluate page replacement algorithms such as FIFO and LRU.

Uploaded by

priya selvakumar
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as DOCX, PDF, TXT or read online on Scribd
You are on page 1/ 2

Memory and Resource Management in

Modern Operating Systems


1. Virtual Memory
Virtual memory is a memory management technique that provides an abstraction of a large,
uniform memory space to processes, regardless of the actual physical memory size.
• Allows processes to use more memory than physically available.
• Provides process isolation and security.
• Implemented using paging and segmentation.

Diagram:
[Process Virtual Address Space] → [Virtual Memory Manager] → [Physical Memory + Disk]

2. Demand Paging
Demand paging is a technique where a page is loaded into memory only when it is required
during execution.
• Reduces initial load time.
• Page fault occurs when the requested page is not in memory.
• Requires page replacement policies to handle memory full conditions.

Diagram:
[Process Request] → [Page Table Check] → (If page not present → Page Fault → Load from
Disk)

3. Page Replacement Policies


When physical memory is full, page replacement policies decide which page to evict:
• FIFO (First-In-First-Out): Replaces the oldest page.
• LRU (Least Recently Used): Replaces the least recently accessed page.
• Optimal: Replaces the page that will not be used for the longest future duration.
• Random: Selects a random page for replacement.

4. Huge Pages
• Huge pages are memory pages significantly larger than standard (e.g., 2MB or 1GB vs.
4KB).
• Reduce TLB (Translation Lookaside Buffer) misses.
• Improve performance for memory-intensive workloads like databases and virtual
machines.
5. NUMA-aware Memory Management
NUMA (Non-Uniform Memory Access) systems consist of multiple CPUs, each with its own
local memory.
• Access to local memory is faster than remote memory.
• NUMA-aware OS schedules processes close to their memory.
• Balances performance by reducing remote memory access.

Diagram:
[CPU1 ↔ Local Memory1] [CPU2 ↔ Local Memory2] (Remote access is slower)

6. Resource Allocation in Cloud-native Environments


• In cloud-native systems (Kubernetes, Docker), resources like CPU, memory, and storage
are dynamically allocated.
• Resource allocation policies ensure fairness, isolation, and scalability.
• Techniques:
- Cgroups and namespaces for isolation.
- Horizontal scaling (adding more instances).
- Vertical scaling (increasing resources of a single instance).
• Auto-scaling based on workload demand.

7. Activity: Simulate Demand Paging and Page Replacement Algorithms


Objective: Simulate demand paging and evaluate page replacement algorithms.
Steps:
1. Define a reference string of page requests.
2. Allocate limited memory frames.
3. Apply page replacement algorithms (FIFO, LRU, Optimal).
4. Count page faults and compare performance.
Example:
Reference string: 7, 0, 1, 2, 0, 3, 0, 4
Frames = 3 → Apply FIFO → Track page faults.

You might also like