My Notes

Study Timer
25:00
Today: 0 min
Total: 0 min
🏆

Achievement Unlocked!

Description

+50 XP

Chapter 1: Network Architecture, OS, Data Storing and Fault Tolerance

Reading Timer
25:00
Chapter 1: Network Architecture, OS, Data Storing and Fault Tolerance

Chapter 1: Network Architecture, OS, Data Storing and Fault Tolerance

From Mainframes to Modern Storage: Building Resilient Network Infrastructure

🖥️ Architecture 💾 Storage 🛡️ Fault Tolerance
🏛️ Section 1: Computing Architecture Fundamentals

1.1 Mainframe Architecture: The Backbone of Enterprise Computing

Mainframe computers represent the pinnacle of reliable, high-performance computing for large-scale organizational needs. Understanding mainframe architecture provides insight into modern enterprise systems.

Modern data center with mainframe systems
Figure 1.1: Enterprise mainframe environment with high-availability configurations

Key Characteristics of Mainframe Architecture:

Reliability, Availability, Serviceability (RAS)

Mainframes are engineered for maximum uptime with redundant components, hot-swappable parts, and predictive failure analysis. They achieve 99.999% availability (less than 6 minutes downtime per year).

Massive I/O Throughput

Designed to handle thousands of simultaneous connections and transactions. Mainframes excel at processing massive volumes of data with consistent performance, making them ideal for banking, airlines, and government systems.

Vertical Scalability

Scale-up architecture allows adding processors, memory, and I/O channels without disrupting operations. Modern mainframes can have hundreds of processors and terabytes of RAM.

Mainframe Components:

  • Central Processor Complex (CPC): Contains processors, memory, and I/O subsystems
  • Logical Partitions (LPARs): Virtual machines that divide physical resources
  • Channel Subsystems: Dedicated I/O processors that offload work from CPUs
  • Coupling Facilities: Specialized hardware for clustering and data sharing
💡

Modern Relevance: Despite cloud computing growth, mainframes still process over $3 trillion in daily commerce transactions. They remain critical for industries requiring extreme reliability and security.

1.2 Client-Server Architecture Models

Client-server architecture revolutionized distributed computing by separating requesters (clients) from providers (servers) of services and resources.

Client-server network architecture diagram
Figure 1.2: Multi-tier client-server architecture with distributed components

Client-Server Models:

Model Description Use Cases Advantages
Two-Tier Client directly connects to server (database) Small applications, desktop software Simple, fast development, low latency
Three-Tier Client → Application Server → Database Server Web applications, enterprise systems Scalability, security, maintainability
N-Tier Multiple layers with specialized functions Large enterprise applications Flexibility, reusability, independent scaling
Peer-to-Peer Nodes act as both clients and servers File sharing, blockchain Decentralized, no single point of failure

File Server Architecture:

Specialized servers that provide shared storage and file access to network clients.

  • Centralized Storage: Single location for data management and backup
  • Access Control: Permissions and authentication for file security
  • File Locking: Prevents conflicts when multiple users access same file
  • Caching: Improves performance by storing frequently accessed files locally
Upsizing vs. Downsizing

Upsizing: Moving from smaller systems (PCs) to larger systems (servers/mainframes) for better performance and reliability. Downsizing: Moving from large systems to distributed smaller systems for cost savings and flexibility. Right-sizing: Choosing the optimal platform based on requirements.

🖥️ Section 2: Server Types and Hardware

2.1 Physical Server Form Factors

Understanding different server hardware configurations is essential for infrastructure planning and deployment.

Different server types: tower, rack, and blade servers
Figure 2.1: Comparison of tower, rack-mounted, and blade server architectures

Tower Servers:

Characteristics

Standalone servers resembling desktop PCs in upright tower cases. Ideal for small businesses and entry-level applications.

