MPLS Benefit???????

Hi all,
I am investigating why we, as an enterprise, should move toward MPLS. What benefit does it give you? Anyone implements MPLS (not buying MPLS service) on their network and do you notice any difference? Do you gain any more speed? Are you able to control traffic better than tuning/redesigning IGP? Why QoS on MPLS is better? Please let me know. Thank you!

Hi,
here is some links.. I think they would be useful for you.
http://www.eventreg.com/cisco_emea_nw2005/content_viewer/
- IPM-D02 - Designing an Enterprise MPLS Network
- IPM-D03 - Advanced MPLS Deployment in Enterprise Networks
http://www.cisco.com/en/US/netsol/ns483/networking_solutions_customer_profile0900aecd802b1d7f.html
http://www.cisco.com/en/US/netsol/ns656/networking_solutions_program_category_home.html
- Next Generation Enterprise MPLS VPN-Based MAN Design and Implementation Guide
A little bit off topic :-)
http://www.juniper.net/solutions/literature/white_papers/200183.pdf
Regards,
Alexander

Similar Messages

MPLS network design questions

we have in our company 230 remote sites, and we are changing all of our circuits to MPLS. wondering if i need to get a high end router in our Data Center? currently we have 3925. also what is the best routing protocol to use in this kind of network? Eigrp or ospf? MPLS will be hosted by the service provider

I have found that the provider typically wants to know exactly what routes you will be advertising when using EIGRP or OSPF. This is something they will have to configure on their network to allow. For example EIGRP routes flow from Site A to Site B and you have a new subnet to use for an application. You put the proper network statements in EIGRP and are not learning routes on the other side. You would then have to fill out a form or call a support number to get your new network to the other side.
With BGP there is much more control over what you can advertise with adding networks. With 230 sites you will feel the benefit quickly if you start growing and adding subnets in data centers or additional sites.

MARS and MPLS networks

Hi folks,
I have an 80+ node network connected via a service provider managed MPLS VPN cloud. Each one of my 'spokes' connects to the provider edge (PE) router via a /30 subnet. All of my customer edge (CE) routers are in MARS, but of course the PE routers are not. When MARS graphs my network it shows 80 'stub' networks with no connectivity between them.
I've spoken with the TME and I think the feature is planned for a future release, but I had a idea I wanted to bounce off of the folks here:
What if I created a virtual generic router in MARS, and populated it with 80 interfaces. Each interface would have the corresponding PE router's /30 IP address. I think this would let MARS tie all of the stub networks together.
Has anyone tried this? Will it only fix the graphing problem, or is there added benefit for MARS when everything is tied together as a cohesive network.
Any input would be greatly appreciated.

I think the idea of creating a virtual router and then assigning its interfaces with corresponding IP addresses will work for MARS to tie the networks. It could fix the graphing problem but I dont think anything more it will deliver.

6500 Virtual Switching System (VSS) design in MPLS PE PoP

...

Hello,
The VSS is not designed to function as a PE , its best fit in a DataCenter Environment where I have Multiple Links from Core/Access to the Distribution layer.
Although VSS provides Higher backblane Throughput, Ease and Single Administartion Managment, and Redundancy SSO between SUP engines at the Control Plane, These features wont help at the MPLS edge, why? because Both Chasis have identical configuration, So I cant have two links from the CE or from the P router to the same VSS Chasis, So I have No redundancy here.
what if a VSL link fails between the Two Chasis, Each chasis will revert to be as a standlone chasis with Identical Configuration, So I dont have redundancy here.
The VSS would be an ideal option as a PE router if the CE and P routers can form a direct Layer-3 Eitherchannel to each of the VSS chasis in order to benefit from a redundancy, in this case we can say we have (Intra Chasis Redundancy).
Since this Type of Scenario is not an option, then its normaly Service Providers prefer (Inter Chasis Redundancy) by having Two Different PE routers with two different Configuration to have resilency.
Other than that, it could quitely fit.
Regards,
Mohamed

MPLS benefits over P-to-P Leased Links

Hi,
Benefit of MPLS over traditional P-to-P
is off course a cost. But apart from cost what are techonoloical benefits of having MPLS over leased links. Is'nt MPLS less secure as compared to leased circuit.
In a nutshell what are convincing factor for MPLS over P-to-P leased line.
Sounds too basic :)
Any link on cisco.com or any other is appreciated.
Thanx in advance
Subodh

