Addis Ababa Institute of Technology

School of Information Technology

and Engineering(SiTE)
Cyber Stream

A report for AAiT data center visit

Milko Shuma UGR/4707/13

Submitted to: Mr. Daniel

Dec 31, 2024
 Table contents

1. Data Center Infrastructure Fundamental – International Perspective

1.1. Type of Data Center

1.2. Main Components of a Data Center

1.3. Data Center systems and solutions

1.4. Data Center Standards

1.5. Hot and Cold Aisle Data Center

1.6. Data Center Tier Levels or Classification

1.7. Data Center Design

2. AAiT Data Center Infrastructure (AAiT Perspective)

2.1. AAiT Datacenter Type

2.2. Main Components of AAiT Data Center

2.3. AAiT Data Center systems and solutions

2.4. AAiT Data Center Standards

2.5. AAiT Hot and Cold Aisle Data Center System

2.6. AAiT Data Center Tier Classification

2.7. AAiT Data Center Design

3. Comparison of International Data Center and AAiT

4. Summary of Data center visit

1. Data Center Infrastructure Fundamental – International Perspective

1.1. Type of Data Center

Data centers are critical facilities for storing, managing, and disseminating data and
applications. They can be classified into four main types based on their function,
ownership, and operational requirements.

1. Enterprise Data Centers

These data centers are owned, managed, and operated by a single organization for its
exclusive use.

Purpose: Designed to support the specific business needs of the organization,

such as hosting internal IT systems, applications, and databases.

Location: Typically located on-premises or near the organization’s headquarters

or major facilities.

Advantages:

Full control over the infrastructure, security, and data access.

Customizable to meet the organization's specific requirements.

Disadvantages:

High operational and maintenance costs.

Requires significant initial investment in infrastructure and expertise.

Example: A banks’ data centers , AAU and etc

2. Colocation Data Centers

These data centers are owned and operated by a third-party service provider, offering
space, power, cooling, and networking infrastructure to multiple tenants.

Purpose: Allows businesses to rent physical space and leverage a provider’s

infrastructure rather than building their own.

Location: Located in strategic areas to optimize network connectivity and

disaster recovery.
 Advantages:

Lower upfront costs compared to building a private data center.

Shared operational expenses reduce individual tenant costs.

Scalability to adjust resources as business grows.

Disadvantages:

Less control over the physical infrastructure.

Dependency on the colocation provider for uptime and reliability.

Example: A startup renting server racks in a colocation facility to host its web
application.

3. Cloud Data Centers

These data centers support cloud computing services, with resources delivered over the
internet by cloud service providers.

Purpose: Focused on providing scalable, on-demand computing resources,

including storage, processing power, and applications.

Location: Often distributed globally to provide redundancy and minimize latency

for users in different regions.

Advantages:

Cost-effective pay-as-you-go model.

High scalability and elasticity for fluctuating workloads.

Minimal upfront investment and maintenance.

Disadvantages:

Limited control over data and infrastructure.

Potential security concerns due to data being hosted off-premises.

Example: AWS, Microsoft Azure, or Google Cloud

4. Edge Data Centers

These are small-scale data centers located closer to end users and devices to minimize
latency and improve performance.

Purpose: Designed to process data locally and support latency-sensitive

applications like IoT, autonomous vehicles, and real-time analytics.

Location: Deployed in geographically distributed locations near population

centers or key access points.

Advantages:

Reduces latency by processing data locally before sending it to a central

data center.

Enhances the performance of real-time applications.

Supports the growth of 5G and IoT ecosystems.

Disadvantages:

Limited capacity and scalability compared to larger data centers.

Potentially higher deployment costs for widespread distribution.

Example: A telecom company using edge data centers to power 5G networks for
urban areas.

Again the classification of cloud data centers based on deployment models

The are 4 types of cloud cloud data center deployment models :

1. Private Cloud

A private cloud is a data center infrastructure exclusively used by a single organization,

either hosted on-premises or by a third-party provider.

2. Public Cloud
A public cloud is a shared infrastructure where resources like computing power, storage,
and applications are offered over the internet by a third-party provider.
3. Hybrid Cloud

A hybrid cloud combines private and public cloud environments, allowing data and
applications to move seamlessly between the two

4. Multi-Cloud

A multi-cloud environment uses multiple public or private cloud providers to meet

specific business needs.

