Monday, August 3, 2020

VPLS.US 05/06/2011 Chambers Listened to VPLS.US

http://vpls.us/?p=672 Fri, 06 May 2011 15:09:25 +0000 timc
http://vpls.us/?p=672

wall street Journal they announced


“Cisco to Reduce Its Bureaucracy”


Nice… About 5 weeks ago, I published a blog saying that John Chambers reads my blog! I recounted a timeline that made me think that John had read about my advise to Cisco. Well, I still feel that they had read the blog. Today Guess what?!?!?! It happened again. The Wall Street announced that Cisco was going to start reducing bureaucracy.

At the end of my timeline I offered advice to them about what else they could do to become profitable again. I said to get rid of the committee system. Reduce the management overhead. I used General Electrics management style as an example of how the CEO needs to take charge and also have a right hand man that he can trust and allow to make decisions. So what is Cisco’s new direction????


  • Get rid of the committee’s
  • Divide authority up to regions
  • Give more authority to regional people

So, now I guess it’s kind of my responsibility to give some more direction… This is gonna hurt.


John, I know you are trying to protect your employee’s… It’s honorable. In fact if I was working for you, I’d really appreciate your direction. It’s the way the country used to be ran. Employee’s are like family. You treat them with respect and you protect them. I know it hurt pretty hard to lay off the 550 people in the Flip consumer video camera business. But you are going to have to get used to it. The rest of the industry has gone cut throat. In order for you to be competitive in this industry’s landscape you are going to have to become cut throat. As an employee I never want to recommend that people loose their jobs, but if Cisco is going to survive you are going to have to start cutting the fat.

Here is a question you need to ask yourself. How many of my employee’s are millionaires? so when a millionaire is working it’s either because he’s just comfortable in the job or passionate about his job. If he’s passionate. keep him, if he’s comfortable, he’s probably just hanging around because he’s too young to retire, or not motivated enough to start out on his own. Either way, he’s not the best employee.

When you bought companies, you pretty much operated them independently. How many billing systems do you have? How many sales people do you have that don’t have any idea how to sell your core products? Your 6 new lines? Should a sales guy be specialized? NOT. Should a SE be specialized? probably. Having technical people that know everything about everything is probably unrealistic. However having sales people that cannot sell a router? that’s a hoot.

So as tough as this might be, you are going to have to do some major restructuring and laying off of staff.

Trust me, as a technical person, who might one day come to you for a job, the last thing I want is a flood of technical staff out in the market looking for jobs. However, I’m guessing that the people you need to lay off are not competition for me, and they probably not benefiting you either. When my company went into bankruptcy we cut into the fat. At one round of lay offs we went from 25,000 people to around 15,000 cutting more than 40% of the company…. Guess what. everything stayed running. We hardly noticed. Yeah, we had huge stacks of routers laying around and huge stacks of servers laying around. that nobody was using any longer… But everything stayed running. No customers left, and nobody was really unhappy other than the 10,000 employee’s that where no longer employed.

Great start. Let me know if I can help.
]]>

VPLS.US 04/30/2011 Ethernet VPN was MACVPN

http://vpls.us/?p=660 Sat, 30 Apr 2011 00:04:23 +0000 timc
http://vpls.us/?p=660

I've heard it called the "next new thing"

I need to cover :


  • EVPN Architecture and Benefits
  • Pros and cons of EVPN versus VPLS
  • Pros and cons of EVPN versus OTV
  • Pros and cons of EVPN versus PBB and that whole debacle
  • H-EVPN (hierarchical)
  • EVPN Scaling / Resiliency
  • evpn-NNIs
  • EVPN to 4761 VPLS Interconnections
  • EVPN to 4762 VPLS Interconnections
  • A couple enhancements / changes I would like

Probably a whole bunch more stuff before the first release of EVPN comes out. I wonder if EVPN.US is available. :)

So, just from a preliminary review of Ethernet VPN (EVPN) I can tell you that if some of the things I’ve mentioned above are worked out, then yeah this could be the next big thing. It’s going to address and solve a lot of the issues that we have been facing with VPLS. From what I understand it’s basically a complete rewrite of what a Layer 2 VPN should look like. I’m looking forward to start digging under the covers of what these guys are proposing.

draft-raggarwa-sajassi-l2vpn-evpn-02.txt

Should be a good read. :)]]>

VPLS.US 05/05/2011 Ethernet VPN to RFC 4762

http://vpls.us/?p=658 Thu, 05 May 2011 01:19:25 +0000 timc
http://vpls.us/?p=658

I could see a couple different ways of addressing this issue:


  1. Through the use of NNIs. Which by the way is something I’m seeing as missing from the Ethernet VPN draft. The ability to cross different AS realms. Having a link between two PEs become the path between a BGP signaled Ethernet VPN and a LDP Signaled VPLS domain.

