Re: [Bloat] Terminology for Laypeople

Re: [Bloat] Terminology for Laypeople

[Ingemar Johansson S]

[-- Attachment #1: Type: text/plain, Size: 1347 bytes --]

Hi

Yes. "Idle latency"  and "Working latency" make sense.

Note however that if you think of idle latency as sparse ping, then these sparse ping can give unreasonably high values over cellular access (4G/5G). The reason is here mainly DRX which is a battery saving function in mobile devices. More frequent pings like every 20ms over the course of 100ms or so can give more correct values.

/Ingemar


> Message: 1
> Date: Tue, 11 May 2021 21:26:21 +0000
> From: Greg White <g.white@CableLabs.com>
> To: Jonathan Foulkes <jf@jonathanfoulkes.com>, "Livingood, Jason"
> 	<Jason_Livingood@comcast.com>
> Cc: bloat <bloat@lists.bufferbloat.net>
> Subject: Re: [Bloat] Terminology for Laypeople
> Message-ID: <0A5DF790-7A71-4B84-A20B-559A5E0CE65F@cablelabs.com>
> Content-Type: text/plain; charset="utf-8"
> 
> I recently heard Stuart Cheshire (sort of tongue-in-cheek) refer to “idle
> latency” as “the latency that users experience when they are not using their
> internet connection” (or something along those lines).
> 
> I think terminology that reinforces that the baseline (unloaded) latency is not
> always what users experience, and that latency under load is not referring to
> some unusual corner-case situation, is good.  So, I like “idle latency” and
> “working latency”.
> 
> -Greg
> 
> 


[-- Attachment #2: smime.p7s --]
[-- Type: application/pkcs7-signature, Size: 5437 bytes --]

Re: [Bloat] Terminology for Laypeople

[Dave Collier-Brown]

[-- Attachment #1: Type: text/plain, Size: 2002 bytes --]

"lag" is often understood by non-technical folks, as in "the lag between 
the time you step on the gas and the time the car actually speeds up".

Some folks who've been exposed to video enough will know about "lag and 
jitter" (;-))

--dave

On 2021-05-12 11:50 a.m., Ingemar Johansson S via Bloat wrote:
> Hi
>
> Yes. "Idle latency"  and "Working latency" make sense.
>
> Note however that if you think of idle latency as sparse ping, then these sparse ping can give unreasonably high values over cellular access (4G/5G). The reason is here mainly DRX which is a battery saving function in mobile devices. More frequent pings like every 20ms over the course of 100ms or so can give more correct values.
>
> /Ingemar
>
>
>> Message: 1
>> Date: Tue, 11 May 2021 21:26:21 +0000
>> From: Greg White <g.white@CableLabs.com>
>> To: Jonathan Foulkes <jf@jonathanfoulkes.com>, "Livingood, Jason"
>> 	<Jason_Livingood@comcast.com>
>> Cc: bloat <bloat@lists.bufferbloat.net>
>> Subject: Re: [Bloat] Terminology for Laypeople
>> Message-ID: <0A5DF790-7A71-4B84-A20B-559A5E0CE65F@cablelabs.com>
>> Content-Type: text/plain; charset="utf-8"
>>
>> I recently heard Stuart Cheshire (sort of tongue-in-cheek) refer to “idle
>> latency” as “the latency that users experience when they are not using their
>> internet connection” (or something along those lines).
>>
>> I think terminology that reinforces that the baseline (unloaded) latency is not
>> always what users experience, and that latency under load is not referring to
>> some unusual corner-case situation, is good.  So, I like “idle latency” and
>> “working latency”.
>>
>> -Greg
>>
>>
>
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

-- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
dave.collier-brown@indexexchange.com |              -- Mark Twain


[-- Attachment #2: Type: text/html, Size: 3363 bytes --]

Re: [Bloat] Terminology for Laypeople

[john]

Hello,
I have been on this mailing list since March this year. I am one of the 
laypeople who is very interested in bufferbloat and have been trying to 
understand what the bufferbloat actually is by reading your email 
exchanges.

