YT YouTube2Text

YouTube2Text → Transcripts → Structuring and Modularizing the Network Part 3

Structuring and Modularizing the Network Part 3 — Transcript

by Santelmo · 6,031 words · 1,337 segments · language en · Watch on YouTube

Open in transcriber Summarize with AI Download SRT/VTT

Full transcript

0:05services
0:06within modular networks
0:11business that operate large enterprise
0:13networks strive to create
0:15an enterprise-wide network
0:17infrastructure and interactive services
0:20to serve as a solid foundation for
0:23business
0:24and collaborative applications this
0:27section explores
0:28some of the interactive services with
0:31respect to
0:32the modules that form the cisco
0:35enterprise architecture
0:37so the key point here is a network
0:39service is a supporting and necessary
0:41service
0:42but not an ultimate solution for example
0:46security and quality of service are not
0:48ultimate goals
0:50for a network they are necessary
0:53to enable other services and
0:54applications
0:56and are therefore classified as network
0:58services
1:00however ip telephony might be an
1:03ultimate goal of a network and is
1:06therefore
1:06a network application or solution rather
1:09than
1:10a service so interactive services
1:14since the inception of a packet-based
1:16communication
1:18networks have always offered a
1:20forwarding service
1:21forwarding is the fundamental activity
1:24within the internetwork
1:26in ip this forwarding service was built
1:29on the assumption
1:31that end nodes in the network were
1:33intelligent
1:35and that the network core did not have
1:39so with advances in networking software
1:41and hardware
1:42the network can offer an increasingly
1:44rich intelligent set of mechanisms
1:47for forwarding information interactive
1:51services
1:52add intelligence to the network
1:54infrastructure
1:55beyond simply moving a datagram between
1:59two points for example
2:02through intelligent network
2:03classification the network distinguishes
2:06and identifies traffic based on
2:08application content
2:10and context advanced network services
2:14use the traffic classification to
2:16regulate performance
2:18ensure security facilitate delivery and
2:22improve manageability network
2:25applications such as
2:26ip telephony support the enter
2:29enterprise network environment
2:31from the teleworker to the campus to
2:34the data center so these applications
2:37are enabled by critical network services
2:40and
2:41provide a common set of capabilities
2:44to support the application's
2:46network-wide requirements
2:48including security high availability
2:52reliability flexibility responsiveness
2:57and compliance
3:05examining the role of the infrastructure
3:07services
3:08recall the layers of the cisco sona
3:10framework illustrated in the future
3:13so the sona interactive services layer
3:15includes
3:16both application networking and
3:19the infrastructure services so for
3:22example
3:23the following infrastructure services
3:25shown earlier in the figure
3:26enhance classic network functions to
3:29support
3:30today's application environment by
3:32mapping the application's requirements
3:35to the resources that they require from
3:37the network
3:39so these are security services
3:42it ensures that all aspects of the
3:44network are secure
3:46from devices connecting to the network
3:49to secure
3:50transport to data theft prevention
3:53you also have mobility services
3:56it allows users to access network
3:59resources regardless of their physical
4:01location
4:03you also have this storage services
4:06okay so which provide distributed and
4:09virtual storage
4:10across the infrastructure you also have
4:14a voice and collaboration services
4:17it deliver the foundation by which voice
4:20can be carried across the network such
4:24as security and high availability
4:28you also have this compute services it
4:31connect and virtualize compute
4:33resources based on the application
4:36and you've got the identity services
4:40which map resources and policies
4:43to the user and the device
4:47now what are the reasons for internal
4:49security
4:51strongly protecting the internal
4:53enterprise campus by including security
4:55functions
4:56in each individual element is important
4:59for the following reasons
5:02so that would be if the security
5:03established at the
5:05enterprise edge fails an unprotected
5:08enterprise campus
5:10is vulnerable so deploying several
5:13layers of security increases the
5:15protection
5:16of the enterprise campus where the most
5:19strategic asset usually resides
5:24so relying on physical security is not
5:26enough so for example
5:28as a visitor to the organization a
5:30potential attacker
5:32could gain physical access to devices in
5:35the enterprise campus
5:37okay often external access does not stop
5:42at the enterprise edge so some
5:44applications
5:45require indirect external access
5:50to the enterprise campus okay
5:53so strong security must protect access
5:57to this resources
6:00modularizing internal security so
6:04examples of network services embedded
6:07in the infrastructure services includes
6:09the following
6:10so first you've got the network
6:13management okay so it includes
6:16land management for advanced management
6:19of a multi-layer switches
6:21routed one management for monitoring
6:24traffic management and access control
6:27to administer the routed infrastructure
6:30of
6:31a multi-service networks service
6:34management for managing and monitoring
6:36service level agreements or slas
6:38and vpn security management for
6:41optimizing
6:42vpn performance and security
6:44administration
6:46so high availability ensures