Main Purpose is: To avoid reliance on a single vendor and ensure redundancy,
flexibility, and specialized services.

1.2. Main components of Data center

a. Compute Systems

What It Is?:

The servers and processors that perform computing tasks.

Example: Blade servers, rack servers

b. Storage Systems

What It Is?

Systems used to store data securely and access it efficiently.

Example:

Direct Attached Storage (DAS), Network Attached Storage (NAS), and

Storage Area Networks (SAN).

Cloud-based storage solutions like AWS S3 or Google Cloud Storage.

Solutions:

Backup and Archival Solutions: Systems like Veeam or Commvault for

data retention.

Disaster Recovery Solutions: Mitigate risks by ensuring data backups

and failover systems are in place to protect against data loss.
c. Network Systems

What It Is?

Infrastructure to connect servers, storage, and external systems.

Example: Switches, routers, and firewalls.

d. Power and Cooling Systems

What It Is?

Ensures continuous operations by providing reliable power and controlling heat.

Example: Electric power , generators,UPS systems ,air conditioning units, and

hot/cold aisle designs.

e. Security Systems

What It Is?

Protects the data center from physical and cyber threats.

Example: CCTV cameras, biometric access systems, and cybersecurity tools.

f. Management and Monitoring Systems

What It Is?

Tools to oversee and optimize data center operations.

Example:

Data Center Infrastructure Management (DCIM) software.

Tools like SolarWinds or Nagios for network and server monitoring.

Neticos and etc

13. Data Center systems and solutions

The solution Data center came with are many things but I elaborate some it as below:

Virtualization: This technology allows multiple virtual instances (e.g., virtual

machines, virtual storage) to run on a single physical resource, maximizing
hardware utilization and reducing costs by improving resource efficiency.

Disaster Recovery Solutions: These solutions involve creating backup systems

and failover mechanisms to ensure data is securely backed up and systems can
quickly recover during failures, ensuring minimal downtime and business
continuity.

Monitoring Tools: These tools track real-time performance and health of data
center systems, including hardware, software, environmental conditions, and
security, allowing for proactive issue detection and performance optimization.

Energy Efficiency Systems: These systems aim to reduce the environmental

impact of data centers by optimizing energy usage, such as through
energy-efficient hardware, cooling systems, and utilizing renewable energy
sources

1.4. Data Center Standards

TIA-942: A set of telecommunications infrastructure guidelines for data centers.

It defines requirements for network cabling, power, cooling, and other aspects to
ensure reliable and high-performance operations.

Uptime Institute Standards (Tier Classification): A system that classifies data

centers into four tiers (Tier I to Tier IV) based on their reliability, uptime, and
infrastructure. Higher-tier data centers offer greater levels of redundancy, fault
tolerance, and operational resilience.

1.5 Hot and Cold Aisle Configurations in Data Centers

Cold Aisles: Where servers intake cool air from air conditioning systems.

Hot Aisles: Where hot air is expelled, preventing mixing and optimizing cooling
efficiency.
 1.6 Data Center Tier Levels and Classifications

Data centers are classified into four tiers by the Uptime Institute based on uptime and
redundancy:

❖ Tier I:
➢ Uptime: 99.671% (Approx. 28.8 hours of downtime per year)
➢ Fault Tolerance: Minimal redundancy, with only a single path for power
and cooling. Suitable for non-critical operations where occasional
downtime is acceptable.
❖ Tier II:
➢ Uptime: 99.749% (Approx. 22 hours of downtime per year)
➢ Fault Tolerance: Partial redundancy. Some critical components have
backups, but there’s still a single path for power and cooling, which can
lead to downtime during maintenance or failure.
❖ Tier III:
➢ Uptime: 99.982% (Approx. 1.6 hours of downtime per year)
➢ Fault Tolerance: High availability with multiple power and cooling paths. It
allows for maintenance without impacting operations, offering greater
redundancy and ensuring more reliable service.
❖ Tier IV:
➢ Uptime: 99.995% (Approx. 0.4 hours of downtime per year)
➢ Fault Tolerance: Fully fault-tolerant systems with multiple redundant
paths for power, cooling, and network, ensuring maximum uptime with no
single point of failure. Provides the highest level of reliability and
availability.