  • Advantages: Low cost, quiet operation, easy to expand, no rack required
  • Disadvantages: Large footprint, limited scalability, difficult cable management
  • Best For: Small offices, file/print servers, development environments

Rack Servers:

Characteristics

Servers designed to mount in standardized 19-inch racks. Measured in rack units (U), where 1U = 1.75 inches height.

  • Advantages: High density, efficient space utilization, organized cabling, easy maintenance
  • Disadvantages: Requires rack infrastructure, can be noisy, needs proper cooling
  • Common Sizes: 1U (compact), 2U (balanced), 4U (high-performance)

Blade Servers:

Characteristics

Modular servers (blades) that slide into a chassis providing shared power, cooling, networking, and management.

  • Advantages: Maximum density, reduced cabling, shared infrastructure, hot-swappable
  • Disadvantages: Higher initial cost, vendor lock-in, complex cooling requirements
  • Best For: Data centers, cloud providers, high-performance computing clusters
Feature Tower Rack Blade
Density Low Medium-High Very High
Cost $ $$ $$$
Scalability Limited Good Excellent
Power Efficiency Medium Good Excellent
💾 Section 3: File Systems and Storage Management

3.1 File System Fundamentals

File systems provide the structure for storing, organizing, and retrieving data on storage devices. Different operating systems use different file systems optimized for their environments.

File system hierarchy and structure diagram
Figure 3.1: File system organization showing directory structures and data blocks

Windows File Systems:

FAT (File Allocation Table)

Legacy file system with simple structure. Variants: FAT16, FAT32, exFAT. Limitations: 4GB file size limit (FAT32), no journaling, limited security. Use: USB drives, SD cards, compatibility with multiple OS.

NTFS (New Technology File System)

Modern Windows file system with advanced features. Features: Journaling, file compression, encryption (EFS), disk quotas, ACLs, large file support (16EB max). Use: System drives, enterprise storage.

Linux/Unix File Systems:

EXT (Extended File System)

Linux native file systems. EXT3: Added journaling to EXT2. EXT4: Current standard with extents, delayed allocation, and improved performance. Features: Journaling, large file support (16TB), backward compatibility.

XFS

High-performance 64-bit journaling file system. Strengths: Excellent for large files, parallel I/O, fast recovery. Use: Enterprise servers, media storage.

ZFS (Zettabyte File System)

Advanced file system with volume management. Features: Snapshots, compression, deduplication, RAID-Z, data integrity verification, copy-on-write. Use: High-end storage servers, NAS appliances.

Network File Systems:

NFS (Network File System)

Unix/Linux network file sharing protocol. Versions: NFSv3 (stable), NFSv4 (improved security, stateful). Use: Unix-to-Unix file sharing, home directories.

CIFS/SMB (Common Internet File System)

Windows network file sharing protocol (SMB is the protocol, CIFS is an implementation). Use: Windows file/printer sharing, Samba for Linux-Windows interoperability.

🔧

File System Selection: Choose based on requirements: NTFS for Windows compatibility, EXT4 for Linux servers, XFS for large files, ZFS for data integrity and snapshots, NFS for Unix networks, SMB/CIFS for mixed environments.

3.2 Mounting and Managing File Systems

Understanding how to mount, unmount, and manage file systems is essential for system administration.

Mounting Concepts:

Mount Point

A directory in the file system hierarchy where a storage device or partition is attached and becomes accessible.

bash
# Linux Mounting Commands

# View currently mounted file systems
mount

# Mount a device manually
mount /dev/sdb1 /mnt/data

# Mount with options (read-write, no execution)
mount -o rw,noexec /dev/sdb1 /mnt/data

# Unmount a file system
umount /mnt/data

# View disk space usage
df -h

# Check file system integrity
fsck /dev/sdb1

# Permanent mounting via /etc/fstab
# /dev/sdb1  /mnt/data  ext4  defaults  0  2

File System Management Tasks:

  • Creating File Systems: mkfs.ext4 /dev/sdb1 or mkfs.ntfs /dev/sdb1
  • Checking Disk Usage: df -h (disk free), du -sh /path (disk usage)
  • File System Check: fsck to repair corrupted file systems
  • Quota Management: Limit user/disk space with edquota, repquota
  • Permissions: chmod (change mode), chown (change owner)
⚠️

Critical Warning: Never run fsck on a mounted file system. Always unmount first or boot from rescue media. Running fsck on mounted filesystems can cause data corruption.

