Layer 3 Vs Router

Hi..
I am new in this field.
How a Layer 3 switch is technically different from a Router?

Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
A router is a router because the vendor so labels it and a L3 switch is a L3 switch because the vendor so labels it.
As Jon also notes, L3 switches generally have more hardware dedicated to packet forwarding (often very Ethernet only oriented) while routers usually have a much richer feature set and/or media support, but the line is blurred.
I've worked with some vendor X equipment labeled SSR (for smart switch router), so even vendors can blur the line.
Years ago, the 6500 and 7600 could use not only the same sups and some of the same line cards, they ran the same IOS image! However the 6500 was a "L3 switch" and the 7600 was a "router". The did differ, as the 6500 supported service modules that the 7600 didn't, while the 7600 supported special WAN cards that the 6500 didn't. Personally, I suspected the difference was more due to vendor marketing then what could be done technically.

Similar Messages

Which layer is the Routing protocol represented in the OSI model?

Hello friends,
I am a student with surface level of knowledge in networking and looking for some ideas about OSI layer models.I am confused "Is routing protocol belong to network layer of the OSI model or it is the Application layer?".Please help me out with the logical reasoning.
Thanks in advance!
Subash Basnet
[email protected]

Hello Subash,
There is a lot of discussions happend earlier even.
Now as per my understanding first of all routing job is done be a layer3 device using these protocols. All of them affect the routing table of a layer 3 device...
We call a device layer 3 when it does a network routing job....
BGP=TCP port 179(L4 or L7) uses a layer 4 tcp port as if an application
RIP = UDP port 520 (L4 or L7) uses a layer 4 udp port as if an application
OSPF = IP protocol 89 (L3)
EIGRP = IP protocol 88 (L3)
IS-IS = CLNS protocol (L2) connection less network service it runs at data link layer..
Please rate the helpful posts
Regards
Thanveer
"Everybody is genius. But if you judge a fish by its ability to climb a tree, it will live its whole life believing that it is a stupid."

Multi-layer/layer3 switch VS. Router

Multi-Layer Switch or Layer3 switch vs. router; How they are different?
1.7

In a router the route calculation and packet processing take place in the software on layer 3. This means that packets need to be moved from the layer 2 hardware interface to layer three and so it takes some time. In a layer 3 Switch Routing calculations takes place at layer 3 in hardware or software, while the actual packet processing takes place at layer 2. The speed gain is accomplished by reducing the amount of features supported and moving as much logic as possible into hardware.

Transport layer and transport package

I have a fundamental query on transport package and layer.
Objects are classified in packages depending on the project and put in $TMP or test package else they are local ? Only objects belonging to a transport package can be transported , others are not ? How transport package is defined ? All objects must be a part of transport package else can not be transported ?
ABout transport layer, I understand this is used to define which path objects will take during transport particularly when there are multiple transport routes. But what are the advantages of it ? How we can define it (trx code/menu path) ?
Is it mandatory to use transport package and layers ?

Hi Mike,
Just to clarify a few basic concepts:
A transport package (formerly called a "development class", a term you may still encounter) groups related development objects. Whenever you create a new development object (table, program, ...) you have to assign it to a package; the object is then placed into a transport request. Alternatively you can indicate it is a local object (its package is then "$TMP") but such an object is non-transportable.
A transport layer identifies the route that transportable devmt objects must follow. Typically in your system you will have two layers: a layer usually called Zxxx (with xxx = SID of your development system) is used for your own developments. Another layer called "SAP" is used for transporting SAP objects, for example changes to standard SAP code implemented via transaction SNOTE and recorded in repairs. Without the "SAP" layer in place, repairs would be non-transportable.
Each transport package belongs to a layer. You assign the layer when you create a new package in transaction SE80. You can change the layer later on, but this is rarely needed and can cause quite a few side-effects. SAP objects always have layer "SAP". The layers themselves are managed in STMS, under "Transport Routes".
Hope this clarifies things,
Rgds, Mark

SG300: How to set up routing between VLANs?

I have recently purchased a Cisco SG300-10. I need it to perform routing between two VLANs on the switch. Seems like this should be quick and easy to do from the built in GUI. When I configure it according to the documentation, it does not ropute between the VLANs.
I have set the system mode to L3 (for level 3 switching).
I have followed the instructions on pages 26 through 33 of the attached PDF (which I obtained from the Cisco site). I used the same ports on the switch and the same IP addresses as shown in the document.
Everything works until I attempt the step "ping 10.1.1.10" on page 33. This is the step to verify the level 3 switching between the 2 PCs (on separate VLANs).
The switch Firmware Version (Active Image): 1.3.5.58
I have attached the running configuration from the switch. It is the file named "running-config.txt".
The 2 PCs that I am using are running Windows 7 and Windows 8.