6:50end-to-end availability for services
6:54so that includes clients and sessions
6:57so implementation includes reliable
7:00fault tolerant
7:01network devices to automatically
7:04identify
7:05and overcome failures and resilient
7:08network technologies
7:10so also we need the quality of service
7:14so manages the delay delay variation or
7:17also known as the jitter
7:19bandwidth availability and packet loss
7:22parameters
7:23of a network to meet the diverse needs
7:26of voice
7:27video and data applications so quality
7:30of service features
7:31provide value-added functionality such
7:34as
7:34network-based application recognition
7:37for classifying traffic
7:39on the application basis so cisco ios
7:43ipslas or the previously called
7:46the service assurance agent for an
7:49end-to-end
7:50quality of service measurements so
7:53resource
7:54reservation protocol signaling for
7:56admission control and reservation of
7:57resources
7:59and a variety of configurable queue
8:01insertion
8:02and servicing function
8:05so you also have multi casting or ip
8:09multicasting
8:10so it provides a bandwidth conserving
8:13technology that reduces network traffic
8:15by delivering a single stream of
8:17information intended for many recipients
8:20throughout the transport network so
8:22multicasting enables
8:24distribution of video conferencing
8:27corporate communications
8:29distance learning software and other
8:31applications
8:33so multicast packets are replicated only
8:36as necessary
8:37by routers enabled with protocol
8:39independent multicast
8:41and other supporting multicast protocols
8:44that result
8:46in the most efficient delivery of data
8:49to multiple receivers so to support
8:51network applications efficiently
8:54so deploy the underlying infrastructure
8:56services in some
8:57or all modules of the enterprise network
9:00as
9:01required so this design elements can be
9:04replicated simply to
9:06other enterprise network modules
9:09as network changes so as a result
9:13modularization to small subsets of the
9:15overall network simplifies the network
9:17design
9:18and often reduces the network's cost
9:22and complexity so the following are some
9:26of the recommended security practices
9:28in each of the module so you've got at
9:30the building
9:32okay so for building access access is
9:35controlled
9:36at the port level using data link layer
9:39information
9:41some examples are filtering based on
9:43media access control addresses
9:46and the ieee 802.1x
9:49port authentication okay
9:53now on the building distribution it
9:55performs
9:56filtering to keep unnecessary traffic
9:59from the campus core
10:01this packet filtering can be considered
10:04a security function because it does
10:06prevent
10:07some undesired access to other modules
10:11so given that switches in the building
10:13distribution layer
10:15are typically multi-layer switches and
10:17are there for layer 3 aware
10:20this is the first place on the data path
10:22in which
10:23filtering based on the network layer
10:25information can be performed
10:28okay so you also have the campus core
10:31layer
10:32so this is a high speed switching
10:34backbone
10:35and should be designed to switch packets
10:38as quickly as possible
10:40so it should not perform any security
10:42functions
10:43because doing so would slow down the
10:46switching of packets
10:49so next would be the server for modules
10:52which primary goal
10:53is to provide application services to
10:56end users
10:57and devices so enterprises
11:00often overlook the server for module
11:03from a security perspective
11:05okay so given the high degree of access
11:08that most employees
11:10have to these servers they often become
11:13the primary
11:14goal of internally originated attacks
11:17so simply relying on the effective
11:20passwords
11:21does not provide a comprehensive attack
11:23mitigation strategy
11:25so using a host based in network based
11:28ips
11:29and idss so private vlans
11:33and access control provides a much
11:36more comprehensive attack response so
11:39for example
11:40onboard ideas or ips within the server
11:44farm
11:44multi-layer switches that inspects
11:48traffic flows okay
11:51now the server farm module typically
11:53includes
11:55network management systems to securely
11:58manage
11:58all the devices and hosts within the
12:01enterprise architecture
12:03so for example syslog provides an
12:05important information on security
12:07violations
12:08and configuration changes by lagging
12:10security related
12:12uh events authentication authorization
12:16and accounting and so on so on
12:18authentication authorization and
12:20accounting or triple
12:21a that is a security server
12:24okay so also works on a one-time
12:26password or otp
12:28server that is to provide a high level
12:31of security
12:33to all local and remote users
12:36so a and otp authentication reduces the
12:39likelihood of
12:41successful password attack let me give
12:43you some
12:44information about aaa or the
12:46authentication
12:48authorization and accounting
12:51so aaa is a crucial aspect of network
12:53security
12:54that should be considered during the
12:55network design an
12:57aaa server handles the following
13:00so first is authentication with
13:04authentication that pertains to who
13:07authentication checks the user's
13:09identity typically
13:11through a username and password
13:13combination
13:15the next a is authorization
13:18this pertains to what so after the user
13:21is authenticated
13:23the aaa server dictates what activity
13:26the user is allowed to perform on the
13:28network
13:30so the last a is accounting accounting
13:33pertains to
13:34when the aaa server can record the