I was very happy when the subject: [Bloat] Questions for Bufferbloat 
Wikipedia article, came up. I thought someone could finally explain it 
with very easy word which I could understand. But the more discussion 
goes on the more I got confused and I thought this mailing list is for 
aliens exchanging idea not for me human.
:-)

I also watched Dave's
Making Wifi Fast + Slides - BattleMeshV8 - YouTube
https://www.youtube.com/watch?v=Rb-UnHDw02o&t=1657s
But I still could not really understand exactly what Dave was trying to 
explain in it.

Then, another subject: [Bloat] Terminology for Laypeople, came up. I 
thought I could finally understand what the bufferbloat is. But so far I 
am not even close to the understanding. Now it seems to me that this 
mailing list is only for Albert Einstein class scientists to discuss 
something super difficult to understand. This is because no one seems to 
be able to explain in a manner laypeople can understand. Or probably my 
effort to understand it is not really enough yet. But I think I have 
been really doing my best trying to understand what all of you are 
trying to explain about.

I thought I could not be qualified to even drop a line to this mailing 
list because I can not write nor explain in the manner others on this 
mailinglist do. I have been hesitating to ask a question. But please 
allow me to share with you what image came up in my mind after reading 
all the Emails exchanged here since March.

As of now, I am wondering if the bufferbloat could be explained with 
something like a sushi belt-conveyor, the one you may find at the sushi 
restaurant equipped with automated belt-conveyor to bring plates with 
sushi to customer's seat.

In my mind, I am seeing the plate as a packet and the conveyor as 
network. For Europeans and Americans, it may be easier to picture the 
KrispyKreme's donuts belt conveyor in mind instead of sushi one.  Donuts 
are the packets and they go around the network on conveyor in my mind.

After watching Dave's YouTube video, it seems to me, the congestion of 
the packets which Dave was explaining is equivalent to sushi plats on 
conveyor stuck on the route and colliding each other on the conveyor 
since the conveyor keeps bringing more packets one after another to the 
collision point, then plats overflow from the conveyor and dropped on 
the floor.

So now, my question is the picture I described above is close to what 
bufferbloat is? Or I am still very far from understanding? If I am still 
far from understanding, will you be able to explain it to me, the 
laypeople, using the sushi or donuts conveyor? Is the problem the speed 
adjustment of the conveyor? Or too many plates or donuts are placed on 
the conveyor? If so, why the rate or speed of each factors can not be 
adjusted? I even wonder if you could explain it using the door to door 
package delivery service since you are talking about delivering packets.

By the way, thank you very much for your effort to make the explanation 
of bufferbloat to laypeople!

john




--------- Original Message ----------
Subject:  Re: [Bloat] Terminology for Laypeople
From:     Dave Collier-Brown <davecb.42@gmail.com>
To:       bloat@lists.bufferbloat.net
Cc:       
Date:     Wed, 12 May 2021 17:51:16 -0400

"lag" is often understood by non-technical folks, as in "the lag between 
the time you step on the gas and the time the car actually speeds up".
Some folks who've been exposed to video enough will know about "lag and 
jitter" (;-))

--dave 
On 2021-05-12 11:50 a.m., Ingemar Johansson S via Bloat wrote:
>  Hi

Yes. "Idle latency"  and "Working latency" make sense.

Note however that if you think of idle latency as sparse ping, then 
these sparse ping can give unreasonably high values over cellular access 
(4G/5G). The reason is here mainly DRX which is a battery saving 
function in mobile devices. More frequent pings like every 20ms over the 
course of 100ms or so can give more correct values.

/Ingemar


 Message: 1
Date: Tue, 11 May 2021 21:26:21 +0000
From: Greg White <g.white@CableLabs.com>
To: Jonathan Foulkes <jf@jonathanfoulkes.com> , "Livingood, Jason"
	<Jason_Livingood@comcast.com>
Cc: bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] Terminology for Laypeople
Message-ID: <0A5DF790-7A71-4B84-A20B-559A5E0CE65F@cablelabs.com>
Content-Type: text/plain; charset="utf-8"

I recently heard Stuart Cheshire (sort of tongue-in-cheek) refer to "
idle
latency" as "the latency that users experience when they are not using 
their
internet connection" (or something along those lines).

