Cloud Networking

Understanding Cloud-based Data Center Networks

Gary Lee

• Publisher: Morgan Kaufmann
• Year: 2014
• ISBN:
  9780128008164
• Topic: Informatik
• Subtopic: Computernetzwerke

Cloud Networking: Understanding Cloud-Based Data Center Networks explains the evolution of established networking technologies into distributed, cloud-based networks. Starting with an overview of cloud technologies, the book explains how cloud data center networks leverage distributed systems for network virtualization, storage networking, and software-defined networking. The author offers insider perspective to key components that make a cloud network possible such as switch fabric technology and data center networking standards. The final chapters look ahead to developments in architectures, fabric technology, interconnections, and more. By the end of the book, readers will understand core networking technologies and how they're used in a cloud data center.

• Understand existing and emerging networking technologies that combine to form cloud data center networks
• Explains the evolution of data centers from enterprise to private and public cloud networks
• Reviews network virtualization standards for multi-tenant data center environments
• Includes cutting-edge detail on the latest switch fabric technologies from the networking team in Intel

Chapter 1

Welcome to Cloud Networking

Abstract

The first chapter of the book covers general information regarding cloud networking. In this chapter, we set the stage for the rest of the book by providing some basic networking background for readers who are new to the subject, along with providing an overview of cloud computing and cloud networks. This background information should help the reader better understand some of the topics that are covered later in the book.

Keywords

Network stack

OSI

Ethernet

TCP/IP

Packets

L2 header

L3 header

NIC

LAN

Cat5

Cat6

10GBase-T

1GbE

10GbE

Virtualization

Convergence

OCP

SDN

Welcome to a book that focuses on cloud networking. Whether you realize it or not, the “Cloud” has a significant impact on your daily life. Every time you check someone’s status on Facebook®, buy something on Amazon®, or get directions from Google® Maps, you are accessing computer resources within a large cloud data center. These computers are known as servers, and they must be interconnected to each other as well as to you through the carrier network in order for you to access this information. Behind the scenes, a single click on your part may spawn hundreds of transactions between servers within the data center. All of these transactions must occur over efficient, cost effective networks that help power these data centers.

This book will focus on networking within the data center and not the carrier networks that deliver the information to and from the data center and your device. The subject matter focuses on network equipment, software, and standards used to create networks within large cloud data centers. It is intended for individuals who would like to gain a better understanding of how these large data center networks operate. It is not intended as a textbook on networking and you will not find deep protocol details, equations, or performance analysis. Instead, we hope you find this an easy-to-read overview of how cloud data center networks are constructed and how they operate.

Introduction

Around the world, new cloud data centers have been deployed or are under construction that can contain tens of thousands and in some cases hundreds of thousands of servers. These are sometimes called hyper-scale data centers. You can think of a server as something similar to a desktop computer minus the graphics and keyboard but with a beefed up processor and network connection. Its purpose is to “serve” information to client devices such as your laptop, tablet, or smart phone. In many cases, a single web site click on a client device can initiate a significant amount of traffic between servers within the data center. Efficient communication between all of these servers, and associated storage within the cloud data center, relies on advanced data center networking technology.

In this chapter, we will set the stage for the rest of this book by providing some basic networking background for those of you who are new to the subject, along with providing an overview of cloud computing and cloud networking. This background information should help you better understand some of the topics that are covered later in this book. At the end of this chapter, we will describe some of the key characteristics of a cloud data center network that form the basis for many of the chapters in this book.

Networking basics

This book is not meant to provide a deep understanding of network protocols and standards, but instead provides a thorough overview of the technology inside of cloud data center networks. In order to better understand some of the subject presented in this book, it is good to go over some basic networking principals. If you are familiar with networking basics, you may want to skip this section.

The network stack

Almost every textbook on networking includes information on the seven-layer Open Systems Interconnect (OSI) networking stack. This model was originally developed in the 1970s as part of the OSI project that had a goal of providing a common network standard with multivendor interoperability. OSI never gained acceptance and instead Transmission Control Protocol/Internet Protocol (TCP/IP) became the dominant internet communication standard but the OSI stack lives on in many technical papers and textbooks today.

Although the networking industry still refers to the OSI model, most of the protocols in use today use fewer than seven layers. In data center networks, we refer to Ethernet as a layer 2 protocol even though it contains layer 1 and layer 2 components. We also generally refer to TCP/IP as a layer 3 protocol even though it has layer 3 and layer 4 components. Layers 5-7 are generally referred to in the industry as application layers. In this book, we will refer to layer 2 as switching (i.e., Ethernet) and layer 3 as routing (i.e., TCP/IP). Anything above that, we will refer to as the application layer. Figure 1.1 shows an example of this simplified model including a simple data center transaction.