13:38length of the session
13:39the services accessed during the session
13:43and so on so everything will be logged
13:45in on the accounting
13:47server
13:52how about external threats so
13:55when designing a network security in an
13:58enterprise network
13:59the enterprise edge is the first line of
14:02defense
14:03at which potential outside attacks
14:06can be stopped so the enterprise edge is
14:09like a wall
14:11with small doors and strong guards that
14:14efficiently control any access
14:17so the following for attack methods are
14:19commonly used in attempts
14:21to compromise the integrity of the
14:23enterprise network
14:24from the outside so first would be
14:27ip spoofing okay
14:31so ip spoofing attack occurs
14:34when a hacker uses a trusted computer to
14:38launch
14:38an attack from inside or outside the
14:40network
14:42the hacker uses either an ip address
14:45that is in the range of the network's
14:48trusted ip addresses or
14:50trusted external ip addresses that
14:53provides access to specified
14:55resources on the network ip spoofing
14:58attack
14:59often lead to other types of attacks
15:03for example a hacker might launch a
15:06denial of services
15:07or dos using spoofed
15:11source address to hide his identity
15:14okay so you also have password attacks
15:18so using a packet sniffer to determine
15:21username and passwords
15:23is a simple password attack however
15:26the term password attack usually
15:29referred to repeated brute force
15:31attempts
15:32to identify username and password
15:34information
15:35so trojan horse programs are another
15:38method
15:39that can be used to determine this
15:41information
15:42so a hacker might also use ip spoofing
15:45as a first step in system attack by
15:49violating the trust relationship
15:51based on the source ip addresses
15:54so first however the system would have
15:57to be configured to bypass password
15:59authentication
16:00so that only username is required
16:04so you only ha you also have this dos
16:07attacks
16:08or the denial of service attacks so dos
16:11attacks
16:12focus on making a service unavailable
16:15for normal use
16:16and are typically accomplished by
16:18exhausting some resource limitation
16:21on the network or within an operating
16:24system or applications
16:26okay so also we have this application
16:29layer tab
16:31so application layer attacks typically
16:33exploit
16:34well-known weaknesses in common software
16:37programs
16:38to gain access to a computer
16:44designing high availability so most
16:48enterprise networks
16:49carry mission critical information
16:52organizations
16:53that run such networks are usually
16:56interested in protecting the integrity
16:58of
16:59that information along with security
17:03this organization expects the
17:05internetworking platform
17:06to offer sufficient level of resilience
17:10so this section introduces another
17:13network infrastructure service
17:15high availability to ensure adequate
17:18connectivity for mission critical
17:20applications
17:21high availability is an essential
17:23component
17:24of an enterprise environment okay
17:28now designing for high availability in a
17:30network so redundant network design
17:34duplicate network links and devices
17:36eliminating single point of failure
17:39on the network that is high availability
17:43so the goal is to duplicate components
17:46whose
17:47failure to disable critical applications
17:51so because redundancy is expensive to
17:54deploy and maintain
17:55redundant topologies should be
17:57implemented with care so redundancy adds
18:01complexity to the network topology
18:03and to network addressing and routing so
18:06the level of redundancy should meet
18:08organizations availability
18:10and affordability requirements so what's
18:14the key point here
18:15so before selecting redundant design
18:17solutions
18:19analyze the business and technical goals
18:22and constraints
18:23to establish the required availability
18:26and affordability
18:28so critical application systems internet
18:31working devices
18:32and link must be identified
18:35so analyze the risk tolerance and the
18:38consequences
18:39of not implementing redundancy and
18:42ensure
18:43that you consider the trade-offs of
18:46redundancy versus cost
18:47and simplicity versus complexity
18:51so duplicate any component whose failure
18:54could disable critical applications
18:57and redundancy is not provided
19:01by simply duplicating all links so
19:04unless
19:05all devices are completely fault
19:07tolerant
19:08so redundant links should terminate at
19:11different devices
19:13otherwise devices are not
19:16fault tolerant becoming a single points
19:19of
19:22failure
19:24how about designing for route redundancy
19:28so the following types of redundancy may
19:30be used
19:31in the modules of an enterprise so
19:34you've got sort of
19:35device redundancy including card
19:38and port redundancy you also have
19:42redundant physical connections to
19:44critical workstations and servers
19:47route redundancy can also be considered
19:50link redundancy and even power
19:53redundancy
19:54including redundant power supplies
19:56integral to the network devices
19:59and redundant power to building's
20:01physical plan
20:04now designing for redundancy so
20:06redundant routes have two purposes
20:09so that is to minimize the effect of
20:11link failures
20:13and to minimize the effect of the
20:15internet working device
20:17failure so redundant routes might also
20:20be used
20:21for load balancing when all routes
20:24are up okay now possible ways
20:28to make connection redundant includes
20:30the following