I think terminology that reinforces that the baseline (unloaded) latency 
is not
always what users experience, and that latency under load is not 
referring to
some unusual corner-case situation, is good.  So, I like "idle latency" 
and
"working latency".

-Greg


 

>  _______________________________________________
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat
 -- 
David Collier-Brown,         | Always do right. This will gratify
System Programmer and Author | some people and astonish the rest
dave.collier-brown@indexexchange.com |              -- Mark Twain 
-------------------------------text/plain-------------------------------
 
_______________________________________________
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat
 

Re: [Bloat] Terminology for Laypeople

[Jonathan Morton]

> On 16 May, 2021, at 9:48 pm, john <email@matrix8.org> wrote:
> 
> After watching Dave's YouTube video, it seems to me, the congestion of 
> the packets which Dave was explaining is equivalent to sushi plats on 
> conveyor stuck on the route and colliding each other on the conveyor 
> since the conveyor keeps bringing more packets one after another to the 
> collision point, then plats overflow from the conveyor and dropped on 
> the floor.
> 
> So now, my question is the picture I described above is close to what 
> bufferbloat is? Or I am still very far from understanding? If I am still 
> far from understanding, will you be able to explain it to me, the 
> laypeople, using the sushi or donuts conveyor? Is the problem the speed 
> adjustment of the conveyor? Or too many plates or donuts are placed on 
> the conveyor? If so, why the rate or speed of each factors can not be 
> adjusted? I even wonder if you could explain it using the door to door 
> package delivery service since you are talking about delivering packets.

Here's an analogy for you:

Today there is a logging operation going on just up the road - not unusual in my part of the world.  They have a team felling trees, another team trimming off the branches, and the trunks are then stacked for later delivery to the sawmill (*much* later - they have to season first).  The branches are fed into a chopping machine which produces firewood and mulch, which is then weighed and bagged for immediate sale.

I need firewood for my sauna stove.  I know that if I load my trailer full of firewood, it'll last me about a year.  I figure I'll pay these guys a visit, and it shouldn't take more than half an hour of my time to get what I need.

Under normal circumstances, that would be true.  However, the hardware store in the town an hour away has also chosen today to replenish its stock of firewood, and they have a representative on site who's basically buying the branches from every tree as it comes down; every so often a big van turns up and collects the product.  He graciously lets me step in and buy the branches off one tree for my own use, and they're tagged as such by the loggers.

So instead of just loading ready-made bags of firewood into my trailer, I have to wait for the trimming team to get around to taking the branches off "my" tree which is waiting behind a dozen others.  The branches then go into a big stack of branches waiting for the chopping machine.  When they eventually get around to chopping those, the firewood is carefully put in a separate pile, waiting for the weighing and bagging.

It takes a full hour before I have the branches from "my" tree in a useful format for firing a sauna stove and in my trailer.  Which is now only half full.  To fill it completely, I have to go through the entire process again from the beginning - only the felling team has been going gangbusters and there are now *twenty* trees waiting for trimming.

I planned for half an hour.  It actually took me three hours to get my firewood.  Not for lack of throughput - that was one pretty effective logging operation - but because of the *queues*.

 - Jonathan Morton

Re: [Bloat] Terminology for Laypeople

[Michael Richardson]

[-- Attachment #1: Type: text/plain, Size: 1178 bytes --]


Jonathan Morton <chromatix99@gmail.com> wrote:
    > So instead of just loading ready-made bags of firewood into my trailer,
    > I have to wait for the trimming team to get around to taking the
    > branches off "my" tree which is waiting behind a dozen others.  The
    > branches then go into a big stack of branches waiting for the chopping
    > machine.  When they eventually get around to chopping those, the
    > firewood is carefully put in a separate pile, waiting for the weighing
    > and bagging.