💽 Section 4: Disk Technologies and Storage

4.1 Disk Interface Technologies

Storage interfaces determine how disks connect to computers and significantly impact performance, cost, and use cases.

Different disk drive types and interfaces
Figure 4.1: Evolution of storage interfaces from ATA to NVMe

SATA (Serial ATA):

Characteristics

Consumer-grade interface for HDDs and SSDs. Speeds: SATA I (1.5 Gb/s), SATA II (3 Gb/s), SATA III (6 Gb/s). Advantages: Low cost, wide compatibility, hot-swappable. Use: Desktop PCs, budget servers, bulk storage.

SAS (Serial Attached SCSI):

Characteristics

Enterprise-grade interface with higher performance and reliability. Speeds: SAS-3 (12 Gb/s), SAS-4 (24 Gb/s). Advantages: Better reliability, dual-port capability, higher IOPS, longer cable lengths. Use: Enterprise servers, high-performance storage arrays.

SCSI (Small Computer System Interface):

Characteristics

Legacy parallel interface, largely replaced by SAS. Historical Significance: Dominated enterprise storage before SATA/SAS. Modern Use: Still found in legacy systems and specialized applications.

NVMe (Non-Volatile Memory Express):

Characteristics

Modern protocol designed specifically for SSDs over PCIe. Speeds: NVMe 1.3 (up to 32 Gb/s per lane), NVMe 2.0 (even faster). Advantages: Lowest latency, highest IOPS, parallelism. Use: High-performance computing, databases, caching layers.

Interface Max Speed Cost Use Case Reliability
SATA III 6 Gb/s $ Consumer/Bulk Medium
SAS-3 12 Gb/s $$ Enterprise High
NVMe PCIe 4.0 64 Gb/s $$$ High-Performance High

4.2 Storage Area Networks (SAN)

SANs provide block-level storage access over dedicated networks, separating storage from servers for better management and scalability.

SAN Fundamentals:

What is a SAN?

A dedicated high-speed network that provides access to consolidated block-level storage. Servers see SAN storage as locally attached disks, not network shares.

SAN Protocols:

Fibre Channel (FC)

Dedicated high-speed network technology (2/4/8/16/32 Gb/s). Advantages: Low latency, high throughput, lossless. Disadvantages: Expensive, requires specialized hardware (HBAs, FC switches).

iSCSI (Internet SCSI)

Encapsulates SCSI commands over TCP/IP networks. Advantages: Uses existing Ethernet infrastructure, lower cost. Disadvantages: Higher latency than FC, depends on network quality.

FCoE (Fibre Channel over Ethernet)

Encapsulates FC frames over Ethernet networks. Advantages: Converges storage and data networks, maintains FC features. Requirements: Lossless Ethernet (DCB), specialized switches.

SAN Components:

  • Storage Arrays: High-end disk systems with RAID, caching, and management
  • HBAs (Host Bus Adapters): Specialized network cards for SAN connectivity
  • Switches: FC switches or converged Ethernet switches
  • Zoning: Logical segmentation for security and traffic management
  • LUN Masking: Controls which servers can access which storage volumes
🌐

SAN vs NAS: SAN provides block-level storage (appears as local disk), while NAS provides file-level storage (appears as network share). SAN is faster but more complex; NAS is simpler but has more overhead.

🛡️ Section 5: Fault Tolerance and RAID

5.1 RAID (Redundant Array of Independent Disks)

RAID combines multiple physical disks into a single logical unit for improved performance, redundancy, or both. Understanding RAID levels is critical for data protection strategies.