20:31so you've got parallel physical links
20:34between switches and routers
20:36you could also have back a plan and one
20:38links
20:39for example ddr backup for less lines
20:45now the following are possible ways to
20:47make the network redundant
20:50so a full mess to provide complete
20:53redundancy
20:54and good performance a partial mesh
20:58which is less expensive and more
21:01scalable
21:02the common approach when designing the
21:04route redundancy is to implement
21:06partial redundancy by using a partial
21:09mesh instead of a full mesh
21:11and backup links to alternative devices
21:15so this protects only the most vital
21:18parts
21:19of the network such as links
21:22between the layers and the concentration
21:24devices
21:28now example of campus infrastructure
21:31redundancy
21:32so a full mesh forms any to any
21:36connectivity and is ideal for connecting
21:39a reasonably small number of devices
21:42however as the network topology grows
21:45the number of links required to maintain
21:47a full mesh increases exponentially
21:50so the number of links in a full mesh
21:53is uh n okay times n
21:57minus one divided by two okay so where n
22:00is the number of routers so as the
22:03number of router appear increases
22:05the bandwidth and cpu resources devoted
22:07to processing routing updates
22:09and service requests also increases
22:13now in a partial mesh network that is
22:16similar to the full mesh network with
22:18some of its connections
22:19removed okay now a partial mesh
22:23backbone might be appropriate for a
22:25campus network
22:26in which traffic predominantly goes
22:29into one centralized server farm module
22:33now this figure here illustrates an
22:35example of
22:36route redundancy in the campus okay
22:40now in this example the access layer
22:42switches
22:44are fully meshed with the distribution
22:46layer switches
22:48okay if the link or distribution switch
22:50fails
22:51an access layer switch can still
22:53accommodate and communicate
22:55with the distribution layer
22:58now the multi-layer switches select the
23:00primary and back up path
23:02between the axis and distribution layers
23:04based on the links
23:06metric as computed by the routing
23:08protocol algorithm in use
23:10the best path is placed in the
23:13forwarding table
23:15and in case of equal cost path
23:18load sharing takes place
23:22how about example of edge redundancy
23:26so designing for link redundancy it is
23:29often necessary to provision
23:31redundant media in locations where
23:33mission critical application traffic
23:35travels
23:37so in layer 2 switch networks
23:40redundant links are permitted as long as
23:42stp is running
23:45so stp guarantees one and
23:48only one active path within the
23:50broadcast domain
23:51avoiding problems such as broadcast
23:54storms
23:55when a broadcast storms continuously
23:57loops
23:58okay now the redundant path
24:01automatically activates
24:03when the active path goes down so
24:06because
24:06one links are often critical pieces of
24:09the internetwork
24:10redundant media are often deployed in
24:13one environment
24:15now if you will observe on this figure
24:17here where
24:18frame relay circuit is used in parallel
24:21with
24:22a backup ipc connection over the
24:25internet
24:26so backup links can use different
24:28technologies
24:29and so with the frame relay which i said
24:32from the previous video
24:33is almost obsolete okay now it is
24:37important that the backup provides
24:38sufficient capacity
24:40to meet the critical requirements if the
24:42primary route fails
24:44so backup links can always on or can
24:47always be on
24:49or and become active when a primary link
24:51goes down
24:52or becomes congested so you have this
24:56connectivity okay so one is active the
24:59other one is standby
25:03now how about high availability in the
25:05server farm module
25:07so improving the reliability of critical
25:10workstations and servers
25:12usually depends on the hardware and
25:15operating system software in use
25:17some common ways of connecting includes
25:19the following
25:20so you could have the single attachment
25:23okay so which is not recommended
25:26so when a workstation on a server has
25:29traffic to send to a station that is not
25:31local
25:33it must know the address of the router
25:35on each network segment
25:38so if that router fails the workstation
25:40or server needs a mechanism to discover
25:43an alternative router
25:45so if the workstation or server has a
25:47single attachment
25:49it needs layer 3 mechanism to
25:52dynamically find an alternative router
25:54therefore the single attachment method
25:57is not
25:58recommended okay so
26:01the available mechanism includes address
26:03resolution protocol or the arp
26:06you've got a router discovery protocol
26:09or rdp
26:11routing protocols such as routing
26:13information protocol
26:15rep ospf eigrp
26:18you could also have this hot standby
26:20router protocol
26:22okay or the hsrp the gateway
26:25load balancing product called jlbp and
26:29the virtual router redundancy protocol
26:31vrp
26:34now example attachment through a
26:36redundant transceiver
26:38okay so this is an attachment through
26:41redundant transceiver
26:42so physical redundancy with redundant
26:46transceiver
26:47is suitable in environments where the
26:49workstation hardware or software
26:51does not support redundant attachment
26:54options
26:55okay
26:59now in this diagram here it is an
27:01example of attachment
27:02through redundant and icd okay
27:06so attachment through redundant network
27:08interface cards or nic
27:10some environments for example most unix
27:12servers