Your analogy is definitely the result of optimizing for batches rather than latency.
(JIT manufacturing in general and much of _The Goal_ talks about the business side of
this, btw)

But, I don't think that it's a great explanation for grandma.
The fetching milk analogy is a bit better, but still not great.

John@matrix8, how did it work for you?

Explaining this is pretty important.

(Thanks for the slide Jonathan)

--
]               Never tell me the odds!                 | ipv6 mesh networks [
]   Michael Richardson, Sandelman Software Works        |    IoT architect   [
]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on rails    [


[-- Attachment #2: signature.asc --]
[-- Type: application/pgp-signature, Size: 487 bytes --]

Re: [Bloat] Terminology for Laypeople

[Aaron Wood]

[-- Attachment #1: Type: text/plain, Size: 3181 bytes --]

I think the "I Love Lucy" chocolate factory scene is perhaps a good analogy:

https://www.youtube.com/watch?v=WmAwcMNxGqM

The chocolates start to come in too fast, and they can't keep up, but
because they aren't telling the kitchen to slow down, they keep piling up
until it collapses into a mess.

Except with networks, many of the senders keep sending packets until the
receiver says that they've missed one (or three, or whatever it is), and
then the sender slows down again.  But if you're hoarding packets, that
signal to slow down is delayed.  And then that creates bufferbloat.

I also like to think of buffers as time.  The buffer in front of a link is
basically a bucket of time the size of the buffer divided by the speed of
the link.  1MB of buffer, in front of a 10Mbps link, is 800ms: (1,000,000
MB) * (8 bits/byte) / 10,000,000 bits /sec => 0.8 seconds.

And so the sender is going to keep sending faster and faster until they go
over 10Mbps, and start to fill that buffer, and then when they do fill it,
they have to resend the missing packets, AND cut their sending rate.

If the buffer is large enough (and therefore the delay long enough), the
sender "overshoots" by so far that they have to just sit and deal with all
the "hey, I missed packets after X" messages from the receiver, until
everything's caught up, and they they can start going faster again (we call
this congestion collapse, because the sender can't send anything new at
all, and once they've sorted out the state of things with the receiver,
they can start again (slowly)).

Congestion collapse is the candy factory from the above clip:  That mess
that needs to be cleaned up before things can start over again (slowly).




On Sun, May 16, 2021 at 1:44 PM Michael Richardson <mcr@sandelman.ca> wrote:

>
> Jonathan Morton <chromatix99@gmail.com> wrote:
>     > So instead of just loading ready-made bags of firewood into my
> trailer,
>     > I have to wait for the trimming team to get around to taking the
>     > branches off "my" tree which is waiting behind a dozen others.  The
>     > branches then go into a big stack of branches waiting for the
> chopping
>     > machine.  When they eventually get around to chopping those, the
>     > firewood is carefully put in a separate pile, waiting for the
> weighing
>     > and bagging.
>
> Your analogy is definitely the result of optimizing for batches rather
> than latency.
> (JIT manufacturing in general and much of _The Goal_ talks about the
> business side of
> this, btw)
>
> But, I don't think that it's a great explanation for grandma.
> The fetching milk analogy is a bit better, but still not great.
>
> John@matrix8, how did it work for you?
>
> Explaining this is pretty important.
>
> (Thanks for the slide Jonathan)
>
> --
> ]               Never tell me the odds!                 | ipv6 mesh
> networks [
> ]   Michael Richardson, Sandelman Software Works        |    IoT
> architect   [
> ]     mcr@sandelman.ca  http://www.sandelman.ca/        |   ruby on
> rails    [
>
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>

[-- Attachment #2: Type: text/html, Size: 4386 bytes --]

Re: [Bloat] Terminology for Laypeople

[Jonathan Morton]

> On 16 May, 2021, at 11:44 pm, Michael Richardson <mcr@sandelman.ca> wrote:
> 
> Your analogy is definitely the result of optimizing for batches rather than latency.

