Packet switching Puns

What would it take to make a robust and effectual packet-switching network protocol that has 256-bit addresses which contain no topological information?

The idea is that nodes will hash a public key to derive their addresses, allowing for nodes to prove their identities. Since spatial distribution of addresses will be random, routing information will have to be empirically derived. Additionally, there will be no central governing bodies controlling who gets what allotment of the address space. It would be a fully decentralized and self-organized network layer 3 protocol that could switch packets over any underlying layer 2+ network layers.

There are several problems to solve with this type of network:

1. How to apply distance vector or link state routing to such a network.
2. How to conserve memory when generating routes.
3. How to conserve bandwidth when sharing routing information between peers.
4. How to structure a packet such that it has sufficient information for cross-layer routing.

Any ideas on how to solve any of these problems would be helpful.

Here is a free PT lab that can be used to practice switching and VLANs, including a little troubleshooting.

Base lab can be downloaded from https://theKeithBarker.com

Happy studies everyone.

Hey Everyone,

I recently switched a UDM (non-pro) and am experiencing extreme packet loss while online gaming. No one else is using any of the network (400 down 30 up) and my games just completley locked up.

I disabled all beta features (DPI etc) and nothing is fixing the issue.

I switched back to my old router (some crappy asus one) and the packet loss issue does not happen.

Has anyone experienced this? Should I just reach out to UI and do an exchange?

Thanks

Hi All,

New to the community.  After an acquisition of a small office, I’ve recently inherited a DELL network environment. I’ve always worked in Cisco shops, but the past few years I've focused entirely on voice/video. Our sr. network engineer just quit, so i'm back in the hot seat.  The previous sysadmin of this new site left during the transition, so I’m doing the dance of discovery/support.

After migrating the site from onprem VOIP to a cloud PBX(last week), we started getting complaints of call quality.  A few pingplotter scans to the cloud pbx showed all our packet loss occurring on the LAN side.  after some more testing I isolated the packet loss to the user VLAN . ( pingplotter on a local server hit the internet with no packets dropped).  The VLAN gateway on the core switch (s5000) seems to consistently drop packets (3-5%), which is killing voice quality.

After a quick call to DELL I found the switches are out of warranty, so a call with support would be BIG $$$.  I decided to kick the cores after business hours to try and shake things out (this was an hour ago). It's worked before! Unfortunately, packet loss started popping up again 10 minutes after the switch reload.

Feeling a bit defeated, but I have tomorrow morning to troubleshoot further, as I’m ET based and this office is on the west coast.  I’ll be calling DELL tomorrow, but wanted to see what the community thought.  I’ll be posting some config files once I hit the office; sorry I know it’s a party foul to ask questions without giving you guys hard data.

The mile high environment...

MDF

• S5000- core
• 3048
• 3148

IDF

• 3048 - fiber run to S5000
• 3148

while the network has a small footprint, the only oddity is that internet comes into the IDF and hits a sonicwall FW then the L3 3048 switch.  This traffic is then routed to the MDF S5000 switches.

I finished the night by starting fresh pingplotter scans.  One PC serviced by the MDF access switches, the other PC by the IDF.  My thought being,  if one comes back clean and one doesn’t, I’ll know to focus on either the IDF 3048 or the MDF S5000 ( from a routing perspective)

Any advice would be huge, sorry for the wall of text!!

I'm just wondering as latency from Dubai to Johannesburg is changing every 10-15 minutes from around 130 on the low end and around 200 on the high end, is anyone else having latency issues to Dubai from South Africa and is this something to do with the AAE-1 cable being upgraded, this started happening from 0800hrs UTC today.

I'm a SWE with 1 year experience working at a large network devices manufacturer (think Cisco, Juniper etc.). I'm on a platform team, working closely with the switching ASICs on our ToR switches. I've been getting more work in the areas of Packet Processing and Traffic Management lately, and was wondering if these are reasonable areas to try and specialize in for my career going forward. This requires a lot of general networking knowledge as well as an in-depth understanding of the hardware architecture and specific chip revisions. Pretty much I think this is really interesting stuff but I'm a bit concerned that it might be a bit niche, and that the large investment I made in learning how these systems work in the detail required would only help me if I work for a small group of large networking companies for my whole career.

If anyone with experience in these areas could give some advice, that would be much appreciated.

I recently switched from Verizon FIOS ISP to Optimum Altice (Cablevision on Long Island). I actually went from 300mbps to 400mbps when I made the switch. Bear with me as I am not very knowledgeable in this area but I'm trying. I have my Xbox set up in my basement, and the main cable box/router on the main level. I have a wifi range extender in my basement next to my xbox in the basement, and have been connected to that. Note that the wifi range extender is also a cable box. Each optimum cable box has the capability to act as a wifi extender, and this function must be turned on manually.