27:13support redundant attachment through
27:15dual analysis
27:17so primary and backup so the device
27:20drivers
27:21represents this attachment as a single
27:24interface
27:25to the operating system
27:28now for fast ethernet or giga ethernet
27:31port bundles so fast easter channel
27:35and gigabit ether channel port bundles
27:38group
27:38multiple pass or gigabit ethernet ports
27:42into a single logical transmission path
27:45between a switch
27:46and a router so host or another switch
27:50so stp treats the easter channel as one
27:53logical length
27:55so the switch distributes frame across
27:57the ports in an insert channel
28:00so this load balancing was originally
28:02done
28:03based only on map addresses so however
28:07newer implementations can also load
28:09balance
28:10based on ip addresses or layer 4 port
28:13numbers
28:15so source destination or source and
28:18destination addresses
28:19or port numbers can be used so if a port
28:23within an ether channel fails traffic
28:26previously carried over the field port
28:28reverts to the remaining ports
28:31within the insert channel so if you want
28:33to learn more about insert channel you
28:35could go ahead and check the
28:36supplementary videos on easter channel
28:42voice transport overview so voice
28:45services
28:46in a modular network design so to ensure
28:49successful implementation of voice
28:51applications
28:53network designers must consider the
28:56enterprise
28:57services in infrastructure in edge
28:59configuration
29:00for example to support voip
29:03the underlying ip infrastructure must be
29:06functioning
29:06and robust in other words
29:10don't even think of adding voice to a
29:12network experiencing other problems such
29:14as
29:14congestion or network failures
29:18okay now two voice implementations
29:22voice transport is general term so that
29:24can be divided
29:25into the following two implementations
29:27here okay
29:29so you've got the voip or the voice over
29:31ip
29:32and the ip telephony now voip
29:37uses voice enabled routers to convert
29:40analog voice
29:41into ip packets or packetized
29:45digital voice channels and route those
29:48packets between the corresponding
29:50locations
29:52users do not often notice that voip is
29:55implemented in the network
29:57so they use their traditional phones
30:00which
30:01are connected to the pbx however
30:04the pbx is not connected to pstn or to
30:07another
30:08pbx okay so back to a voice enabled
30:11router that
30:12is an entry point to voip
30:16so voice enabled routers can also
30:18terminate ip
30:19phones using session initiation protocol
30:23for call control and signaling so the
30:26next one would be
30:27ip telephony so for ip telephony
30:32traditional phones are replaced with
30:35iphones
30:37so a single or a server for call
30:40control and signaling such as this cisco
30:43unified communication manager
30:45okay so the ip phone
30:49itself performs voice to ip conversion
30:52and no voice enabled routers are
30:54required within the enterprise network
30:57however if a congestion to the pstn is
31:01required
31:01a voice enabled router or other gateway
31:05in the enterprise edge is added
31:08where calls are forwarded to the pstn
31:13so both implementation voip and ip
31:16telephony
31:17require properly designed networks
31:21so using a modular approach in voice
31:24transport design
31:26that is especially important because of
31:28the voice sensitivity to delay
31:30and the complexity of troubleshooting
31:32voice networks
31:34so all cisco enterprise architecture
31:36modules
31:37are involved in voice transport design
31:44ip telephony components so ip telephony
31:48components
31:49an ip telephony network contains four
31:52main voice specific components okay
31:56so this includes an iphone's okay
31:59an iphone are used to place calls in an
32:02ip telephony network
32:04they perform voice to ip and vice versa
32:08coding and compression
32:09using special hardware so iphones
32:13offer services such as directory lookup
32:17and internet access so the phones
32:20are active network devices that require
32:23power to operate
32:25so power is applied through the lan
32:26connection using the poe
32:29or with external power supply okay
32:32so the next one would be switches with
32:34inline power so you've got switches
32:36there
32:37so switches with inline power are poi
32:39enabled the modular wiring closet
32:42to provide centralized power for the
32:45telephony networks
32:46so these switches are similar to
32:49traditional switches
32:50with an added option to provide power to
32:54the lan ports
32:55where iphones are connected
32:58now the switches also perform some basic
33:00uos tasks
33:02such as packet classification which is
33:05required
33:06for prioritizing voice through the
33:09network
33:10so also you have this
33:13voice or the call processing manager
33:17okay or the call processing engine now
33:20the call processing manager
33:22such as the cisco unified communication
33:24manager
33:25it provides central control and
33:27configuration management for iphones
33:30so it provides the core functionality to
33:33initialize
33:34ip telephony devices and to perform
33:38all setup and call routing throughout
33:40the network
33:42okay now this cisco unified
33:44communication manager can be clustered
33:46to provide distributed scalable and
33:49highly available ip telephony model
33:52so adding more servers to plaster of
33:55servers
33:56provides more capacity to the system
34:00okay next would be the voice gateway
34:05okay so voice gateways also called voice
34:08enabled routers
34:10or voice enabled switches provide voice
34:13services such as
