0% found this document useful (0 votes)
16 views17 pages

Data Center Networking Overview

The document discusses data center networking, highlighting the architecture and functions of data centers used by major internet companies like Google and Amazon. It covers the challenges faced in data center networks, such as reliability and load management, and outlines the hierarchical design and trends in data center technology, including SDN control and virtualization. Additionally, it provides insights into network elements, routing, and the evolving nature of data center networks.

Uploaded by

yyogeshwar768
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
16 views17 pages

Data Center Networking Overview

The document discusses data center networking, highlighting the architecture and functions of data centers used by major internet companies like Google and Amazon. It covers the challenges faced in data center networks, such as reliability and load management, and outlines the hierarchical design and trends in data center technology, including SDN control and virtualization. Additionally, it provides insights into network elements, routing, and the evolving nature of data center networks.

Uploaded by

yyogeshwar768
Copyright
© All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd

DATA COMMUNICATION AND NETWORKING

Data Center Networking

Dr. Veena S
Department of Computer Applications
DATA COMMUNICATION AND NETWORKING

Data Center Networks


▪ Internet companies such as Google, Microsoft, Amazon, and Alibaba
have built massive data centers, each housing tens to hundreds of
thousands of hosts.
▪ Data Center Network is the one which interconnect their internal
hosts.

Inside a 40-ft Microsoft container, Chicago data center


DATA COMMUNICATION AND NETWORKING

Data Center Networks


Data Centers supports:
▪ e-business (e.g. Amazon)
▪ content-servers (e.g., YouTube, Akamai, Apple, Microsoft)
▪ search engines, data mining (e.g., Google)

▪ Broadly speaking, data centers serve three purposes. - ---


- First, they provide content to web pages. (email, video
streaming, web pages)
- Second, they serve as massively-parallel computing
infrastructures for specific Data (index computation for
search engines)
- Third, they provide cloud computing to other
companies.
DATA COMMUNICATION AND NETWORKING

Data Center Networks


challenges:
▪ Multiple applications, each serving massive numbers of
clients
▪ Reliability
▪ Managing/balancing load, avoiding processing, networking,
data bottlenecks
DATA COMMUNICATION AND NETWORKING

Data Center Networks - Architecture


▪ The worker bees in a data center are the hosts. The hosts in
data centers, called Blades.
▪ The hosts are stacked in racks, with each rack typically having
20 to 40 blades.
▪ At the top of each rack, there is a switch, aptly named the
Top of Rack (TOR) switch, that interconnects the hosts in the
rack with each other and with other switches.
▪ The data center network supports two types of traffic: traffic
flowing between external clients and internal hosts and
traffic flowing between internal hosts.
DATA COMMUNICATION AND NETWORKING

Data Center Networks - Architecture


▪ To connect to public internet, it uses Border Routers.
▪ Inside they may have one or more Load balancers. It can also
perform NAT like functionality (for security)
▪ Below that three layers of switches (core, Aggregate and
edge switches).
DATA COMMUNICATION AND NETWORKING
Datacenter networks: network elements

Border routers
▪ connections outside datacenter

Tier-1 switches
▪ connecting to ~16 T-2s below

Tier-2 switches
▪ connecting to ~16 TORs below
… … … …
Top of Rack (TOR) switch
… … … … ▪ one per rack
▪ 100G-400G Ethernet to blades
Server racks
▪ 20- 40 server blades: hosts
DATA COMMUNICATION AND NETWORKING

Data Center Networks


▪ Follow Hierarchical design.
▪ Number of switches depends on size of the data center.
▪ It has redundant links to provide high availability.
▪ To avoid ARP broadcasts, network may be divided into
subnetworks.
▪ Popular Data center design is Clos Network Topology.
DATA COMMUNICATION AND NETWORKING

Data Center Networks – Clos Design

Highly Interconnected Data Center Design


DATA COMMUNICATION AND NETWORKING

Data Center Networks - Trends


▪ It is evolving.
▪ Cost reduction and improve performance.
▪ Centralized SDN Control and Management.
▪ Virtualization.
▪ Physical Constraints - Require extremely Low delay, High
throughput and availability
▪ Hardware Modularity and Customization
DATA COMMUNICATION AND NETWORKING
Unit 5: Summary

▪ Principles behind data link layer services:


▪ Error detection, correction
▪ Sharing a broadcast channel: multiple access
▪ Link layer addressing - ARP
▪ Instantiation and implementation of various link
layer technologies
▪ Ethernet
▪ Switched LANS, VLANs
▪ MPLS
▪ Data Center Networks
DATA COMMUNICATION AND NETWORKING

Additional Information
Datacenter networks: network elements
Facebook F16 data center network topology:

[Link] (posted 3/2019)


Link Layer: 6-13
Datacenter networks: multipath
▪ rich interconnection among switches, racks:
• increased throughput between racks (multiple routing paths possible)
• increased reliability via redundancy

9 10 11 12 13 14 15 16

two disjoint paths highlighted between racks 1 and 11


Link Layer: 6-14
Datacenter networks: application-layer routing
Internet
load balancer:
application-layer
routing
▪ receives external
Load client requests
balancer
▪ directs workload
within data center
… … … … ▪ returns results to
external client
… … …
(hiding data center

internals from client)

Link Layer: 6-15


ORION: Google’s new SDN control plane for internal
datacenter (Jupiter) + wide area (B4) network
▪ routing (intradomain, iBGP), traffic Orion SDN architecture and core apps
engineering: implemented in applications
on top of ORION core
▪ edge-edge flow-based controls (e.g.,
CoFlow scheduling) to meet contract SLAs
▪ management: pub-sub distributed
microservices in Orion core, OpenFlow for
switch signaling/monitoring

Note:
▪ no routing protocols, congestion control (partially) also managed by SDN rather
than by protocol
▪ are protocols dying?
Link Layer: 6-16
THANK YOU

Dr. Veena S, Chairperson


Department of Computer Applications
sveena@[Link]

+91 80 26721983 Extn 829

You might also like