Image courtesy of Wikipedia
Whenever I talk to people who are just getting started in networking, there’s a part of me that wishes I was in their shoes. I say that because I know several of the things they are going to learn or figure out in the next couple of years and I remember having to go through the same process. Before I understood variable length subnet masks(VLSM), the numbers in the subnet mask field of a workstation’s TCP/IP settings didn’t really mean a whole lot to me. If someone used slash notation(ie /24, /16, /27), I had no idea what that meant. Like a lot of people, I relied on someone to tell me what the subnet mask was. However, once I learned about VLSM, it was as if a whole new world opened up. That was one of my absolute favorite “Ah-ha!” moments. You’ve had those yourself haven’t you? It is the point in time in which a certain technical concept just clicks in your head. You go from not really understanding it, to comprehending it. In fact, it’s almost as if that concept is only represented in binary inside your head. You go from a 0 to a 1 with no in between.
As you progress along in networking, more and more of these “Ah-ha” moments come. Unfortunately, over time they become fewer and fewer. That’s not to say that they go away completely. They don’t. They are just harder to come by. I’ve found that I am able to keep a steady stream of these “Ah-ha” moments coming as long as I look at technology without taking anything for granted. What I mean by that is that I don’t assume anything when it comes to trying to understand a protocol or technology. What I “think” I know might actually be wrong. My understanding might only be partial. I have to continually ask “why/what/how/when/where” when dealing with technology.
Let me give you a personal example. I have known for many years that a T-1 is 1.544Mbps in terms of bandwidth. It is comprised of 24 64kb channels. The only problem is that 24×64,000 is 1536000 and not 1544000. Oops. Where did the other 8k go? To further drive this home, a “show interface” on a serial link that is configured as a full T-1 shows the interface bandwidth to be 1536kbps. Why the discrepancy? I could have just moved on and ignored the reason behind the discrepancy. However, by researching the issue and figuring out what the issue with this discrepancy was, I learned a whole lot more about T-1’s. I learned how alarms over the circuit get propagated. I learned what the extended super frame(ESF) actually was. In other words, had I not been curious as to why the math didn’t add up when it came to T-1 bandwidth, I would be far more deficient in the inner workings of the T-1.
In the spirit of chasing the “Ah-ha” moments, take a look at the 4 questions below. Go find the answers if you don’t already know them.
1. Why is MPLS faster than conventional IP based routing?
2. What are the differences between a multi-layer switch and a router?
3. Why do you need different antennas for wireless access points and where would you use each antenna type? Sure, this is rather open ended, but what I am getting at is the radiation pattern of each antenna.
4. How does traceroute really work? Not just the TTL mechanics, but look at the various ICMP type codes as well.
Can you remember the last “Ah-ha” moment you had? If not, why? If so, does it make you want to go out and find more of those moments?
These Ah Ha! moments drive me through my career. It’s almost like I’m standing in the middle of stage and a giant spotlight suddenly turns on. Then everything gets clearer. I figure whenever I stop having these kinds of moments, I’ll be done in networking/IT because my desire to learn will have been long gone.
As for my last Ah Ha! moment, it was writing my followup BPDU post. In the middle of a long paragraph, I glanced over at a comment and it was like a lightbulb flashed on. Everything suddenly made sense. I furiously typed out everything I was thinking so I could commit my epiphany to eternity before my thoughts were replaced by something more mundane, like what a doorbell sounds like.
Actually, I had the same question years ago, when i kept hearing about T1’s and DS1’s. I wanted to know the difference. That’s when I figured out where the extra 8k went. LOL. It’s funny how one question, leads you to so many more!
I think my last ‘Ah-ha’ moment, was when you were trying to explain to me how the WAN Accelerators worked. Then it all just clicked! ((grin))
I think my most recent ‘ah-ha’ was around the bandwidth-delay product and how the latency of a connection will ultimately affect the maximum bandwidth you can eke out of an individual TCP stream. Totally agree though, when you learn something and it just clicks – it’s a good feeling! 🙂
Great Post! Just the kick-in-the-pants I’ve needed after these last few weeks. One of the last Ah-hah moments I’ve had was understanding the differences between the various OSPF LSA types. I do agree though that the VLSM *click* is one of the biggest light bulbs in networking.
One of my most memorable a-ha!s was like 2 hours before taking CCNA exam, when colleague explained summarizing to me the way I finally understood it. It’s amusingly simple, but I just was not able to get it to my head by reading numerous books and howtos.
Another I remember was when I saw document lying on shared drive talking about turning up E1/T1 service showing various ping test patters we use when testing line. Given what I learned about digital circuits, it finally clicked in my head why some patterns (0x00, 0xFF, 0x40 etc.) are more significant than just random traffic.
My latest one was listening to Packet Pushers at the gym and they were talking about VPLS. My ah-ha was relating it to my own situation and realizing I could use GRE tunnels over my providers MPLS network to get around relying on them for routing.
Mathew nice article! though I am yet to find a experiment that says MPLS is faster than traditional IP routing.. especially when you are using CEF for packet forwarding these days on IP routers some say any advantage label lookup and forwarding can give is negligible. Where else do you think the speed comes from for MPLS…. hey that just can be a aha moment 🙂
” 1. Why is MPLS faster than conventional IP based routing? ”
Bad example. In any modern router there is no speed difference.
Thanks for pointing that out. Vin also mentions that in his comment from December 24th. I wonder if a better way to frame the question would have been: “Why would MPLS be preferred over traditional IP routing?”