  2. Through the use of something like Juniper’s option E… Having a PE that can speak both Ethernet VPN and LDP VPLS and having it keep two LARGE forwarding tables that allows it to do the forwarding between the different domains.

  3. What else???

This is going to have to be addressed, or they risk another fraction as we saw between 4761 and 4762 or perhaps that’s what they want?
]]>

VPLS.US 05/06/11 MAC NAT

http://vpls.us/?p=656 Fri, 06 May 2011 02:22:10 +0000 timc
http://vpls.us/?p=656

So, I was reviewing OTV and MACVPN and Ethernet VPN and VPLS, and had an idea to reduce some of the overhead that's associated with L2 boundaries. Today, Layer 3 boundaries represent the boundary between a Layer 2 domain and a Layer 3 domain. Why couldn't one of today's switches do this? They are looking into the packet for TOS / DSCP information which means that a switch is already looking way into the layer 3 header of the IP frame. So why couldn't every Source MAC coming from a particular location be hidden behind the MAC address of PE?

This would reduce the MAC-address-table of the PE and reduce the load and overhead of the whole system.

Anyway that's an idea. It might have some issues with resiliency ]]>

VPLS.US 04/22/2011 Consolidations

http://vpls.us/?p=631 Fri, 22 Apr 2011 08:01:15 +0000 timc
http://vpls.us/?p=631

  • Anda and Telco Systems
  • Hatteras and Overature.

    I did an RFP for all 4 of these corporations in reviewing our requirements for a NID, Network Interface Device. Plus a couple more that are still on the market.

    • Sprint & Nextel
    • ATT & TMobil
    • Global Crossing & Level 3

    Consolidations in the marketplace have been needed in the carrier market for a LONG TIME. I think everybody was just wondering when it would happen. Well we know about the ISP / carrier market now. I still suspect that a couple more will happen.

    On the NID / smart demarc for Ethernet? Well when I went to RFP we had 8 different companies that we were reviewing. EIGHT. That’s too many. Most of them did the same thing. One of my biggest concern 4 years ago when we went through the exercise was: Which of these companies are going to be economically viable 5 years from now :) Guess what we are about 5 years out and there is justifiable concern over the boxes. One large Carrier in the southeast was using Hatteras. I wonder what they are thinking now. The vendor we picked wasn’t my choice, but the project got moved away from me so I didn’t have the final say so on direction. But our choice is still around, I’m not sure the guy who took over the project put as much weight on company financials and viability as did I, but that’s probably another story.

    I don’t suspect our vendor will be around much longer as it exists today. With this consolidation starting, it won’t be long before somebody larger will pick them up and their customer base. I just hope the one we picked does the buying rather than the getting bought. It usually ends up with equipment vendors that the buyee’s products goes away.

    But then again almost 4 years ago, I rambled about the NID market going away entirely as the functions were integrated into the customer based premises gear. Well, I guess that hasn’t happened or isn’t going to happen. Why? Well the main reason I see is that the carrier market likes to have a piece of gear that’s their own at the premises. It gives them a *KNOWN* good piece of equipment to test against versus trusting that the customer has enough knowledge and experience to place a device with NID capabilities at the premises

    So I was wrong in my predictions about the NID market. Or maybe I”m not wrong I just predicted the demise too early? It was a relatively long germination period for Ethernet services to get rolled out as widely and as ubiquitous as the market pundits predicted. Here we are 1011 and the Standards have been out for 6 or 8 years now. It’s just now hitting stride and appearing to make money. Yeah I know we are still in the first adopters phase. I just wonder if it’s going to go gangbusters like they predicted?

    Anyway, I can predict this. Stay tuned for more consolidation. It’s inevitable. ]]>