Hi jkst,
There is a very minimum requirement to obtain layer 3 intervlan routing
1- 2 VLAN in layer 3 mode assigned an IP address
config t
vlan database
vlan 2
int vlan 1
ip address 192.168.1.1 /24
int vlan 2
ip address 192.168.2.1 /24
2 - Active link state on each VLAN - Define a port for the second vlan then connect an IP device to that port and another device to another port since the rest of the ports will default to vlan 1
config t
int gi2
switchport mode access
switchport access vlan 2
3 - Assign your device #1 that connects to any port an ip address on the same subnet as vlan 1
Computer in vlan 1 IP info=
192.168.1.100
255.255.255.0
192.168.1.1
Computer in vlan 2 IP info-
192.168.2.100
255.255.255.0
192.168.2.1
Assuming these devices respond to ping and do not have external wireless communication, this will provide basic IP connectivity through the switch across vlans.
-Tom
Please mark answered for helpful posts

ChaRM: transport route migration

Hi!
we have a lot of different transport routes with different transport layer and would like to implement ChaRM.
We have a lot of different transport routes with different transport layer where we assign the developer classes/packages.
Now we would like use ChaRM.
As we know ChaRM uses standard transport layer and transport route.
Therefore we plan to consolidate all the developer classes/packages assigning to none-standard transport layer and route and migrate them to standard transport layer and transport route
Questions:
1) Can some one confirm whether or not ChaRM uses the standard transport layer with standard transport routes?
2) Or is it possible to create own transport layer u201CZCRMu201D, assign them all the development class/package relevant for ChaRM and leave the other none-transport layer unchanged?
3) Can one development class/package be assigned to several transport layers?
4) Has some one experience dealing with the approach of migration?
(tcode SE16 --> T_DEVC)
It will be great get some information regarding these questionsu2026
Thank you very much!

Hello,
We set up a testing environment for change request management. In a solution manager we created five clients. One single development which feeds two quality systems, and final two production system. As it was said in the thread we used transport groups in the transport route. The development system feeds this transport group which feed to import queus in quality systems. We tested this transport route and transports worked well and arrives into both production 1 and production 2 systems.
Then created a logical component for that landscape, with five sytems. Then we created a maintenance project using that logical component. We created a maintenance cycle and then a task list was created which shows two delivery system, and the two productive systems. All the groups have the right actions.
However when we created a SDHF correction the system creates the task list only for 3 systems, not for the five of them.
we see that only action groups are created for development, quality 1, and production 1. No entries for quality 2 and production 2
We try to run the action which is in the general part of the task list to execute the import in multiple systems, it cames out with the correct selection of importing in quality 2 system because it is the transport group. However an error is raised saying that the actions for quality 2 are not there in the task list which is true. Finally, as short term solution, we were able to import into quality 2 and later on pr1 but done manually in the stms transactions which is not what we want to do.
Regards
Esteban Hartzstein
Edited by: Esteban Hartzstein on Aug 24, 2010 2:06 AM
Edited by: Esteban Hartzstein on Aug 24, 2010 2:07 AM

In TMS, can you have routes from a dev to multiple consolidation systems?

My developers want to be able to choose from a list of target systems when creating a transport so that transport would go to a target group, or individual systems.
Currently, I have my my ECC 6.0 Dev system transport routes configured to send to a transport group which adds the transports to two other systems queues. there is a SAP layer and zxxx layer. All custom development classes
use the zxxx layer.
My problem is that sometimes they only want the transport to go to one system or the other, but not both. They would like to choose which system to use in the target system field when creating a transport. Is this possible?
All development classes (packages) use the same
transport layers, either sap or zxxx, I have tried to create routes from DEV to these to 3 different locations, but TMS only lets me use the layer for one route. if I create a SAP route to the /ZXXX/ target group it will not let me also create a SAP route to the individual systems. It appears you can only use a layer to go to one destination.
Is there some way to give them a choice? The developers swear they have seen this before in development systems.
Thanks, Pat

Hello Pat,
The customer i work for has a complex landscape. We have one development system and 6 quality systems.
I have just checked the transport routes in STMS and i see that there are a lot Z layer created for the transport routes to the multiple systems.
Hence i agree with you from an SAP standard point of view all options are closed out. (and believe me i am working on the old R/3 3.1 version)
But what is the problem with transport target group. Just because a request goes and sits in the buffer of a system, it does not mean the request has been imported. right? why cant you convince them on those lines?
But at the same time, i hope your company does not use the standard SAP concept od doing an import all every few hours.
Regards,
Prashant

Layer 2 vs Layer 3

We are looking at purchasing a new switch. What I am wondering is I want to buy a layer 3 switch, but only use it as a layer 2 switch for the time being. I want the option of using layer 3 functionality later, should we see the need. This way the equipment is in place and we don't have to purchase anything new in the middle of a budget year.
Thanks in advance!

