CCNA concentrates on systems for SMBs (small, medium companies).
A hierarchical design model is suggested.
Simpler to handle and expand.
Troubles are solved more rapidly.
Hierarchical design divides the network into 3 layers.
Core, (CL).
Distribrution, (DL).
Access, (AL).
Each layer provides specific functions.
This "modularity" facilitates scalability and gratifaction.
Access Layer, (AL): cheapest
Connects using the finish device (user).
Includes hubs, switches, bridges, modems and wireless APs.
Supplies a way of hooking up products towards the network and controlling which communicate around the network.
Distribution Layer, (DL):
Aggregates data caused by the AL before transmitting towards the CL for routing.
Controls traffic flow using guidelines and delineates broadcast domain names with VLANs defined in the AL.
VLANs allow traffic segmentation (separate subnetworks).
DL switches are usually high-performance products which have high availability and redundancy to make sure reliability.
Core Layer, (CL):
Our prime-speed backbone or even the internetwork.
Crucial for interconnectivity between distribution layer products ? must be highly available and redundant.
Frequently connects to Internet assets.
Aggregates tfc fm all products, so it should be able to sending considerable amounts of information rapidly.
Note: more compact systems frequently mix the distribution and core layers.
Three Logical Laye3rs are broken into a properly-defined hierarchy.
It's more difficult to determine these layers physically.
Advantages of a Hierarchical Network:
Scalability:
Hierarchical systems scale perfectly.
The modularity enables you to definitely replicate elements of design.
Expansion is simple to organize and implement.
Redundancy:
Like a network develops, availability gets to be more important.
Availability increases significantly with hierarchical systems.
E.G. AL switches connect with 2 DL switches. If a person DL switch fails, the AL switch can switch to another one.
Redundancy is restricted reaches the access layer. Typically, finish products don't connect with multiple switches.
Performance:
Correctly designed systems is capable of near wire speed btwn all products.
Security:
AL switches could be set up to supply treatments for which products are permitted for connecting towards the network.
More complex security guidelines offered at the DL.
Some AL switches support L3 functionality, but it's normally the job from the DL switches, simply because they can process it a lot more effectively.
Manageability:
Changes could be repeated across all products inside a layer simply because they most probably carry out the same functions.
Deployment of recent switches is simplified because configs could be replicated with couple of modifications.
Consistency within each layer simplifies troubleshooting.
Maintainability:
Due to their modularity and scalability, hierarchical systems are simple to maintain.
This means these systems are less costly.
Along with other designs, manageability becomes progressively complicated because the network develops.
Concepts of Hierarchical Network Design:
Hierarchical design isn't any guarantee of excellent design.
Simple recommendations help differentiate btwn well-designed and poorly designed hierarchical systems.
Network Diameter:
Normally the first factor to think about.
The # of products a packet crosses to achieve its destination.
Small diameter guarantees low and foreseeable latency.
Bandwidth Aggregation, (adding together):
Combines links btwn switches to attain up throughput.
'cisco' includes a proprietary link aggregation technology known as EtherChannel.
Aggregated links are shown by multiple dotted lines by having an oblong or perhaps a single, dotted line by having an oblong.
May be used at each layer (less frequent @ AL).
Redundancy:
Redundancy could be provided in many ways.
E.G. 2x connections btwn products, or 2x products.
Redundant links could be costly.
Creating redundancy begins in the AL. You make sure that you accommodate all network products ? 3 of AL switches.
This can help determine 3 of DL switches ? CL switches.
Exactly what is a Incorporated network?
SMBs are progressively running voice, video and knowledge.
Convergence is the procedure of mixing these.
Until lately it was restricted to large businesses.
Legacy (older) equipment hinders convergence.
Because analog phones haven't yet been changed, additionally, you will see legacy PBX telephone and IP-based systems.
Convergence has become simpler and fewer costly.
Having a convergence there's only one network to handle.
This is cheaper to apply and manage.
IT cabling needs are simplified.
Convergence also produces new possibilities.
You are able to tie voice and video straight into a worker's PC.
No requirement for an costly phone or video equipment.
