In the modern enterprise landscape, ensuring robust and efficient connectivity between on-premises infrastructure and cloud environments is a critical challenge. The demand for high bandwidth, low latency, and fault tolerance in data traffic flows is ever-increasing, especially with the adoption of multi-cloud strategies. Traditional network architectures often struggle to meet these requirements, leading to performance bottlenecks and increased vulnerability to service disruptions. This thesis addresses these challenges by proposing a flexible, high-performance network architecture designed to support enterprise connectivity in multi-cloud environments. The solution used cloud exchange provider services which provides advanced network technologies and connectivity through multiple layers. The architecture implements Layer 1 physical fiber connections and redundant ports, Layer 2 connectivity through Ethernet Virtual Private Line (EVPL) transport technology with VLAN 802.1q encapsulation, and Layer 3 services utilizing BGP routing protocols and virtual network functions. This comprehensive approach enables direct, private connections between on premise and cloud service providers (CSPs), eliminating the limitations of traditional VPN-based solutions. The cloud exchange provider's infrastructure operates as an Ethernet network, enabling efficient traffic management and simplified connectivity between multiple endpoints. Through detailed implementation and testing, this research demonstrates significant improvements in network performance, including reduced latency, enhanced fault tolerance through automated failover mechanisms, and simplified management of complex multi-cloud environments. The results validate that enterprises can optimize their cloud connectivity strategy while ensuring reliable, secure, and high-performance communication between their on-premises infrastructure and multiple cloud providers.

In the modern enterprise landscape, ensuring robust and efficient connectivity between on-premises infrastructure and cloud environments is a critical challenge. The demand for high bandwidth, low latency, and fault tolerance in data traffic flows is ever-increasing, especially with the adoption of multi-cloud strategies. Traditional network architectures often struggle to meet these requirements, leading to performance bottlenecks and increased vulnerability to service disruptions. This thesis addresses these challenges by proposing a flexible, high-performance network architecture designed to support enterprise connectivity in multi-cloud environments. The solution used cloud exchange provider services which provides advanced network technologies and connectivity through multiple layers. The architecture implements Layer 1 physical fiber connections and redundant ports, Layer 2 connectivity through Ethernet Virtual Private Line (EVPL) transport technology with VLAN 802.1q encapsulation, and Layer 3 services utilizing BGP routing protocols and virtual network functions. This comprehensive approach enables direct, private connections between on premise and cloud service providers (CSPs), eliminating the limitations of traditional VPN-based solutions. The cloud exchange provider's infrastructure operates as an Ethernet network, enabling efficient traffic management and simplified connectivity between multiple endpoints. Through detailed implementation and testing, this research demonstrates significant improvements in network performance, including reduced latency, enhanced fault tolerance through automated failover mechanisms, and simplified management of complex multi-cloud environments. The results validate that enterprises can optimize their cloud connectivity strategy while ensuring reliable, secure, and high-performance communication between their on-premises infrastructure and multiple cloud providers.

Design and Implementation of Multi-cloud Network Architecture for High-Performance Connection

AGABA, BOUCHRA
2024/2025

Abstract

In the modern enterprise landscape, ensuring robust and efficient connectivity between on-premises infrastructure and cloud environments is a critical challenge. The demand for high bandwidth, low latency, and fault tolerance in data traffic flows is ever-increasing, especially with the adoption of multi-cloud strategies. Traditional network architectures often struggle to meet these requirements, leading to performance bottlenecks and increased vulnerability to service disruptions. This thesis addresses these challenges by proposing a flexible, high-performance network architecture designed to support enterprise connectivity in multi-cloud environments. The solution used cloud exchange provider services which provides advanced network technologies and connectivity through multiple layers. The architecture implements Layer 1 physical fiber connections and redundant ports, Layer 2 connectivity through Ethernet Virtual Private Line (EVPL) transport technology with VLAN 802.1q encapsulation, and Layer 3 services utilizing BGP routing protocols and virtual network functions. This comprehensive approach enables direct, private connections between on premise and cloud service providers (CSPs), eliminating the limitations of traditional VPN-based solutions. The cloud exchange provider's infrastructure operates as an Ethernet network, enabling efficient traffic management and simplified connectivity between multiple endpoints. Through detailed implementation and testing, this research demonstrates significant improvements in network performance, including reduced latency, enhanced fault tolerance through automated failover mechanisms, and simplified management of complex multi-cloud environments. The results validate that enterprises can optimize their cloud connectivity strategy while ensuring reliable, secure, and high-performance communication between their on-premises infrastructure and multiple cloud providers.
2024
Design and Implementation of Multi-cloud Network Architecture for High-Performance Connection
In the modern enterprise landscape, ensuring robust and efficient connectivity between on-premises infrastructure and cloud environments is a critical challenge. The demand for high bandwidth, low latency, and fault tolerance in data traffic flows is ever-increasing, especially with the adoption of multi-cloud strategies. Traditional network architectures often struggle to meet these requirements, leading to performance bottlenecks and increased vulnerability to service disruptions. This thesis addresses these challenges by proposing a flexible, high-performance network architecture designed to support enterprise connectivity in multi-cloud environments. The solution used cloud exchange provider services which provides advanced network technologies and connectivity through multiple layers. The architecture implements Layer 1 physical fiber connections and redundant ports, Layer 2 connectivity through Ethernet Virtual Private Line (EVPL) transport technology with VLAN 802.1q encapsulation, and Layer 3 services utilizing BGP routing protocols and virtual network functions. This comprehensive approach enables direct, private connections between on premise and cloud service providers (CSPs), eliminating the limitations of traditional VPN-based solutions. The cloud exchange provider's infrastructure operates as an Ethernet network, enabling efficient traffic management and simplified connectivity between multiple endpoints. Through detailed implementation and testing, this research demonstrates significant improvements in network performance, including reduced latency, enhanced fault tolerance through automated failover mechanisms, and simplified management of complex multi-cloud environments. The results validate that enterprises can optimize their cloud connectivity strategy while ensuring reliable, secure, and high-performance communication between their on-premises infrastructure and multiple cloud providers.
Cloud Exchange
IPSEC
point of presence
network edge
multi cloud
File in questo prodotto:
File Dimensione Formato  
Bouchra_Agaba.pdf

Accesso riservato

Dimensione 5.21 MB
Formato Adobe PDF
5.21 MB Adobe PDF

The text of this website © Università degli studi di Padova. Full Text are published under a non-exclusive license. Metadata are under a CC0 License

Utilizza questo identificativo per citare o creare un link a questo documento: https://hdl.handle.net/20.500.12608/87074