MPLS VPN's have a direct comparision to other shared access media VPN's like Frame Relay ATM etc. It cant be compared directly to a P2P Private Circuit.
Here is a link comparing MPLS VPN sec to FR and ATM.
http://www.cisco.com/warp/public/cc/so/neso/vpn/prodlit/mpvpn_wp.pdf
Having said that it does not mean there are no loopholes in P2P private circuit and it is completely secure. Looking at the economy and the security provided by MPLS VPN's its comparable to P2P Private Circuit if not better.
But again providing MPLS based VPN;s has more benefits to the service provider that the end customer. For the customer the only major benefits is with scale more number of services would be available to him if using a MPLS VPN at lesser cost when compare to other technology based VPN;s.
HTH-Cheers,
Swaroop

Why we choose mpls over point to point only because it is cheap

I need a document explaining the advantages and disadvantages of converting point to point connectivity to MPLS and the plan / suggestion to overcome the disadvantages

Hi,
Converting a network that consists of point-to-point links to run MPLS can have benefits. This is regardless of the type of links you have in the network.
One starting point could be this one:
http://en.wikipedia.org/wiki/Multiprotocol_Label_Switching
The big benefit that comes with deploying MPLS in your network is the ability to run the MPLS applications. Here are the most important ones:
-MPLS VPN - Layer 3 (with multicast)
-MPLS VPN - Layer 2 - point-to-point or point-to-multipoint
-MPLS Traffic Engineering - point-to-point or point-to-multipoint
Each of these have their benefits, while running over a one common network : an IP network with MPLS enabled.
The most important benefits are:
L3 VPN : obviously gives you seperated VPNs, with the possibility of overlapping IP subnets
extreme easy provisioning of layer 3 VPNs
L2 VPN: supports transporting most supported layer 2 protocols, even protocols with clocking information
MPLS TE: steering of traffic through the network and extremely fast traffic protection
These are just the most important ones.
All of these applications run on one common network, while before MPLS, one needed to have multiple carrying networks (e.g. one Frame-Relay network, one ATM network, one IP network) in order to provide all the same functionality.
Regards,
Luc

DFC3 's for MPLS ?

I'm looking for some advice on running MPLS on the 6500 platform. I've got several 6500's with Sup 720 (PFC 3A) which I am looking to upgrade to PFC 3Bs to support MPLS. However, I'm wondering if I should upgrade the modules to DFC3's to support MPLS. I imagine that the answer might be "it depends", but thought I would reach out for advice.
Anyone have recommendations or links that might help with this decision?
Thanks,
Jim

Hello, what line card do you have? DFC is installed in the Line card and you need to check frist. DFC will handle the traffic locally, this is the big benefit that you will have. The Mpps also will increase from 30 to 48.
Best regards,
Cassio

MPLS "VPN" - really a VPN?

I keep reading about the MPLS VPN, including a recent Cisco document. They keep referring to MPLS VPNs, but the closest reference to any real VPN was to "Layer 2 VPNs". What L2 VPN? The best that I could decipher was some possible inference to L2TP and/or PPTP encoding (etc.) - maybe. Or not. WPA is probably the only real L2 VPN tech (other than the rare Ethernet encryptors). Most descriptions of MPLS VPNs actually sound similar to VLANs (which is also L2 tech), since the MPLS seem to be isolating similar to VLAN.

Hi there,
First: Here's a MPLS FAQ for beginners
http://www.cisco.com/en/US/tech/tk436/tk428/technologies_q_and_a_item09186a00800949e5.shtml
When it comes to info on the security of MPLS VPN's, take a look at this document (written by a Cisco-employee):
M. Behringer, Analysis of the Security of BGP/MPLS IP Virtual Private Networks (VPNs), Internet informational RFC 4381, February 2006.
"This document analyses the security of the BGP/MPLS IP virtual private network (VPN) architecture that is described in RFC 4364, for the benefit of service providers and VPN users.
The analysis shows that BGP/MPLS IP VPN networks can be as secure as traditional layer-2 VPN services using Asynchronous Transfer Mode (ATM) or Frame Relay. This memo provides information for the Internet community."
link: ftp://ftp.rfc-editor.org/in-notes/rfc4381.txt
Did it help? If so, please rate it.