Softphones ('cisco' IP Communicator) offer lots of versatility.
With software companies can rapidly become incorporated systems with little capital expense.
With cheap webcams videoconferencing could be added.
Separate Voice, Video and knowledge Systems:
Voice systems contain isolated phone lines running to some PBX (Private BDXT Exchange) switch situated inside a Telco wiring closet PSTN (Public Switch Telephone Network).
Telco closet frequently separate fm the information and video closets.
New phone ? a brand new line towards the PBX.
Utilizing a correctly designed hierarchical network voice lines could be added with little if any impact.
Since systems can hold the BW it seems sensible to converge.
Factors for Hierarchical Network Switches:
Tfc Flow Analysis:
The entire process of calculating BW usage and examining it for performance tuning, planning, and HW improvement.
To choose the right gear inside a hierarchical network, you have to spec out tfc flows, customers and servers.
Systems should be made with track of growth.
Done using tfc flow analysis software.
Should think about port densities and sending rates to make sure sufficient growth capacity.
Analysis Tools:
Many tfc flow analysis tools can be found.
E.G. Solarwinds Orion 8.1 NetFlow Analysis.
User Towns Analysis:
Identifies user grpings as well as their effect on internet performance.
Affects port density and tfc flow, which influences picking a network switches.
Typically customers are grped based on job function.
E.G. HR one floor and Finance on another.
Each dept. has different customers and requires, as well as use of different assets with the network.
Choose switches which have enough ports to satisfy the dept needs and pwrful enough to support tfc.
Good network design also factors within the growth.
Investigate tfc produced by finish-user programs.
Some user towns produce a lot, some don't.
The position of the user towns influences where data stores and server farms are situated.
By finding customers near to their servers, you are able to reduce network diameter, lowering the effect on other customers.
However, usage isn't necessarily bound by department or location.
Data Stores and knowledge Servers Analysis:
Data stores could be servers, SANs, NAS, tape bu models, or other hard drive or component.
Views both client-server and server-server tfc.
Client-server tfc typically traverses multiple switches.
BW aggregation and switch sending rates might help eliminate bottlenecks for this kind of tfc.
Some server applications generate high volumes btwn servers.
These server shouls be situated near to one another (i.e. guaranteed data centers).
Tfc across data center switches is usually high.
Requires greater carrying out switches.
Topology Diagrams:
A graphical representation of the network infrastructure.
Shows how all switches are interconnected, including which ports interconnect products.
It shows where and just how many switches are being used.
May also contain information on device densities and user grps.
Helps aesthetically identify potential bottlenecks.
Tough to create afterwards.
Switch Features:
Switch Form Factors:
Fixed or modular config, and stackable or non-stackable.
Thickness is expressed in rack models. (i.e. 1U, 3U).
Fixed Config Switches - Cannot add hardware.
Modular Switches - chassis enables for multiple line cards that have the ports.
The bigger the chassis, the greater modules it may support.
Stackable Switches:
Could be interconnected utilizing a special backplane cable that gives high-bandwidth throughput btwn the switches.
Cisco's StackWise technology enables you to definitely interconnect as much as nine switches using fully redundant backplane connections.
Stacked switches effectively operate like a single bigger switch.
Desirable where fault tolerance and BW availability are critical along with a modular switch is simply too pricey.
Performance:
Port Density:
Port density may be the 3 of ports available per switch.
Fixed sitches typically 1,000 ports!
Large enterprise systems require high density, modular switches to help make the best utilization of space and pwr.
Also prevent uplink bottlenecks.
A number of fixed swtches consume many additional ports for BW aggregation btwn switches.
Whith a modular switch, aggregation is a smaller amount of an problem since the chassis backplane offers the BW.
Sending Rates:
The processing abilities of the switch in bps.
Switch products are sorted by sending rates.
If the rate is not high enough, it can't accommodate wire-speed across all ports.
Wire speed = rate that every port is capable of doing (10Mbps etc).
E.G. 48-port GbE switch at wire speed = 48Gbps of tfc.
When the switch only supports 32 Gbps (internally), it can't run at full wire speed across all posrts concurrently.