1.7 Data Center Design

DC-design involves planning and implementing the physical infrastructure, systems,

and processes to ensure a data center operates efficiently, reliably, and securely. The
design focuses on optimizing space, ensuring scalability, and meeting both performance
and operational goals. Here are the key aspects to consider:

Key Considerations in Data Center Design:

★ Scalability:
○ Data centers should be designed to scale as business needs grow. This
includes adding servers, storage, and networking components without
requiring significant redesign or disruption to services.
○ Modular designs allow for incremental expansion, ensuring that
additional capacity can be integrated smoothly over time.
★ Redundancy:
○ Redundancy ensures that the data center can maintain operations in
case of equipment failure or maintenance. This includes:

Power Redundancy: Backup power sources like Uninterruptible

Power Supplies (UPS) and generators ensure the data center
remains operational during power outages.

Cooling Redundancy: Backup HVAC systems or cooling units help

prevent overheating during failures of primary cooling systems.

Data Redundancy: Storing copies of critical data across different

locations or using technologies like RAID (Redundant Array of
Independent Disks) ensures no data is lost during system failures.

★ Energy Efficiency:
○ Data centers are large consumers of electricity. Efficient design can
reduce power consumption and operational costs. Strategies include:

Energy-efficient hardware: Using servers, storage, and

networking equipment designed to consume less power.

Optimized cooling systems: Utilizing methods like free cooling,

where outside air is used for cooling during certain conditions, or
liquid cooling technologies.

Green energy sources: Incorporating renewable energy sources

such as solar or wind power to lower the environmental footprint.

Monitoring and management: Systems to track energy

consumption and optimize usage.

★ Security:
○ Physical Security: Protecting the data center from unauthorized physical
access is critical. This can include:

Restricting access with biometric scanners, keycard systems,

security guards, and surveillance cameras.

Maintaining secure perimeters and implementing disaster

preparedness protocols.
 ○ Digital Security: Safeguarding data and networks from cyber threats is
equally important. This includes:

Firewalls and encryption: Protecting internal systems and

communication channels.

Intrusion detection systems (IDS) and intrusion prevention

systems (IPS) to detect and respond to cyber threats in real-time.

Access control: Ensuring that only authorized users and devices

can access sensitive data.

2. AAiT Data Center Infrastructure -AAiT Perspective

The AAiT Data Center is a state-of-the-art facility designed to meet the diverse and
growing IT demands of AAiT and Addis Ababa University (AAU). It connects 14
campuses via a secure Virtual Private Network (VPN), ensuring seamless
communication and data sharing. The infrastructure’s hybrid design combines
traditional and modular architectures, providing:

Scalability: Modular setups enable cost-effective and efficient future expansion.

Resilience: Traditional systems deliver high reliability and adherence to industry

standards.

Flexibility: Supports various IT workloads, including data storage, processing,

and transmission, with the capacity for growth.

2.1. AAiT Datacenter Type

The AAiT Data Center is a hybrid facility that optimally blends:

Traditional Architecture: Ensures stable, time-tested operations for critical

systems.

Modular Architecture: Allows flexibility for upgrades, maintenance, and

adaptation to evolving technological needs with minimal disruption.

The hybrid nature makes the data center a centralized, highly secure, and efficient
hub, serving the dynamic needs of both AAiT and AAU campuses.
2.2. Main Components

1. Power House

The power infrastructure is the backbone of the data center, ensuring uninterrupted
service and operational reliability:

Primary Power Source: Supplied by Ethiopian Electric Power (ELPA), with

provisions for:

Voltage Stabilization: Through an Automatic Voltage Regulator (AVR).

Automatic Transfer Switch (ATS): Swiftly switches power sources

between ELPA, backup generators, and the UPS within 40–45 seconds.

Uninterruptible Power Supply (UPS):

A Huawei UPS 5000E, featuring 12 cascaded batteries with a total

capacity of 32*12 volts, providing up to 10 hours of uninterrupted
operation.

Includes:

A management screen displaying battery and system performance.

Remote diagnostics via static IP for proactive monitoring.

Load support for various power types (apparent, reactive, active).

➔ Redundancy: Duplication of critical devices ensures high availability and fault

tolerance.
➔ Cooling Systems:

Advanced air conditioning (AC) systems manage equipment heat

efficiently.

Vacuum-sealed environments ensure energy optimization and prevent

external contamination.

➔ Security:

Dual access control systems (e.g., biometric scanners, RFID).

 Fire suppression systems (e.g., fire extinguishers and smoke detectors).