In reviewing the Cisco Licensing Guide, there is a license part number for MPLS on the Nexus 7010. Part Number is N7K-MPLS1K9.

In reviewing the Cisco Licensing Guide, there is a license part number for MPLS on the Nexus 7010. Part Number is N7K-MPLS1K9. Is there a corresponding part number for the Nexus 7710?

For the benefit of others, here is what we found. The N7K was hitting the bug CSCtg95381.
Symptom
Nexus 7000 may punt traffic to CPU; so that the traffic may experience random delay or drop.
Further looking, ARP is learned and FIB adjacency is in FIB adjacency table.
Conditions
The problem is caused by race condition. Some hosts have not responded to the ARP refresh sent by
N7k which in turn trigger to delete ARP entry due to expiry. As a result the route delete notification is
sent to URIB from the process. However there is still traffic coming to given IP address as a result the next packet that hit glean resulting in triggering ARP and hope ARP is learnt from the host this time.
Workaround
(s):
Clear ip route < host>.
Not totally explains why it was working for certain client-server combination but yet the workaround is holding well for end-points when implemented.
There would be no host route for the destination server in the adjacency manager on N7K-01. The only thing thats there is the subnet route pointing towards the vlan gateway address. Implementing the work-around, a new /32 route can now be seen in the adjacency manager for the server.
The bug is fixed in releases starting 5.1(5). Planning to upgrade to 5.2(3a).
Regards, Rashid.

Port Channel over L2 MPLS links

Hello.
I was hoping that someone could over some suggestions on best practice or recommendations for configuring port channel to bundle to layer 2 MPLS links that we have.
We have racks in two geographically separated data centres, each rack has a stack of Cisco 3850 switches, and there are two 100Mbps layer 2 connections linking these stacks together. The links are provided by our supplier and runs over their core network (I can get more info on their setup if need be).
I had initially just configured a simple port channel bundling the two ports on "Switch A", and the same on "Switch B". However I have since noticed that one of the ports was put into error disabled state, and I have my doubts that we ever had 200Mbps throughput over the portchannel.
The logs showed:
UDLD-4-UDLD_PORT_DISABLED: UDLD disabled interface Gix/x/x, aggressive mode failure detected
%PM-4-ERR_DISABLE: udld error detected on Gix/x/x, putting Gix/x/x in err-disable state
How I understand is that one of ports recieved an unexpected (or incorrect) BPDU packet, and shut the port down as part of spanning tree process to stop a loop forming.
For the time being I have removed the port channel config and re-opened all the ports, so I believe one of the ports is now in blocking state.
Whilst redundancy for these links is great, ideally I would like to have the links bundled so I also get the benefit of higher throughput.
Any thoughts are greatly appreciated.
Thank you

Hi,
Thank you for your reply and suggestions. I have been reading up on layer 2 protocol tunnelling and I'm not sure if this will work for us. According to these guidelines for the catalyst 3550 (http://www.cisco.com/c/en/us/td/docs/switches/lan/catalyst3550/software/release/12-1_19_ea1/configuration/guide/3550scg/swtunnel.html#wp1006657) it states:
•If you enable PAgP or LACP tunneling, we recommend that you also enable UDLD on the interface for faster link-failure detection.
•Tunneling is not supported on trunk ports. If you enter the l2protocol-tunnel interface configuration command on a trunk port, the command is accepted, but Layer 2 tunneling does not take affect unless you change the port to a tunnel port or an access port.
•EtherChannel port groups are compatible with tunnel ports when the 802.1Q configuration is consistent within an EtherChannel port group.
(maybe its a different and is supported on 3850's)
Also, I cant use layer 3 over this link as I am looking to extend VLANs over the link so devices and virtual machines in each of the datacentre are in the same subnet and broadcast domain.
I will speak with the supplier to see what they are willing to do.
Thanks again.

In our enterprise MPLS network we are using 192.168.20.0/24 subnet, in this subnet we have not assigned the IP 192.168.20.200/30 & 204/30, But still these subnets are reachable . Are these NNI IP ...Please explain.