34:14voice to ip coding and compression so
34:17pstn access
34:19ipp packet routing backup call
34:22processing
34:23and voice services so backup
34:27backup call processing allows voice
34:30gateways
34:31to take over call processing in case of
34:34the primary call processing manager
34:36fails
34:37so voice gateways typically support
34:40a subset of a call processing
34:42functionality supported by
34:44this unified communication manager
34:48so other components of ip telephony
34:50network includes
34:52a robust ip network of course okay
34:55so voice messaging and applications
34:58and a digital signal processor resources
35:01to process voice functions in hardware
35:04so which is much faster than doing it in
35:07software
35:08so these components are located
35:10throughout the enterprise network
35:12as you can see here on this diagram
35:19modular approach in voice network design
35:24so implementing voice requires deploying
35:26delay sensitive services
35:28such as end-to-end in all enterprise
35:31network modules
35:34so use the modular approach to simplify
35:36design
35:38implementation and especially
35:40troubleshooting
35:42so voice implementation requires some
35:44modification to the existing enterprise
35:46network
35:47okay so that is in terms of performance
35:50capacity and availability because it is
35:52an end-to-end solution
35:54so for example so clients iphones are
35:58located
35:59in the building access layer
36:02okay and then
36:05the call processing manager is located
36:07in
36:08the server farm okay so therefore
36:11all modules in the enterprise network
36:14are involved
36:14in voice processing and must be
36:17adequately considered
36:20voice affects various modules of the
36:23network
36:24as follows now on the building
36:27access layer iphones and and user
36:30computers are attached
36:32to a layer two switches here so switches
36:35provides
36:36power to ip phones and provide
36:39quality of service packet classification
36:41and marking
36:42which is essential for proper voice
36:45packet manipulation
36:47through the network okay now
36:50on the building distribution layer this
36:53layer performs
36:54packet reclassification if building
36:57access
36:58is unable to classify packets or is not
37:02within
37:02trusted boundary so it aggregates
37:06the building access layer switches or
37:08the wiring closets
37:10and provides redundant objects to the
37:12campus core layer
37:15okay now on the campus core layer
37:19this forms the network score so all
37:22enterprise
37:23network modules are attached to it
37:25however
37:26okay so this could be virtually
37:29all traffic between application servers
37:32and clients
37:32traverses the campus core
37:36and with the advent of wire speed
37:38multi-layer gigabit switching devices
37:41land back bones might migrate it or
37:44might have migrated
37:46to switch gigabit architectures that
37:48combine
37:49all the benefits of routing with wire
37:52speed packet forwarding
37:54okay now on the server farm
37:58so this module includes multi-layer
38:00switches with
38:01redundant connections to redundant cisco
38:04unified communication managers
38:06which are essential for providing
38:09high availability and reliability
38:12now for the enterprise edge okay
38:16now the enterprise edge extends ip
38:19telephony
38:21from the enterprise campus to remove
38:23locations of via ones
38:25okay the pstn or the public switch
38:28telephone network
38:29okay and the internet
38:34example voice network solution so this
38:37figure here shows the voice network
38:39solution
38:40in the cisco enterprise architecture so
38:43it
38:44illustrates how a call is initiated
38:47on the ip phone okay
38:51so going to the destination
38:54okay so how the call setup goes through
38:57the cisco unified communication managers
38:59and how the end-to-end session between
39:02the iep phones or the two iphones
39:04is established so take note that the
39:07cisco unified communications manager is
39:09involved
39:10in only the call setup okay
39:13now calls this thing for remote
39:15locations
39:16traverses the enterprise edge
39:20that is through the one and man
39:23and also if possible that could be via
39:26the side
39:27side vpn module or through the remote
39:29access and vpn module
39:32so calls distinct for public phone
39:34numbers on the pstn are routed over the
39:36enterprise
39:37edge through the remote access and vpn
39:40module so
39:42calls between ip phones
39:45reverses the building access the
39:47building distribution
39:49okay and the campus score layers and of
39:52course
39:53the server farm module so although call
39:56setup uses these modules
39:58speech employs only the building access
40:03building distribution and in some cases
40:06the campus core layers
40:11evaluating the existing data
40:13infrastructure for voice design
40:16so evaluating the existing data when
40:19designing ip telephony
40:21designers must document and evaluate
40:24the existing data infrastructure in each
40:27enterprise module
40:29to help determine upgrade requirements
40:32so
40:32items to consider includes the following
40:35so first of course we have to talk about
40:38performance okay
40:41now enhance infrastructure for
40:43additional bandwidth
40:44consistent performance or higher
40:47availability if required
40:49might be necessary for the conveying or
40:52converging environment
40:54okay now performance evaluation includes
40:57analyzing the network maps
40:59device inventory information and network
41:02baseline information
41:04so links and devices such as with high
41:07picked
41:08or busy rs okay so use might have
41:13to provide sufficient capacity for the