RAID configuration diagrams showing different levels
Figure 5.1: Common RAID levels showing data striping, mirroring, and parity

RAID 0 (Striping):

Performance Without Redundancy

Data is split (striped) across multiple disks. Advantages: Maximum performance, full capacity utilization. Disadvantages: No fault tolerance—failure of one disk loses all data. Minimum Disks: 2. Use: Temporary data, caching, non-critical high-performance needs.

RAID 1 (Mirroring):

Complete Redundancy

Data is duplicated (mirrored) on two or more disks. Advantages: Excellent read performance, complete data protection, simple recovery. Disadvantages: 50% capacity overhead (2 disks store 1 disk worth of data). Minimum Disks: 2. Use: OS drives, critical data requiring high availability.

RAID 5 (Striping with Distributed Parity):

Balance of Performance and Protection

Data and parity information are striped across all disks. Advantages: Good read performance, fault tolerance (survives 1 disk failure), efficient storage (only 1 disk worth of overhead). Disadvantages: Slow write performance (parity calculation), rebuild time can be long. Minimum Disks: 3. Use: File servers, general-purpose storage.

RAID 6 (Striping with Double Parity):

Enhanced Fault Tolerance

Like RAID 5 but with two parity blocks. Advantages: Survives 2 simultaneous disk failures. Disadvantages: Higher overhead (2 disks worth), slower writes than RAID 5. Minimum Disks: 4. Use: Large arrays, critical data with long rebuild times.

RAID 10 (1+0):

Mirroring + Striping

Combines RAID 1 mirrors with RAID 0 striping. Advantages: Excellent performance and redundancy, fast rebuild. Disadvantages: 50% capacity overhead (expensive). Minimum Disks: 4. Use: Databases, high-performance critical applications.

RAID Level Min Disks Capacity Fault Tolerance Read Speed Write Speed
RAID 0 2 100% None Excellent Excellent
RAID 1 2 50% 1 disk Good Good
RAID 5 3 (N-1)/N 1 disk Excellent Fair
RAID 6 4 (N-2)/N 2 disks Excellent Poor
RAID 10 4 50% 1 per mirror Excellent Excellent
⚠️

RAID is NOT Backup: RAID protects against hardware failure but not against deletion, corruption, malware, or disasters. Always maintain separate backups in addition to RAID.

🖥️ Section 6: Network Operating Systems

6.1 Linux Server Administration

Linux dominates server environments due to its stability, security, flexibility, and cost-effectiveness. Understanding Linux server fundamentals is essential for system administrators.

Linux server terminal and command line interface
Figure 6.1: Linux server management through command-line interface

Popular Linux Distributions:

Red Hat Enterprise Linux (RHEL)

Commercial distribution with enterprise support. Use: Enterprise servers, mission-critical applications. Package Manager: YUM/DNF (RPM packages).

Ubuntu Server

Debian-based distribution with long-term support (LTS) versions. Use: Cloud servers, web hosting, development. Package Manager: APT (DEB packages).

CentOS/AlmaLinux/Rocky Linux

RHEL-compatible free distributions. Use: Budget-conscious enterprises, testing environments. Note: CentOS shifted to Stream; AlmaLinux/Rocky are community alternatives.

SUSE Linux Enterprise Server (SLES)

Enterprise distribution popular in Europe. Use: SAP environments, mainframe integration. Package Manager: Zypper (RPM packages).

Core Linux Server Skills:

  • User Management: useradd, usermod, passwd, /etc/passwd, /etc/shadow
  • File Permissions: chmod (755, 644), chown, umask
  • Process Management: ps, top, kill, systemctl
  • Network Configuration: ip addr, nmcli, /etc/network/interfaces
  • Package Management: apt (Debian/Ubuntu), yum/dnf (RHEL/CentOS)
  • Log Management: journalctl, /var/log/, rsyslog
🐧

Linux Philosophy: "Do one thing and do it well." Linux tools are modular and composable. Master the command line, understand file permissions, and learn to chain commands with pipes (|) for powerful automation.