In our enterprise MPLS network we are using 192.168.20.0/24 subnet, in this subnet we have not assigned the IP 192.168.20.200/30 & 204/30, But still these subnets are reachable . Are these NNI IP ...Please explain.

I have checked with ISP, there response is like below:
Those are the NNI to GBNET IPs for Dominican Republic. They are Network IPs. You should be able to ping them-that means they are working.
WANRT01#show ip route | include 192.168.20.20
B        192.168.20.200/30 [20/0] via 192.168.20.226, 02:18:29
B        192.168.20.204/30 [20/0] via 192.168.20.226, 02:18:29
Here its shows from any of our MPLS site we are able to trace the IP and it seems like, 192.168.20.204/30 is one more site but in actual its not.
INMUMWANRT01#ping 192.168.20.205
Type escape sequence to abort.
Sending 5, 100-byte ICMP Echos to 192.168.20.205, timeout is 2 seconds:
Success rate is 100 percent (5/5), round-trip min/avg/max = 224/232/260 ms
INMUMWANRT01#trace
INMUMWANRT01#traceroute 192.168.20.205
Type escape sequence to abort.
Tracing the route to 192.168.20.205
VRF info: (vrf in name/id, vrf out name/id)
1 192.168.20.226 24 msec 24 msec 24 msec
2 192.168.20.206 [AS 8035] 232 msec 232 msec 252 msec
3 192.168.20.205 [AS 8035] 224 msec 224 msec *

How do I know if my MBP will benefit from heatsink paste reapplication?