41:15additional voice traffic
41:17so devices with high cpu use
41:20high backplane use high memory use
41:24cubing drops or buffer misses might have
41:28to be upgraded
41:29okay so the next one would be
41:32availability
41:34okay so redundancy in all the network
41:37modules should be reviewed
41:38to ensure that the network can meet the
41:41recommended ip telephony availability
41:43goals
41:44with the current or new network design
41:47next would be the feature requirements
41:49so examine the router and switch
41:51characteristics
41:52including the chassis module and
41:55software version
41:57that is to determine the ip telephony
41:59features
42:00okay in the existing environment
42:04next would be the potential network
42:06capacity or impact
42:08okay so
42:11for the capacity evaluate the overall
42:14network capacity and the impact
42:16of ip telephony on a module by module
42:19basis
42:20to ensure that the network meets
42:22capacity requirements
42:23and that there is no adverse impact
42:27on the existing network and application
42:29requirements
42:30now for power assess the power
42:33requirements
42:34of the new network infrastructure
42:37ensuring
42:38that additional devices will not over
42:40subscribe
42:41existing power consider taking advantage
42:44of the poi capabilities in devices
42:50how about wireless lan okay so wireless
42:54services in a modular network so a
42:57wireless lan or wlan supports
43:00mobile clients connecting to the
43:02enterprise networks
43:04the module clients do not have a
43:06physical connection to the network
43:07because
43:07wireless lans replace the layer 1
43:10traditional wired network usually cat5
43:12or cat5e okay so with
43:16the radio frequency or rf transmissions
43:18through the air
43:20now wireless lans are for local networks
43:23either in building line of sight outdoor
43:26bridging application
43:28or combination of post so in a wireless
43:31network
43:32many issues can arise to prevent the
43:34radio frequency signal from reaching all
43:36parts of the facility
43:39multi-pass distortion hidden node
43:43problems
43:44interference from other wireless sources
43:47and near far issues
43:50now a site survey helps find the regions
43:54where these issues occur by defining the
43:57contours of
43:58radio frequency coverage in a particular
44:01facility
44:02discovering regions where multi-path
44:04distortion can occur
44:06areas where radio frequency
44:08interferences is high
44:10and finding solutions to eliminate such
44:13issues so privacy and security issues
44:16must also be considered in a wireless
44:18network
44:19okay because wireless lands are
44:21typically connected to the world network
44:24so all the modules within the enterprise
44:26infrastructure
44:28must be considered to ensure the success
44:31of a wireless deployment
44:35centralized to our less than model
44:37components so as
44:38illustrated on this figure here the four
44:41main components in a centralized
44:43wireless lan deployment
44:44are as follows so first is basically
44:47you need the end users devices
44:51so a pc or other end user device in
44:54access layer uses wireless nic to
44:56connect
44:57to an access point okay so that is via
45:01radio waves next would be the access
45:04points
45:04or wireless aps aps or access points
45:08typically in the access layer
45:10are shared devices that function similar
45:13to hub
45:14so the cisco aps can either lightweight
45:16or autonomous
45:18lightweight aps are used in centralized
45:21wireless lan deployments
45:23a lightweight ap receives control and
45:26configuration from the wireless lan
45:28controller
45:28or the wlc with which it is associated
45:33providing a centralized point of
45:36management
45:37and reducing the security concern of
45:39stolen access point
45:41so an autonomous ap has a local
45:43configuration and requires
45:44local management which might make
45:48consistent configurations difficult
45:50and to the cost of network management
45:53of course that would be added okay so
45:56next would be
45:57the wlc or the wireless lan controller a
46:00wlc provides
46:02management and support for wireless
46:04services such as
46:06roaming the wlc is typically in the core
46:10layer of an enterprise network
46:13and of course you have to consider the
46:15existing switched and routed wired
46:17networks
46:18so the wireless ap is connect to the
46:20wired enterprise
46:22network so if you want to learn more
46:24about this wireless lan concepts and
46:26implementation
46:28you could go ahead and check the
46:29supplementary videos
46:34application networking services
46:37okay now traditional networks
46:40handle static web pages email
46:43and routine client server traffic
46:46so today enterprise networks must
46:50handle more sophisticated types of
46:52network applications that
46:54include voice and video so examples
46:57includes
46:58voice transport video conferencing
47:01online learning
47:02online training and audio and video
47:05broadcasts
47:06so applications plays increasing demands
47:10on id infrastructure
47:12as they evolve into highly visible
47:14services
47:16that represent the face of the business
47:19to internal and external audiences
47:22now the large amount of variety of data
47:25requires
47:26that the modern network be application
47:28aware
47:30okay so in other words
47:33be aware of the content carried across
47:36it
47:37to optimally handle that content
47:41now it is no longer enough simply to add
47:44more bandwidth
47:45as needs grown okay
47:48so networks have had to become smarter
47:52a new role of emerging for the network
47:55as
47:55provider of application infrastructure