Strict access protocols to prevent unauthorized entry.

2. Containment/Modular Data Center

The containment section focuses on modular setups for enhanced efficiency and
security:

Fiber Optics Infrastructure:

Ensures high-speed data transfer between departments and campuses

with minimal latency.

Supports large-scale data transmission for academic and administrative

tasks.

Firewall Systems:

Acts as the primary defense mechanism, filtering incoming and outgoing

traffic to mitigate cyber threats and enhance network security.

Noida Equipment: Critical for networking and routing tasks, ensuring robust
interconnectivity.

Cooling Mechanisms:

Supplementary fans provide additional cooling to modular units.

Continuous cooling is essential to maintain optimal performance during

heavy workloads.
 Access Control:

A door status indicator system uses three lights for clear access visibility:

Green: Authorized access granted.

Red: Unauthorized access attempt.

Orange: Door left open, prompting immediate action.

1. Monitoring and Management:

a. Centralized monitoring systems ensure real-time oversight of all critical
data center operations, including power usage, temperature, and network
traffic.
b. AI-driven analytics for proactive issue detection and resolution.
2. Disaster Recovery and Business Continuity:
a. Robust backup systems ensure critical data is saved and recoverable in
case of failures.
b. Off-site backup replication adds an additional layer of protection for
long-term storage.
3. Energy Efficiency:
a. Use of energy-efficient systems and renewable power integration (e.g.,
solar panels) to reduce operational costs and carbon footprint.
4. Compliance and Standards:
a. Adherence to global data center standards (e.g., Uptime Institute’s Tier
Standards) ensures high performance and reliability.
b. Regular audits and upgrades maintain compliance with industry best
practices.
3. Management Room

➔ Centralized Monitoring: Admins use a desktop to monitor live security camera

feeds, ensuring real-time visibility into the data center's operations.
➔ Control Systems: Devices are mapped using AutoCAD for detailed
management of all systems, enabling precise troubleshooting and control.
➔ Session Logs: Tracks and records all administrative actions and user activities,
ensuring transparency and accountability.
➔ Remote Access Prohibition: Restricts admin portal access to on-site use only,
preventing unauthorized remote access for enhanced security.
➔ Diagnostics: A specialized tool (Fluke machine) is used to detect wiring and
connectivity problems, facilitating early maintenance.

2.3. AAiT Data Center Systems and Solutions

 1. Power Continuity:

a. Automated systems like ATS and UPS ensure the data center stays
operational during power transitions, avoiding downtime.
b. Redundant components enhance reliability.
2. Data Security:
a. Multi-layered protection includes firewalls, VPN for secure inter-campus
connections, and strict physical access controls.
3. Cooling Solutions:
a. Efficient air conditioning and vacuum-sealed environments reduce energy
use while maintaining optimal equipment performance.
4. Proactive Monitoring:
a. Tools like Fluke machines and AutoCAD support real-time management
and enable quick resolution of issues.
5. Interconnectivity:
a. A secure VPN links AAU’s 14 campuses for seamless resource sharing
and communication.

2.4. AAiT Data Center Standards

1. Power and Cooling Standards

➔ Reliability: Implements robust systems to minimize downtime and ensure

uninterrupted operation.
➔ Energy Efficiency: Uses advanced cooling mechanisms like vacuum
environments and hot/cold aisle separation to optimize energy use while
maintaining equipment performance.
➔ Redundancy: Adopts N+1 configurations, ensuring backup systems are always
available for power and cooling.

2. Access and Security Standards

➔ Physical Security: Multi-level access controls, such as biometric systems and

surveillance cameras, protect the data center from unauthorized access.
➔ Digital Security: Firewalls and secure Virtual Private Networks (VPNs)
safeguard data and inter-campus communication.
➔ Constant Monitoring: Surveillance and logging systems ensure comprehensive
oversight of both physical and virtual environments.

3. Redundancy Standards

➔ Ensures all critical systems (power, cooling, and networking) are designed with
failover mechanisms, allowing the center to continue operating seamlessly
during component failures or maintenance.

4. Compliance with Industry Standards

➔ TIA Standards (Telecommunications Industry Association): Ensures structured

cabling, power, cooling, and design elements meet international benchmarks.
➔ Device Standards: Uses high-quality equipment from trusted manufacturers like
Cisco and Huawei for networking, power management, and system monitoring.