The connection is good but not great, I have roughly 60ms ping. I experience minor lag from time to time while gaming so I sought to improve this by running a 100 foot CAT 6 Ethernet cable to my xbox from the main router. I did so, and my ping dropped to <30. Now, in some online games, my connection is perfect, however, in others, the lag is way worse than when I was just connected to the wifi extender. I believe the explanation here is packet loss? Unfortunately, the game I have been playing most (apex legends) has major lag issues when wired (even though ping is low) so I reverted back to my wifi connection via the extender for now.

Does anyone have any recommendations on how to improve packet loss on my wired connection? Or is there an explanation other than packet loss?

Note that the wifi extender was still on and projecting the signal right next to my xbox while I was using the wired connection. Could this severely hinder wired performance in some way?

Thanks in advance.

When creating a port mirror on a FortiSwitch, what does "switching-packet" mean? (ref https://docs.fortinet.com/document/fortigate/6.2.3/cli-reference/177620/switch-controller-managed-switch). I understand what a layer 2 switch does but what does this mean in the context of a mirror? Unsure if I should enable it or not. I want to sniff for IDS.

Hi All,

My FiOS router recently died and decided to switch from the MoCA connection to ethernet via the ONT. The router I am using is the TP Link AC1750 Wireless Dual Band Gigabit Router. However, after the switch I'm experiencing high packet loss. When I run ping tests, I notice high spikes in response times even when it is just my device on the network. Can there be something wrong with the ONT ethernet port? I also ran CAT 5e cable to my router. Can the cable also be an issue?

Thank you for any help!

Packet tracer seems to be lacking a ton of commands so I downloaded gns3.

I had the image for the c7200 and got everything configured correctly thanks to the guide, but I don't know how to configure these switches, I have some images I procured but am having trouble and feeling really discouraged, any help would be greatly appreaciated.

Wanna hear a

I am learning basic networking and I got a question below.

In packet switching, simple determinstic queue model:
Assuming fixed Link Rate R,
Q(t) = A(t) - D(t)
where A(t) is cumulative number of bytes arrived up until time t, D(t) is cumulative number of bytes departed up until time t.
D(t) can also be expressed as time spent in the queue by a byte arriving at time t.

Question:
In a store and forward router, every second, a 100 bit packet arrives to a queue at rate 1000 b/s. The maximum departure rate is 500 b/s. What is the average occupancy of the queue?

Solution: During each repeating 1s cycle, the queue fills at rate 500 b/s for 0.1s, then drains at rate 500 b/s for 0.1s.
Over the first 0.2s, the average queue occupancy is therefore 0.5 x (0.1 x 500) = 25 bits.
The queue is empty for 0.8s every cycle, and so average queue occupancy: Q(t) = (0.2 x 25) + (0.8 x 0) = 5

Why average queue occupancy is 0.5 x (0.1 x 500) = 25 bits? 500 should be the queueing rate,0.1 should be the time required for a packet arrived to the router completely, but why times 0.5?

I guess 0.5 means 1/2 which is a representation for 0.1 is half of 0.2. But I am still totally lost on what happened.

Please suggest what are math components of the answer.

Please explain, thanks!

Hey, I'm trying to ping between 2 vlans, 110 and 120.

There are 2 layer 3 switches, and 1 switch. Here's the config

Layer 3 switch 1: https://pastebin.com/0zejwxki

Layer 3 switch 2: https://pastebin.com/0QUnV6cf

Switch: https://pastebin.com/aBTa7GCg

The vlan 110 PC is at the end of fa0/6 of switch 2, and the vlan 120 PC is at the end of fa0/6 of switch.

If there's any more info you need from me, let me know

I feel that I am giving myself enough time to get this cert out of the way. I left school for Health reason and felt that I wasn't learning as much as I wanted to. anyone have any tips ?

I just saw some people recommend l Lammle ccna routing and switching book over odoms book. Which one do y'all suggest?

I'm looking for some resources on switching a small packet of cards. I do a color monte based routine involving the court cards, and I'd like to add a kicker at the end where the cards I'm holding change into the 4 aces. I was wondering if anyone could point me in the direction of some literature that may help me out. Typically this is done standing, and as such a tabled cooler is not really an option.

So I've been reading Networks: A top-down approach, and supplementing with class and on-line, but as often as the term packet-switching is used, I do not clearly understand what the switching part means. I understand the concept of a packet being transmitted, and that data is re-pieced at the receiving end, but what does the switching part actually mean? What is a packet switch/what is a link-layer switch?

Also, as far as protocols, is it correct to see protocols as a format of transmitting data that is then understood at the end/receiver and can thus be used by, say an application, to know what way to piece the data back together as they know the protocol used. Likewise for lower level protocols, would it be that it is already embedded in the circuitry for data to be sent in a certain format?