Dear Mac Users
I know the theme of hot MacBook Pro's has been done to death, but I would like to add my two pennyworth with respect to the specific question: How do I know if my MBP will benefit from reapplication of the heatsink compound? I have trawled numerous very long threads here and not found a satisfactory answer. The reapplication of the heatsink paste is a bit of a schlep and not without some risk of breaking it, so it would be good to know if the benefit is worth the risk.
One common question is: "Are my MBP temperatures unusual?". A useful mac temperature database can be found at the following link:
http://www.intelmactemp.com/list
Consulting this list should set some parameter space for many users - there does seem to be considerable variability - suggesting highly variable quality of heat sink paste application by Apple. In my particular case I have an early 2011 MBP (i7, quad core, 2.2GHz). This certainly ran very hot. I even had a burn on my thigh from it, which took a while to heal. I know, they are notebooks, not laptops, but still . . . . To put very hot into perspective, the CPU was sitting at over 60C at idle, and whole case would be hot to the touch, such that resting your hands on it was uncomfortable. And yes, the computer was idling, there were no rogue processes chewing up cycles. At full load, the CPU would run at 90C and the whole thing would get very hot. I avoided running major projects or rendering jobs on it to preserve it from heat fatigue - the warranty is now up.
My first attempt at reducing the temperatures, was to clean it. The clue was in the USB/Firewire etc ports. All were very fluffy, which suggested that there was a lot of crud in the heatsink fins. Fixing this was simple; adopt anti-static precautions; remove the back panel; blow air into the exhaust vents at the base of the screen. Numerous large fluff bunnies popped out of the fans and elsewhere. Alternatively, you can unscrew the two fans and partially lift them out (without removing the power cables to the motherboard). Major fluff on the heatsink fins can then be tweezered out, prior to blowing it through, to ensure all debris is removed. I did this subsequently on my Core 2 Duo MBP. After defluffing the idle i7 CPU temperatures dropped from 60C to 43C, but perhaps of more importance was that the whole case was much cooler and the keyboard temperature was close to ambient - much nicer to use.
So that fixed the major heat problem. However, the question remained, would I see any further improvements by stripping the thing down and reapplying the heatsink paste? What I was looking for in this forum was a rule of thumb based on temperature measurements to help me make an informed decision on whether to strip the thing down. As my machine was out of warranty, I wasn't worried about voiding that. Also, where I am there is zero support (Africa), so Genius Bar or similar was not an option. But jiggering a 14month (very expensive) laptop was not something I wanted to do for thrills. The data in many of the posts did not provide any answers. In the end I made some measurements and decided to do the fix. The step by step instructions I followed were very clear and can be found for many Macs/models at:
http://www.ifixit.com/Device/Mac
This guide made life a lot less stressful, as pulling a connector the wrong way could trash the motherboard, and those connectors are so very tiny and fragile. It is very important to take it slow and follow each step to the letter (and not skip steps inadvertently). My before and after heatsink paste fix temperature data are shown below. The idle temps are improved, while the full load temps look like a marginal improvement. However, CPU temperature is not the only story. Certainly after the fix, the whole case is generally cooler, so the heat extraction system is working better. Also, the CPU before and after temps are quite similar at full load, but that assumes the CPU is running at the same power. I do not know enough about the i7 architecture to say if the before speed was being limited by heat, while in the after case the turbo-boost was able to run out to the maximum. Unfortunately, I didn't do any CPU performance tests. So, based on temperatures alone, it looks like a marginal improvement, but it is certainly making a difference to case temperatures.
Note all temperatures were measured at an ambient temperature of 20C on a flat, hard surface using Marcel Bresink's Temperature Monitor. Fan speed was measured using the Fan Control software - set to allow firmware control at idle.
Apple original heat sink paste - big dollop of grey crud and lots of extrusion at the sides.
Idle
CPU 43C
GPU 38C
Heat Sink 2 36C
Heat Sink 3 35C
Fan Speeds 2000rpm
Full Load (all cores maxed with a Boinc Distributed Computing Project (Rosetta - the project, not the Apple technology).
CPU 89C
GPU 57C
Heat Sink 2 52C
Heat Sink 3 49C
Fan Speeds 6200rpm
Arctic Silver heat sink paste
Idle
CPU 38C
GPU 34C
Heat Sink 2 33C
Heat Sink 3 32C
Fan Speeds 2000rpm
Full Load
CPU 86C
GPU 59C
Heat Sink 2 50C
Heat Sink 3 50C
Fan Speeds 6200rpm
Back to my original question: How do I know if my MBP will benefit from reapplication of the heatsink compound? I think temperatures alone will not give a good answer - mainly because the CPU probably has variable output - due to turbo boost. One thing I did notice is that the temperature response of the CPU when it gets switched to full load, does vary according to the quality of the heatsink paste. With the original Apple paste, when I switched the CPU to full load (from idle) the temperature of the CPU went up to its maximum almost instantaneously and stayed at that level ie the graph of temperature vs time was a step function. This suggests that the CPU may be controlling the temperature by throttling itself. Otherwise, as the heatsink warmed up, the CPU should also rise in temperature - mine stayed bang on 90C and fluctuated only by a degree either side. After application of heat sink paste, the CPU has a much larger effective thermal mass and consequently heats up much more slowly. Turning the CPU up to maximum from idle resulted in the temperature climbing to a maximum over about 3-4s. After peaking at around 92C it dropped back as the fans kicked in, to around 86C.
So perhaps one way of assessing the quality of your heatsink paste is to ramp the CPU to maximum from idle and look at the shape of the temperature profile. A step function suggests a lousy job and benefit may be had from reapplication. A more gently sloping profile, followed by a dip due to the fans kicking in, may suggest you are in good shape. Obviously, this presupposes that your Mac temperatures are not insanely hot to start with (i.e. high 90sC+). In which case, if defluffing doesn't do the job, then new heatsink paste is almost certainly required.
Regards, BB

I'm sorry but this is too funny to pass up.

Error in updating benefit plan IT0171...for CE

Hi All,
Presently in SAP, the benefit main assignment for concurrent employee is done using the transaction for benefit workbench.
I need to updated Infotype 0171 for concurrent employee from my Z-program. But here I am facing an eror which is given below:
1. I am change the data for IT0171 using HR_INFOTYPE_OPERATION (eg. CE 40, has 2 assignments pernr 40 and 52). Now the default in the benefit workbench is 40, so when i am changing assignment from 40 to 52, I am passing BENGR and BSTAT from 40 to 52 and trying to update 40 with blank value.
Bt here i am getting error in IT0171 for pernr 40, tht BAREA, BENGR and BSTAT cannot be updated with blank value (whereas the same happens in benefit workbench)
2. I am not able to solve this coz I am using HR_INFOTYPE_OPERATION in my program and this FM is also used in HRBEN00CEWB.
PLease help
Edited by: abhay chauhan on Nov 12, 2008 1:22 PM

Hi,
Try having a look at this document
E-WL: Message From the ISAPI Plugin: "Cannot read 8192 bytes of postData from tmp file" [ID 664043.1]
https://support.oracle.com/CSP/main/article?cmd=show&type=NOT&doctype=PROBLEM&id=664043.1

Will I Benefit By Upgrading Video Card Only? or Complete New Build?