5. Tier II Classification

➔ 99.982% Uptime: Guarantees high operational availability throughout the year.

➔ Concurrent Maintenance: Systems are designed to allow maintenance or
replacement without disrupting ongoing operations.
 6. Cable Management

➔ Implements a 40 cm raised floor for organized cabling, preventing clutter and

simplifying maintenance tasks.
➔

2.5. AAiT Hot and Cold Aisle Data Center System

The Hot and Cold Aisle system is a strategic design to optimize cooling efficiency and
energy consumption in the data center.

Hot Aisle

Purpose: Collects the heated air emitted from server exhausts and directs it
away from sensitive equipment.

Location: Typically enclosed or contained to isolate hot air from mixing with the
cooler environment.

Benefit: Prevents heated air from recirculating, ensuring it doesn’t affect

equipment performance.

Cold Aisle

Purpose: Focuses on delivering cool air from air conditioning systems directly to
the server intakes.

Alignment: Server racks are arranged so that the fronts (intakes) face the cold
aisle.

Benefit: Maintains optimal operating temperatures for servers

2.6. AAiT Data Center Tier Classification

Tier II Standards: The center is classified as a Tier II data center, offering:

○ 99.982% Uptime: Ensures availability nearly all year.

○ Redundancy: Power, cooling, and network systems have failover
mechanisms to handle failures.
○ Concurrent Maintenance: Systems can be maintained or replaced
without interrupting operations.

2.7. AAiT Data Center Design

Scalable Layouts: Modular sections allow the data center to expand as IT needs
grow.

Energy Efficiency:

Includes vacuum-sealed environments and hot/cold aisle systems to

reduce energy consumption.
 Cable Management:

A 40 cm raised floor organizes cables, avoiding clutter and making

maintenance easier.

Physical Security:

Dual-layer access controls, surveillance, and security personnel ensure

safety for sensitive equipment.

User-Friendly Interface:

AutoCAD mapping simplifies administration and diagnostics for system

management.

Network Design Hierarchy:

Starts with Ethio Telecom for connectivity, then data flows through:

■ Firewall → Core Switch → Distributed Switch → End Users.

3. Comparison of International Data Center and AAiT

Purpose:

The AAiT Data Center supports the academic and research needs of Addis
Ababa Institute of Technology (AAiT).

The National Data Center provides IT infrastructure for the entire nation, serving
government institutions, businesses, and public services.

Scale and Capacity:

The AAiT Data Center is small to medium-sized, suitable for academic purposes
with limited resources.

The National Data Center is a large-scale facility with significant storage and
computing power to support nationwide operations.

Scalability:
The AAiT Data Center has limited scalability due to budget and space
constraints.

The National Data Center is designed for scalability to handle growing national
demands and future expansions.

Redundancy:

The AAiT Data Center has minimal redundancy, offering basic backup systems
for power and cooling.

The National Data Center features advanced redundancy, including multiple

power and cooling systems, ensuring high availability.

Energy Efficiency:

The AAiT Data Center relies on basic cooling and power systems, which may not
be energy-efficient.

The National Data Center employs advanced energy-saving technologies and

optimized cooling systems.

Security:

The AAiT Data Center has basic physical and digital security measures,
adequate for academic needs.

The National Data Center offers robust physical and digital security, including
biometric access, firewalls, and compliance with global standards like ISO/IEC
27001.

Connectivity:

The AAiT Data Center primarily serves local users within the AAiT community.

The National Data Center provides high-speed connectivity and supports

nationwide IT infrastructure and services.

Funding:
 The AAiT Data Center operates within limited budgets provided by AAiT or
academic grants.

The National Data Center benefits from government funding and international
partnerships, enabling advanced infrastructure investment.

Services Provided:

The AAiT Data Center focuses on hosting academic projects, research

computing, and storage for AAiT users.

4. Summary of Data center visit

The AAiT Data Center is a highly secure, efficient, and reliable facility with a focus on
scalability and energy efficiency. Its advanced cooling, security, and management
systems ensure smooth operations, while compliance with Tier II standards guarantees
high uptime and fault tolerance.

The design incorporates best practices for data centers, ensuring future readiness for
technological advancements.

Reference Books

Data Center Handbook - Comprehensive guide on data center design and

management.

Cloud Computing: Concepts, Technology & Architecture - Basics of

cloud-enabled data centers.

Handbook of Cloud Computing - Insights into cloud and data center

integration.