I’m not a computer science person, really just learning about networking, the like. But, I was wondering, if one was using packet switching as a method to send a single piece of data. In other words, they are essentially creating one message to get to a destination, and duplicating it but each one with different identifiers showing the route it needs to take/the next hub it needs to hit....so that it is sent through a network at different nodes/hubs, what would be the best way to garantee or make the odds high that that one message will be received at the destination...given that each hub works only fifty percent of the time. So they can either forward the data packet or not. Essentially, the hubs probably of sending are fifty percent sending and fifty percent not sending...

This is probably pretty confusing, but if anyone can help me out or point in the right direction, that would be so much appreciated.

Thanks!

Hey all,

Ill start with the problem:

With a user population of M users, give a formula (in terms of p, M, N) for the probability that more than N users are sending data.

Now, I understand that there is a binary outcome to the "trials", i.e. sending or not sending. But I utterly lack an intuitive understanding of scenarios like this- I understand I could just jot down the formula for a binomial distribution, but I am wondering if any of you folks can explain to me why the formula is the way it is. Please let me know if I need to clarify at all. Thanks!

I've been reading through this CCNA book, and I came to wonder about a few things relating to Packet and Circuit Switching. Is circuit switching only for phones? If not, then wouldn't it be ideal for VOIP? Can circuit switching operate with VOIP? Thank you.

There's lots of esoteric explanations out there but not much else...

Can anyone give me a good and simple explanation of packet switching?

What else can cause packet loss and how can I confirm what the faulty part is?

edit:

was that there was no problem--pingtest was the only test that showed packet loss, apparently. All other tests showed a normal connection. This might be because my router can't handle all 250 packets at once. Thanks to homer2320776 for helping me figure this out :)

edit:solved by switching internet providers lol

Hi guys, I have a slow Comcast internet connection at the moment (3mb/s) with Motorola surfboard sbg6580 wi-fi cable modem. The internet was set up about 5 years ago.

My problem is, I'm having terrible lag spikes and packet loss every 20minutes, sometimes even more frequent than that. I can't change my connection atm, because I'll be moving in a month and it would be a bit counter-productive.

I was thinking, maybe switching the modem would help? I've read some forum posts on comcast's website and it helped for some people. The modem seems to be a bit out-dated too.

I'm a gamer and I mostly play games with severe death penalties, where 2 second lag or packet loss can cost me 200 hours of gameplay (diablo 3, tibia, path of exile, etc.)...

Thanks!

edit: messed up the title, it was supposed to be "Question regarding lag spikes, packet loss, and switching modems."

I'm using the Happymodel JR size 2.4g Tx and an assumedly hglrc branded ep1 that was pre installed in a BNF Rekon 7, so I flashed it to the same version as my TX module which is v1.0 which has been fine for 2 months, no problems.

So just kinda out of the blue I decided to change packet rates, (my output power has been the same at 250mw since I set it up) so I switched it to 150hz,on the ground and a power cycle in between. So after I take off and get about 50m away in open air I get an audio alarm from my TX16s that says "Sensor lost" over and over pretty much regardless of how close I am. Also at the beginning of this I switched sd cards in the radio back to the original one with the sound packs and I think it had a different version Elrs lua script than previous, maybe that's affecting it also it seems like when I'm in the ELRS lua script the kinda random numbers at the top don't change to what I set the packet rate to like I think they did before.

I hear "telemetry critical" and "telemetry lost" when I plug-in and unplug the quad, which seems normal.

Does anyone know what"Sensor Lost" means? Has anybody encountered this warning

I was given the Pearson routing and switching simulator for my birthday, if anyone else has had a chance to play around with it could you care to comment on how you found it? I only got it a couple of days ago and from what I see it's pretty decent, not sure if I should focus solely on this or use packet tracer too.

When you open the packet tracer there are no devices there to do the activity and creating devices is locked, anyway i can get the packet tracer activity that is supposed to be there or do anyway have a good copy of it?

~Thanks

I’m not a computer science person, really just learning about networking, the like. But, I was wondering, if one was using packet switching as a method to send a piece of data through a network and the packets themselves are all the same, so the message is not being split up....but each one with different identifiers showing the route it needs to take/the next hub it needs to hit...., what would be the best way to garantee or make the odds high that that one message will be received at the destination...given that each hub works only fifty percent of the time. So they can either forward the data packet or not. Essentially, the hubs probably of sending are fifty percent sending and fifty percent not sending...

This is probably pretty confusing, but if anyone can help me out or point in the right direction, that would be so much appreciated.

Thanks!

Why is a "packet switching network" not a frame switching network? I recently started restudying networking and htis is bugging me.