Figure 1.1 Example of a simple data center transaction.

In this simplified example, the sender application program presents data to the TCP/IP layer (sometimes simply referred to as layer 3). The data is segmented into frames (packets) and a TCP/IP header is added to each frame before presenting the frames to the Ethernet layer (sometimes simply referred to as layer 2). Next, an Ethernet header is added and the data frames are transmitted to the receiving device. On the receive side, the Ethernet layer removes the Ethernet header and then the TCP/IP layer removes the TCP/IP header before the received frames are reassembled into data that is presented to the application layer. This is a very simplified explanation, but it gives you some background when we provide more details about layer 2 and layer 3 protocols later in this book.

As an analogy, think about sending a package from your corporate mail room. You act as the application layer and tell your mail room that the gizmo you are holding in your hand must be shipped to a given mail station within your corporation that happens to be in another city. The mail room acts as layer 3 by placing the gizmo in a box, looking up and attaching an address based on the destination mail station number, and then presenting the package to the shipping company. Once the shipping company has the package, it may look up the destination address and then add its own special bar code label (layer 2) to get it to the destination distribution center. While in transit, the shipping company only looks at this layer 2 label. At the destination distribution center, the local address (layer 3) is inspected again to determine the final destination. This layered approach simplifies the task of the layer 2 shipping company.

Packets and frames

Almost all cloud data center networks transport data using variable length frames which are also referred to as packets. We will use both terms in this book. Large data files are segmented into frames before being sent through the network. An example frame format is shown in Figure 1.2.

Figure 1.2 Example frame format.

The data is first encapsulated using a layer 3 header such as TCP/IP and then encapsulated using a layer 2 header such as Ethernet as described as part of the example in the last section. The headers typically contain source and destination address information along with other information such as frame type, frame priority, etc. In many cases, checksums are used at the end of the frame to verify data integrity of the entire frame. The payload size of the data being transported and the frame size depend on the protocol. Standard Ethernet frames range in size from 64 to 1522 bytes. In some cases jumbo frames are also supported with frame sizes over 16K bytes.

Network equipment

Various types of network equipment can be used in cloud data centers. Servers contain network interface cards (NICs) which are used to provide the server CPU(s) with external Ethernet ports. These NICs are used to connect the servers to switches in the network through data cables. The term switch is generally used for equipment that forwards data using layer 2 header information. Sometimes, an Ethernet switch may also be referred to as an Ethernet bridge and the two terms can be used interchangeably. The term router is generally used for equipment that forwards data using layer 3 header information. Both switches and routers may be used within large cloud data center networks, and, in some cases, Ethernet switches can also support layer 3 routing.

Interconnect

In the data center, servers are connected to each other, connected to storage, and connected to the outside network through switches and routers. These connections are made using either copper or optical cabling. Historically, copper cabling has been a lower-cost solution, while optical cabling has been used when higher bandwidth and/or longer cabling distances are required. For example, shorter, copper cabling may be used as a connection between the servers and switches within a rack, and high bandwidth optical cabling may be used for uplinks out of the rack in order to span longer distances. We will provide more information on cable types later in this chapter.

What is a cloud data center?

In the early days of the world wide web (remember that term?) data was most likely delivered to your home computer from a room full of servers in some sort of corporate data center. Then, the internet exploded. The number of people accessing the web grew exponentially as did the number of web sites available as well as the average data download sizes. Popular web service companies such as Google and Amazon needed to rapidly expand their data centers to keep up with demand. It quickly got to the point where they needed to erect large dedicated server warehouses that are today known as cloud data centers.

The term “cloud” started emerging around the same time wireless handheld devices started to become popular in the marketplace. When accessing the web via a wireless handheld device, it seems like you are pulling data out of the clouds. It is natural, then, that the data centers providing this information should be called cloud data centers. Today, it appears that everyone is jumping on the “cloud” bandwagon with all kinds of cloud companies, cloud products, and cloud services entering the market.

Cloud data centers are being rapidly deployed around the world. Since these installations must support up to hundreds of thousands of servers, data center efficiency and cost of operations have become critical. Because of this, some cloud data centers have been erected near cheap electrical power sources, such as hydroelectric dams, or in colder climates to help reduce cooling costs. Some companies, such as Microsoft®, are building modular data centers using pods, which are self-contained server storage and networking modules the size of a shipping container. These modules are trucked in, stacked up, and connected to power, cooling, and networking. Other data centers use server racks as the basic building block and contain rows and rows of these racks. No matter what the structure, networking is an important part of these large cloud data center networks.

A recent Cisco® white paper entitled Cisco Global Cloud Index: Forecast and Methodology, 2012–2017 provides some interesting insights into cloud data centers. They predict that global IP data center traffic will grow by 25% each year at least through...