I, too, was wondering if I might benefit from upgrading only the graphics card. My current system:
Windows 7 Ultimate 64-bit SP1
CPU
   Intel Core i7 2600K @ 3.40GHz 43 °C
   Sandy Bridge 32nm Technology
RAM
   32.0GB Dual-Channel DDR3 @ 799MHz (10-10-10-27)
Motherboard
   ASUSTeK COMPUTER INC. P8Z77-V PRO/THUNDERBOLT (LGA1155) 37 °C
Graphics
1535MB NVIDIA GeForce GTX 580 Amp! (ZOTAC International) 44 °C
Storage
CACHE & PREVIEWS 111GB ATA KINGSTON SH100S3 SCSI Disk Device (SSD) 34 °C
OS DRIVE 238GB ATA Samsung SSD 840 SCSI Disk Device (SSD) 35 °C
CAPTURE STORAGE 16 TB Intel Raid 0 Volume SCSI Disk Device (4 x 4TB) 34 °C
     DNxHD 220 Footage
     DSLR Footage
     AVCHD
   Optical Blu-ray Drives
   HL-DT-ST BD-RE BH16NS40 ATA Device
I was shooting HDV when I built this system, but am now shooting with a Canon EOS C100 and a 5D Mk III DSLR. I record DNxHD externally onto Atomos Ninja Blade.
Although this system seems pretty well balanced, I have never been able to do much more than basic editing with Premiere's real-time effects, since anything else (like Noise reducers, MB Looks and Red Giant effects) pretty much cripples my system playback, rendering, exporting. Even with HDV footage I was never satisfied with performance. I recently upgraded my storage from a 2-disk RAID 0 to a 4-disk 16 TB RAID 0, which seems to have helped transfer rates a little.
I wonder if the GTX 580 is defective. It seems to run pretty hot under load at 60 degrees or more unless I boost fan speed to the max. I cannot change any GPU clocks speeds or other settings from default without the display crashing. That is why I am considering a new GPU. I plan to build or buy a new PC within the next year, anyway, but it would be great if a GPU could extend the life of this one.
As for a new complete built, I need to know if now is the time to do it. Or, is there some new motherboard/processor series on the horizon that I should wait for? The frustrating thing about these kinds of forum threads is that the information is quickly dated and obsolete.

I am hardware accelerated.
As for the GPU temperature. The NVidia control panel monitors voltages and temperatures, but gives no reference for what the normal operating temperature should be. How do I know if the GPU is running too hot? The same goes for the rest of the components. I'm sure you professional computer builders have that kind of info available, but how do the rest of us know when something is wrong with a component? Is there a troubleshooter I can download? or a chart that shows a normal range of numbers?
When I first built this computer 3 years ago or so, I had a GTX 470, but it fried a year later and I replaced it with the GTX 580, which gave me a good boost in performance over the 470. It would seem that a 700 series GPU would give me better performance over the 580 - with a vast in increase in CUDA cores and more RAM, etc. But, I have read in this forum and other places that a newer model GPU alone will not make much difference, which makes absolutely no sense. If that's true, then what's the point of having all those additional CUDA cores?

Cannot find source code for Employee Benefit application

Hello,
I am trying to find the Employee Benefit Sample Application which used to be provided by Oracle at http://otn.oracle.com/sample_code/products/ias/content.html. Unfortunately I couldnât find the source code anywhere, neither by searching through oracle sites, nor through the internet.
Could someone please help me get it, since I just started working on applications based on this model and it is really important for me to have a complete example that works.
Thanks in advance,
Stefanos

Hello,
This sample application used to work on older versions which is no longer supported. This is the reason, it has been removed.
Thanks.

MPLS Benefit???????

Similar Messages

Maybe you are looking for