DEVELOPMENT OF MULTILAYER AGGREGATE SWITCH ARCHITECTURE FOR DATA CENTER NETWORK USING SOFTWARE DEFINED NETWORK
MSEE Project Proposal
By
Abhash Malviya
Ashmita Chakraborty
Janhavi Tejomurtula
Abhilash Naredla
Signature Date
Academic Advisor:
Prof. Nadir F Mir ________________ __________________
Graduate Advisor:
Prof. Thuy Le ________________ __________________
Department of Electrical Engineering
San Jose State University
San Jose, CA
______________________________________________________
DEVELOPMENT OF MULTILAYER AGGREGATE SWITCH ARCHITECTURE FOR DATA CENTER NETWORK USING SOFTWARE DEFINED NETWORK
MSEE Project Proposal
By
Abhash Malviya
Ashmita Chakraborty
Janhavi Tejomurtula
Abhilash Naredla
Project Advisor
Prof Nadir F Mir
Department of Electrical Engineering
San Jose State University
San Jose, CA ABSTRACT
Applications are the engine for modern businesses driving innovation, operational efficiency and revenue generation. They demand an infrastructure that is highly agile and is easy to manage, while reducing overall costs. Typically residing in the data centers, these applications, which include mission critical systems, multitier Web applications, Big Data, etc. has placed new constraints on the networking infrastructure. Additionally, many such applications often consist of both physical and virtualize workloads, adding to
Brocade is fully invested in making SDN completely modular to deliver New IP networks based on an open ecosystem. The Brocade SDN Controller is continuously built from the OpenDaylight code, with a virtuous cycle of contributions back to the OpenDaylight community. Physical and virtual networks from multiple vendors can be managed within the Brocade SDN Controller, and organizations can gradually introduce ever-larger portions of their existing networks into the controller domain with single-source technical support for the entire
Commercial data centers like Equinix and Amazon have to be judicious expanding capacity considering lengthy construction and heavy capital investment. A data center contains aisles and aisles of server cabinets thousands with each costing more than a quarter million containing 64 specialized laptop sized servers essentially a computer. Networking cabinets go together with server cabinets in order to electronically connect any server to another through high-speed fiber cables. The networking cabinets of today consume 25% of the useful floor space for IT cabinets but consist of 40% of the total cost. The data center is clamoring for IT gear that supports a new technology known as Network virtualization to address the conservation of space and
During your time with us you are going to meet an exciting team of leading engineers and business professionals who are as passionate about computer networking as you are. ISP/Internet, Networking, SDN and Servers/Cloud are in our DNA and in yours as well. We are a global leader in the data center industry and this is the reason the premier cloud providers in the world continue to partner with us.
Cloud computing is an emerging model where users can gain access to their applications from anywhere through their connected devices. A simplified user interface makes the infrastructure supporting the applications transparent to users. The applications reside in massively-scalable data centers where compute resources can be dynamically provisioned and shared to achieve significant economies of scale. A strong service management platform results in near-zero incremental management costs when more IT resources are added to the cloud. The proliferation of smart mobile devices, high speed wireless connectivity, and rich browser-based Web 2.0 interfaces has made the network-based cloud computing model not only practical but
To respond to dynamic workloads and demands, network administrators and operators must be to achieve high utilization of server and network capacity, be able to assign workloads operating in single layer 2 network to any server in any rack in the network, be able to move workloads without re-configuring the network. This is achieved by decoupling the workload’s location from its network address. Therefore, key-design objectives for the modern data centers are:
In this paper, we will conduct a comparison on Fast Ethernet Network Switches and Gigabit Ethernet Network Switches offered by 3COM. We will review a few specific components, as well as compare features, pricing and preferred usage.
The NFV ISG lists various possible scenarios in which NFV may be used in the current network infrastructure to help organizations reduce capital and operational expenditures and increase revenue via VNF services. While in theory, every network element can be virtualized, the ISG has identified specific devices based on feasibility in terms of cost and complexity of implementation.
Because of the diverse network types, there need’s to be a way to interconnect various heterogenous networks. Switching and routing functions are implemented to establish proper loop free links, efficient and to the fact that networks aren’t static but have
In today’s world each sector of industry need the internet services. The way we enjoy the deep services provided by the internet is only possible because of data centers. They play a critical role for enterprises by helping them expanding their capabilities. Incorporating software abstractions with DCN has helped with the evolution of DCN. As the need for cloud based applications increase so does the need of DCN to work more efficiently increases as well. Because of cloud computing the DCN are growing even larger in size and will grow further more in future. DCN contains thousands of servers. Interconnecting all these servers is the challenge the researchers face. They are generally connected via network interface cards, cables, routers and switches. Placement of all these devices
Cloud computing is already at the center of new data center evolution because large amounts of information can be stored in cloud. It is actually estimated that 1/3 of all data will go through cloud by 2020. This will lead to a rise in global cloud revenues as well as the amount of money firms spend on cloud computing and innovation. The development will also see many organizations come up with ways to enhance cloud infrastructure to enhance compute density, ultra-low latency and power density. The most common networking attributes of cloud include; low latency, programmable management, scalable, Open APIs and Self-healing resilience. Cloud networking scaling is achieved using topologies from wide ranging nodes that provide non-resistive fabric
To reduce networks problem, we can use Software Defined Network. Software-Defined Networking (SDN) is an outgoing architecture that is progressive, manageable, cost-effective, and painful, making it ideal for the high-bandwidth, cheaper to build, powerfully faster and more skilled dynamic nature of today 's applications [2]. SDN demands to reduce this effort dramatically [4]. SDN is called the birds-eye perspective on a network. The idea behind it is middle "intelligence" that is able to looking the big picture and is thus able to control and optimistic data streams better and efficiently. Adopting an SDN methodology has a countless of benefits including facility, scalability, thickness, and performance [4]. In addition, a growing network
Abstract—Software-Defined Networking (SDN) is a new intelligent architecture which is composited by several kinds of network technology aimed at making the network as agile and flexible as possible. The SDN Controller is one of the most important components in the SDN architecture. The SDN Controller can help the network architecture to reduce hardware limitations, and allowing network engineers and administrators to respond quickly to changing business requirement. It indicates the perfect relationships between software
Nokia’s CloudBand Network Director: Network Director basically functions as NFV network resources and services orchestrator that is again put together for OpenStack & VMware. As the virtual resources are expected to be scattered all around the world, this entity manages these geographically distributed nodes of NFV infrastructure. Moreover the features of visualization and automation of network
This section contains information about the implementation of the SDN switch core, interconnection of multiple SDN switches with the SDN controller (PowerPC) and the attacker nodes (Microblaze). The main functionality of the SDN switch is to modify packet header fields based on the flow table and forward it to the next port(s). The SDN controller is responsible for programming the flow table in each switch and monitor these switches to observe each packet flow. The Microblaze processor, acting as attacker nodes plays the role of an outside network and transmits packets at different programmable rates to the SDN switch network using an array of packet drivers. The big picture showing the connection between different components is given in Figure 3.1.
We compared the performance of various workloads between SDN (OpenFlow and ProGFE), running in the user space of Linux, and non-SDN (dedicated application that executes the networking tasks directly, e.g., pure Linux forwarding) running in the user space (for comparison), and in the kernel space (to explore the PC potential). The following workloads were examined: IP routing, as a rather simple Forwarding Element task, and VLAN tagging, which represented a more complicated task for an FE. In order to evaluate performance, we ignored the indirect