Depending on your budget and requirements, there are a lot of options in the Cisco product portfolio. The lowest model of switch with Layer-3 capability is Catalyst 3560. Out-of-the-box, these series support basic Layer-3 (static routing and RIP), or you can upgrade the s/w to enhanced layer-3 (EIGRP, OSPF, BGP).
My suggestion is to go straight to enhanced layer-3 model of switches, and disable ip routing. That way, it provides you with the flexibility of migrating to layer-3 later on, using any routing protocols you want.

Intervlan routing

we are inserting two new 6513 into our lan to be our core. We have configure core1 to be root for odd vlans and core2 for even vlans. We can route traffic between odd to odd per vlan and even to even per vlan. How do we route for odd to even. We have an etherchannel connecting between the two they are secondary roots for the opposite vlans.

Pete
It would be helpful to know a bit more about how you configured the 6513s, especially how you have the layer 3 vlan interfaces configured. When you talk about core1 being root for odd vlans this is a layer 2 consideration. It has little to do with how layer 3 will route. (It might have some significance if you configured HSRP to run between the core switch/router, this is one of the things we do not know about your configuration)
If you configure vlan interfaces for each vlan (on both switches) then routing between vlans is easy and it just happens. If you only configure odd vlans on core1 and even vlans on core 2 then you need to configure a layer 3 connection between the cores and you need to run some routing protocol between the cores.
HTH
Rick

Report a bug, if this problem is not solved, p2p of rtmfp will not be of much use

Recently, we have several clients on a p2p application, but we use the development version to test, two-way voice and video connection is not successful, several customers of the network environment is as follows:
First of all, there is no firewall;
These networks have in common is:
After an optical access, the subnet with a lot of routers, one by one, and some series, some parallel, some in series and then parallel, and some in parallel then in series, in short, they can be online, and then not the same segment of the computer, p2p connection is not successful, and other computer connection is not successful, and some also use the vpn.
Analogy: 192.168.5.10 and 192.168.10.10 to connect not, 192.168.5.10 and 192.168.5.11 is ok, is 192.168.5.10 may be outside the network of other computers can connect to, but 192.168.10.10 possible and other computers outside the network connections are not.
Hope you solve this problem as soon as possible, because rtmfp too expensive, otherwise p2p applications, are generally large companies (schools, government) affordable, and their networks are very complex, they do not take, you will no client .
My English is not very good, google translation
I gave you a bug report so long, how did not care for me ah;
This forum nobody, or you did not test this bug, you do not have time to correct this bug;
Either we modify not the flash player platform to the bottom of things, I will not give you advice.
In fact, standing on a technical point of view to think that I have is to use a socket network communication can be OK, network barrier, it is not.
But would like the user's perspective, I can be normal to the Internet, other p2p communications software, I can, on this thing, impossible, it is this software not work, so that this bug is that we must solve the .
I am very concerned about this issue every day and came to see nobody replies, I think this problem, I think, to solve this problem, you can proceed in two ways:
1.dns
I think it may be able to solve this problem, dynamic dns,
The idea is: to allow users to connect to the server, the user computer first upd listener (that is, iis services), then farid + domain to form a secondary domain name server dynamically to resolve the domain name to the user's computer, in a timely manner to refresh dns.
In this way, the user's computer, there is a fixed identity.
2 gateway
From users within the LAN to solve this problem, it can only proceed from the gateway.
For example, to use arp and other means to get to the router, where each segment ip then ip, similar to the route command to specify more than one gateway to the user's computer may be able to solve this problem.
In theory, these two methods may be feasible.
We also expect a new solution, I do not know Daoxing not
Speak before this program, take a look at our test for RTMFP p2p network topology, and the test results (explain, test, do not open the firewall and router are not set up static routing).
Computer A, computer B - "router A -" server S
The above NetConnection.Connect.Success 95% NetStream.Connect.Success 5% NetStream.Connect.Fail
Computer A - "Router A - servers S" - routers B - Computer B
Above, NetConnection.Connect.Success, NetStream.Connect.Success
Computer A - "Router A - servers S" - router B - routers C - Computer B
Above, NetConnection.Connect.Success, 50% NetStream.Connect.Success 50% NetStream.Connect.Fail
(Computer A - Router A, computer B) - router B - server S
Above NetConnection.Connect.Success, 100% NetStream.Connect.Fail
Computer A - "Router A -" router B - server S "- Router C" - router D "- computer B
Above NetConnection.Connect.Success, 100% NetStream.Connect.Fail
Computer A - "Router A -" Router B - "Router C -" server S "- Router D" - Router E "- computer B
Above NetConnection.Connect.Success, 100% NetStream.Connect.Fail
See the above test results, is not very bad, the key is FMS4.5 sell 450,000 RMB in China, their own software does not do a good job, but also to sell such a high price in China, a man named QQ. The above test environment, all 100% NetConnection.Connect.Success 100% NetStream.Connect.Success, free, QQ has more than 100 million users, QQ companies rely on advertising to make money game.
Seems to digress, huh, huh.
The test results, 100% NetConnection.Connect.Success, and NetStream.Connect.Fail the main reason is because the route connected in series or more than two or two and then caused.
From a technical perspective, as a packet travels through a number of the NAT router default all packets can be out, new packets can not be into the second layer of the router caused.
Find the reason, we think of a countermeasure to solve this problem from the software itself.
The last case of the above test environment, for example:
When computer A and server S NetConnection.Connect.Success computer B with the server the S NetConnection.Connect.Success, computer A to Router D sends an upd packets, not because the router can only broadcast the data packet to computer B, so 100 % NetStream.Connect.Fail.
The 100% NetConnection.Connect.Success, if we can NetStream.Connect, and NetConnection.Connect connected together, is not can be 100% NetStream.Connect.Success?
In other words, no matter what circumstances, computer A and computer B can and server S connection is successful, the other way round, that is, the server S and the Computer A or Computer B can connect.
We can not let the computer A or computer B, pretending to server S (In theory, called NAT, technically speaking, is to modify the source ip of the packet header and source Mac server S, and parity), the router not in the NAT, and other communications.