47:57services
47:58that extend the value of applications
48:02now either by improving delivery of
48:04content to users
48:06and other applications or by offloading
48:09infrastructure functions that today
48:11burden development
48:13and operation teams so application
48:16networking services or
48:17ants provides this intelligence
48:22okay all right
48:25so ants can resolve application issues
48:29okay so for example
48:32say we have the consolidation of data
48:35centers
48:36result in remote employees having slower
48:39access to centrally managed applications
48:43so the sample and solution is basically
48:46wide area application services can
48:49compress
48:50okay cache and optimize content
48:54for remote users so that they experience
48:58a land-like responsiveness okay
49:02so another sample deployment issue would
49:05be
49:06a new website or a new web-based
49:09ordering system
49:10experiences a high proportion of
49:13abandoned orders
49:15because of poor responsiveness
49:18during the checkout process so
49:21the sample and solution would be
49:24optimization of the web streams being
49:27sent
49:28to an e-commerce portal which reduces
49:31latency
49:32suppresses unnecessary reloading of web
49:36objects
49:37and offloads low-level tasks
49:40from the web server okay
49:44so what else say you have
49:47uh a business partners need immediate
49:51and secure electronic access to
49:53information
49:55held in the back office applications
49:58such as shipment information
50:02so the solution okay might be
50:05security and remote connectivity
50:07services that automatically validates a
50:09partner's
50:10request route it to the appropriate back
50:13office application
50:15and encrypt and prioritize the response
50:19okay now last one would be
50:23sample deployment issue purchasing
50:26application needs
50:27to log and track orders over a certain
50:31value of
50:31compliance purposes so
50:34the and solution would be application
50:38messaging service that intercepts
50:40purchase orders locates the value
50:44and logs large orders to a database
50:47according to business
50:49policy rules okay so that's how
50:53ants can resolve application issues
50:58now what are the ants components now
51:01this
51:01figure here illustrates an example of
51:04ants
51:05deployed in offices connected over a
51:07wide area network
51:09providing a land-like performance to
51:12users
51:13in the branch regional and remote
51:16offices
51:17so ants components are deployed
51:20symmetrically in the data center and
51:23distant offices
51:25the ants components in this example are
51:28as follows
51:30so basically they've got this wide
51:33application services or us
51:36okay that is the software so the cisco
51:39software
51:41gives remote offices land-like
51:44access to centrally hosted applications
51:48servers storage and multimedia
51:52okay so another component would be this
51:54a wide area application engine
51:57okay or wa appliance
52:01now the cisco wa appliance provide a
52:04high performance
52:06global land like access
52:09to enterprise applications and data now
52:12the wa
52:13is use either was
52:16or was or the application and content
52:20networking system
52:21or the acns software now the waes
52:26help consolidate storage servers and so
52:29forth
52:30in the corporate data center with only
52:32low cost
52:33easy to maintain network appliances in
52:36distant offices
52:38okay next would be series content engine
52:41module
52:43okay now the content engine modules can
52:45be deployed
52:46in a data center or branch offices
52:49to optimize lan and one bandwidth
52:54to accelerate deployment of
52:56mission-critical web applications
52:58add web content security and deliver
53:02live and on-demand business videos
53:10now to summarize okay
53:13so network infrastructure services add
53:16intelligence to the network
53:17infrastructure
53:19supporting application awareness within
53:22the network
53:23so security is a network infrastructure
53:25service that increases the integrity of
53:27the network
53:29by protecting network resources and
53:32users from internal and external threats
53:36high availability services protect
53:38integrity of mission critical
53:40information
53:41with networking platforms and topologies
53:45that offer sufficient level of
53:48resiliency
53:50voice infrastructure services throughout
53:52the enterprise are needed to support ip
53:54telephony
53:55so wireless services support mobile
53:57clients and integrate it
53:59with a wired network and last would be
54:03the cisco ins or ans optimizes website
54:06performance
54:07content delivery and the security and
54:10connectivity applications

About this transcript

This page contains the full transcript of Structuring and Modularizing the Network Part 3 by Santelmo, generated from the public captions YouTube serves with the video. The transcript has 6,031 words across 1,337 segments, with the original timestamps preserved so you can click any line to jump to that moment in the embedded player.

What you can do with it

Use the transcript to take notes, quote the speaker, build a study guide, generate a summary with ChatGPT or Claude via the YouTube Summary tool, or export it as a timed subtitle file with YouTube to SRT. You can also re-open it in the transcriber to translate the transcript into 100+ languages.

Free YouTube transcript tool

YouTube2Text is a free YouTube transcript generator — no signup, no daily limit. Paste any YouTube link and get the full transcript instantly, with timestamps, click-to-jump, translation to 100+ languages, AI prompts for ChatGPT, Claude, and Gemini, and exports to TXT, SRT, VTT, or Markdown.