Access switches typically don't need full wire speed since they're physically restricted to their uplinks towards the DL.
Link Aggregation:
Determine whether you will find enough ports to aggregate to aid the needed BW.
E.G. GbE 24-port switch could generate as much as 24 Gbps.
If it's attached to the network with a single cable, it may only forward 1 Gbps towards the relaxation from the network.
That leads to 1/24th wire speed for each one of the 24 products.
Link aggregation reduces these bottlenecks by permitting as much as 8 ports to become grped, supplying as much as 8 Gbps.
With multiple 10GbE uplinks high throughput rates could be accomplished.
'cisco' uses the word EtherChannel = aggregated ports.
Energy over Ethernet (PoE):
PoE enables a change to deliver energy over existing Ethernet.
May be used by IP phones plus some wireless APs.
Enables more versatility for equipment installations.
Adds considerable cost towards the switch.
PoE switch marked having a `V`for volts.
Layer 3 Functions:
Typically, switches operate at L2 and deal mainly with MAC addresses.
L3 switches offer advanced functionality.
L3 switches = multilayer switches.
Switch features inside a Hierarchical Network:
Access Layer switch Features:
Port security - first type of defense for any network.
The number of or what products are permitted for connecting.
All 'cisco' switches support port layer security.
VLANs - element of incorporated systems.
Voice tfc is usually given another VLAN.
Port speed:
Fast Ethernet is sufficient for Voice over internet protocol and many data tfc.
PoE - a lot more costly, so only use when needed.
Link Aggregation - supported whatsoever 3 lvls.
QoS - required for Voice over internet protocol.
Distribution Layer Switch features:
Collect all AL switch data and forward it towards the CL switches.
Offers the inter-VLAN routing functions.
DL switches have greater processing abilities than AL.
Need L3 to aid inter-VLAN routing.
Security Guidelines:
Need L3 so advanced security guidelines does apply.
ACLs control tfc flows via a network.
ACLs allow switches to filter tfc.
ACLs are CPU-intensive because they have to inspect every packet and match ACL rules.
Placing ACLs in the DL also cuts down on the 3 of switches that need the additional mgmt configuration.
Policy-based connectivity and retailéworkgroup accessibility core layer.
Service quality:
DL switches have to keep up with the focal points of tfc coming fm the AL switches which have implemented QoS.
If not completely the products support QoS, the advantages will disappear -. poor performance and quality.
DL switches they are under popular.
They require redundancy for sufficient availability.
DL switches are usually implemented in pairs.
Suggested they support multiple, hot swappable pwr supplies.
Finally, they have to support link aggregation and-bandwidth aggregated backlinks towards the core.
Core Layer Switch Features:
The CL may be the high-speed backbone.
The sending rate relies upon the amount of products taking part within the network.
When you purchase an insufficient switch fundamentally, you face potential bottleneck issues slowing down lower all tfc.
CL switches should support aggregated 10GbE.
L3 redundancy has faster convergence than L2, so, ensure CL switches support L3 functions.
CL switches should support FULL redundancy features.
QoS is essential fundamentally since high-speed WAN access is frequently prohibitivel costly.
Switches for SMBs:
Find out the 'cisco' switches utilized in SMB programs.
The characteristics of 'cisco' Catalyst Switches:
You can't simply choose a switch by how big a company.
Companies are frequently mix integrated along with other organizations.
A 6500 is sensible being an AL switch where you will find 100s of customers within an area, like a stock market.
Cat Express 500 - sending rates = 8.8 - 24 Gbps.
Cat 2960 - L3, QoS, no PoE, 16 - 32 Gbps.
Cat 3560 - enterprise-class PoE, QoS, 32 - 128 Gbps.
Cat 3750 - stackable high end.
Cat 4500 - DL midrange modular Up to 136 Gbps.
Cat 4900 - data center.
Cat 6500 - DL and CL Up to 720 Gbps.
Miscellaneous:
MDF - Primary Distribution Facility.
Gi é1 - abbreviation for Gigabit Ethernet ports.
Spanning Tree - methods enables redundant pathways, but shuts lower some links to prevent switching loops.
No comments:
Post a Comment