Hi Mr. Wog,
My guess is that once you see an estimate for replacement for hundreds of dollars, those scratches won't look too bad anymore. There are products to cover the outer casing, and some people put stickers on, or even paint (I don't recommend either, but ymmv). Good luck.

Switch and Broadcast filtering

I read this article in the cisco curriculum, but I did not understand it well :
" Occasionally, a device will malfunction and continually send out broadcast frames, which are copied around the network. This is called a broadcast storm and it can significantly reduce network performance.
A switch that can filter broadcast frames makes a broadcast storm less harmful.
Today, switches are also able to filter according to the network-layer protocol. This blurs the demarcation between switches and routers. A router operates on the network layer using a routing protocol to direct traffic around the network. A switch that implements advanced filtering techniques is usually called a brouter. Brouters filter by looking at network layer information but they do not use a routing protocol ".
Can the switch filter the broadcast ? Yes, it can,,,,as Cisco says :"This filtering is achieved through the implementation of virtual local-area networks or VLANs ".,,,,,Is there any other type of filtering ?
What is the main difference between router and brouter

hi
if u would like to control the broadcast and multicast storms you can refer the link for configuring the storm control for both broadcast and multicast.
you can define up the values and shut the port if it exceeds the threshold limit..
http://www.cisco.com/univercd/cc/td/doc/product/software/ios124/124cr/hif_r/int_s4h.htm#wp1229258
About the difference between a router and a Brouter afaik BROUTER u use in most of the SP network where you got customers in either DSL or Metro ethernet network where you will have the aggregation of the whole network traffic and from where it will be forwarded to upstream.
It depends on the ios code too which is available to serve the purpose for the same..
you got to have something like 7200 or 7300 in place to serve your purpose of brouter.
regds

TMS Setup - One package to two seperate landscapes

My customer has one development instance that services two test and production landscapes for seperate geographical regions. This is a new installation and they are developing using packages. They currently belive that they can transport the objects from those packages to each landscape via the same package, transport layer and transport route. I'm contending that that is not possible due to each package having only one transport layer and that they will have to create a seperate package with the same objects for each landscape if they want to transport to both. I'm a bit inexperienced at this and could use any advice available to confirm that either I am right or that it is possible to transport to two seperate landscapes.. Is it possible to setup STMS transport routes to transport to two seperate environments from the same transport layer?
Regards.