6.2 Windows Server Administration

Windows Server provides enterprise-grade services with graphical management tools and deep integration with Microsoft ecosystems.

Windows Server Editions:

Windows Server Standard

General-purpose server OS. Features: File/print services, Active Directory, Hyper-V (2 VMs included). Use: Small to medium businesses, branch offices.

Windows Server Datacenter

High-virtualization environments. Features: Unlimited VMs, Storage Spaces Direct, Shielded VMs. Use: Large enterprises, cloud providers, highly virtualized data centers.

Core Windows Server Roles:

  • Active Directory Domain Services (AD DS): Centralized authentication and directory services
  • DNS Server: Name resolution for the network
  • DHCP Server: Automatic IP address assignment
  • File and Storage Services: File sharing, DFS, Storage Spaces
  • Hyper-V: Native hypervisor for virtualization
  • Web Server (IIS): Host web applications and services
  • Print Services: Network printer management

Management Tools:

  • Server Manager: Central dashboard for managing roles and features
  • PowerShell: Command-line automation and scripting
  • Group Policy: Centralized configuration management
  • Windows Admin Center: Modern web-based management interface
powershell
# PowerShell Server Administration Examples

# Get server information
Get-ComputerInfo

# List installed roles and features
Get-WindowsFeature -Installed

# Install a role (Web Server)
Install-WindowsFeature -Name "Web-Server" -IncludeManagementTools

# Get running services
Get-Service -Status Running

# Check disk space
Get-Volume | Select-Object DriveLetter, SizeRemaining, Size

# View event logs (last 20 errors)
Get-EventLog -LogName "System" -EntryType Error -Newest 20
🪟

Server Core: Windows Server can be installed without GUI (Server Core mode) for reduced attack surface, lower resource usage, and fewer updates. Manage remotely via PowerShell or Windows Admin Center.

📚 Section 7: Key Concepts Review

7.1 Chapter Summary and Essential Takeaways

This chapter established the foundational knowledge of system architecture, storage technologies, and operating systems that underpins all system administration work.

Essential Takeaways:

  • Architecture Matters: Choose the right server form factor (tower/rack/blade) based on density, cost, and scalability requirements.
  • File System Selection: Match file systems to use cases: NTFS for Windows, EXT4/XFS for Linux, ZFS for advanced features, NFS/SMB for network sharing.
  • Storage Performance: Understand interface trade-offs: SATA (cost), SAS (reliability), NVMe (speed).
  • RAID is Essential: Implement appropriate RAID levels for data protection, but remember RAID ≠ Backup.
  • OS Expertise: Master both Linux (command-line, permissions, services) and Windows Server (roles, PowerShell, AD) for comprehensive administration skills.
  • SAN/NAS Knowledge: Understand block vs. file storage, and protocols like iSCSI, Fibre Channel, and NFS/SMB.
🎯

Next Steps: Practice these concepts in a lab environment. Set up virtual machines with different OSes, configure RAID arrays (even in software), experiment with file systems, and practice mounting/sharing storage across systems.

Recommended Practice Exercises:

  1. Install both Linux (Ubuntu/CentOS) and Windows Server in VMs
  2. Create and mount different file systems (EXT4, NTFS, XFS)
  3. Configure software RAID using mdadm (Linux) or Storage Spaces (Windows)
  4. Set up NFS and SMB shares between systems
  5. Practice user management, permissions, and quota configuration
  6. Explore disk partitioning with fdisk/parted (Linux) or Disk Management (Windows)
📖

Further Reading: Refer to the course textbooks: "Essential System Administration" by Æleen Frisch for Unix/Linux, "Principles of Network and System Administration" by Mark Burgess for theoretical foundations, and vendor documentation for specific technologies.