Hi Joe,
Your contention is correct. A package/development class can be assigned to only one transport layer at a time.
My solution may not really match your business requirements. However if it possible to identify objects that will be needed in both the systems then you can assign them a single package which will have a common transport layer which will serve both the test systems and subsequently production systems. A bit of manual work but should solve this for you.
Regards.
Ruchit.

Transport to multiple systems using one package

My customer has one development instance that services two test and production landscapes for seperate geographical regions. This is a new installation and they are developing using packages. They currently belive that they can transport the objects from those packages to each landscape via the same package, transport layer and transport route. I'm contending that that is not possible and that they will have to create a seperate package with the same objects for each landscape if they want to transport to both. I'm a bit inexperienced at this and could use any advice available. Is it possible to setup STMS transport routes to transport to two seperate environments from the same transport layer?
Regards.

Hi Joe,
One thing that you can do is to ask you Basis people to provide the Data and CO files for your transport request
You can then send these files via mail, CD etc to any number of locations and import it into the system

XI and IDOC-Adapter / Restriction to certain IDOC Types?

Hi all!
We have some problems in trying to enable our XI systems for using our central user administration. The reason for that is that the "USERCLONE", ... IDOCs sent to XI by the CUA system are not processed normally in the ALE Layer but XI tries to process it through it's IDOC Adapter hence wants to route the IDOCs through the Integration Engine.
Is there a way to define the IDOC-Types to be processed by the IDOC-Adapter and to exclude certain IDOC types such that those are processed via standard ALE?
Many thanks in advance for any input you can give ...
Andy

Hi all!
For those who are interested - we have found the solution for this problem using some debugging
There is an "excemption table" where you can define required IDOC Types that should be processed via the normal ALE layer (instead of routing it through the IE).
To maintain the table entries you have to use report
"IDX_SELECT_IDOCTYP_WITHOUT_IS".
However - if you want to get rid of entries already made
with this report just have a look at note "790175"
Best regards,
Andy

OSPF Equal Cost Path Selection

This is a nerdy enough qeury in reality.
We have a single area - area 0.0.0.32. All intra-area routes. We have 2 switches in the core of the network, and 10 switches at the edge. All of these switches are connected via layer 3 OSPF routed links.
The cost for all links is 20 - which is based on bandwidth between the boxes - which is 2Gbps.
Have a look at the enclosed jpeg to get an idea.
Very simple.
Query revolves around the path selection available to OSPF.
The path from Core 2 to the 10.32.51.0 network is easy - straight across the link between the core switches for a cost of 20.
Question :- if the link between the cores fail, which path will be chosen by OSPF and why?
It can go through ANY of the other edge switches for a total cost of 40, but it will choose a particular one. What criteria does OSPF use to select this path?
Remember, the path costs are equal, they are all intra-area.
I tried messing with Router ID, but this doesn't seem to be it.
I tried highest interface IP addresses, but it doesn't seem to be this.
It is not a random act, the algorithm chooses the same one every time.
There must be some parameter in the LSDB that is the defining one when it comes to path selection.
Can you help me out please. I need someone who understands the OSPF algorithm better than I do (which might not be hard!).
Appreciate any comments for debate.

Disclaimer
The Author of this posting offers the information contained within this posting without consideration and with the reader's understanding that there's no implied or expressed suitability or fitness for any purpose. Information provided is for informational purposes only and should not be construed as rendering professional advice of any kind. Usage of this posting's information is solely at reader's own risk.
Liability Disclaimer
In no event shall Author be liable for any damages whatsoever (including, without limitation, damages for loss of use, data or profit) arising out of the use or inability to use the posting's information even if Author has been advised of the possibility of such damage.
Posting
My guess (as I haven't re-read the RFC), selection of an ECMP to retain in a routing table (assuming all possible ECMP are not retained) and/or exactly how packets or flows are ECMP routed is implementation dependent.
I recall years ago bumping into a situation where I had 6 (OSPF) ECMP on a Cisco router which had the (then) default allowance of 4 ECMP in the route table. I don't recall exactly what the issue was, but whatever it was doing I considered it a bug. The "fix" was to allow the router to use all 6 ECMP. (Again, don't remember the specifics, but the issue I saw was more involved than 2 of the 6 ECMPs weren't retained.)

Layer 3 Vs Router

Similar Messages

Maybe you are looking for