I really don't know how you got there from here.  What I described is basically a pipeline process, not batch processing.  The delay is caused by the fact that the product already in the pipeline has already been bought by the hardware store, and thus contractually the loggers can't divert it to an individual customer like me.

You can think of one bag of firewood as representing a packet of data.  I've requested a particular number of such bags to fill my trailer.  Until my trailer is full, my request is not satisfied.  The hardware store is just taking whatever manufacturing capacity is available; their warehouse is *huge*.

We can explore the analogy further by changing some of the conditions:

1: If the felling of trees was the bottleneck of the operation, such that the trimming, chopping and bagging could all keep up with it, then the delay to me would be much less because I wouldn't have to wait for various backlogs (of complete trees, branches, and piles of firewood) belonging to the hardware store to be dealt with first.  Processing each tree doesn't take very long, there's just an awful lot of them in this patch of forest.

1a: If the foreman told the felling team to take a tea break when a backlog built up, that would have nearly the same effect.  That's what an AQM does.

2: If the hardware store wasn't involved at all, the bags of firewood would be waiting, ready to be sold.  I'd be done in the time it took to load the bags into my trailer.

3: If the loggers sold the *output* of the process to the hardware store, rather than having them reserve it at the head of the pipeline, then I might only have to wait for the throughput of of the operation to produce what I needed, and load it directly into my trailer.  *That* would be just-in-time manufacturing.

 - Jonathan Morton

Re: [Bloat] Terminology for Laypeople

[Jonathan Morton]

> On 17 May, 2021, at 12:33 am, Jonathan Morton <chromatix99@gmail.com> wrote:
> 
> The delay is caused by the fact that the product already in the pipeline has already been bought by the hardware store, and thus contractually the loggers can't divert it to an individual customer like me.

The reason this part of the analogy is relevant (and why I set up the hardware store's representative buying the branches at the felling stage) is because in internet traffic I don't want just any old data packets, I need the ones that specifically relate to the connection I opened.

We could say for the sake of the analogy that the hardware store is buying all the pine and spruce, and the felling team is thus working only on those trees, but I want a birch tree to fuel my sauna (since it's in less demand, the price is lower).  That also makes it easier to identify my branches as they go through the pipeline.

 - Jonathan Morton

Re: [Bloat] Terminology for Laypeople

[Simon Barber]

Bufferbloat -

At every router there are buffers that hold packets about to go out on a link, incase the packets coming into the router arrive faster than they can go out. If the packets keep arriving too fast the buffers fill up and eventually arriving packets have nowhere to go and are dropped. TCP is a protocol that underlies most data transferred on the internet - every time you load a webpage for example, you are using TCP. The way TCP works is by gradually sending data faster and faster until a packet is dropped, then slowing down a bit before trying to ramp speed up again, and again, and again for ever. This means that TCP is trying to keep the buffers in the router that feeds the slowest link in your connection full (Ramping up it’s speed until a packet is dropped - i.e. the buffers are full). Unfortunately these days routers have way more buffers than necessary and every data packets that goes into a buffer must wait for all the other packets ahead of it in the buffer to be sent first. This can add seconds of delay to your connection in some cases.

How’s that?

Simon


> On May 16, 2021, at 11:48 AM, john <email@matrix8.org> wrote:
> 
> Hello,
> I have been on this mailing list since March this year. I am one of the 
> laypeople who is very interested in bufferbloat and have been trying to 
> understand what the bufferbloat actually is by reading your email 
> exchanges.
> 
> I was very happy when the subject: [Bloat] Questions for Bufferbloat 
> Wikipedia article, came up. I thought someone could finally explain it 
> with very easy word which I could understand. But the more discussion 
> goes on the more I got confused and I thought this mailing list is for 
> aliens exchanging idea not for me human.
> :-)
> 
> I also watched Dave's
> Making Wifi Fast + Slides - BattleMeshV8 - YouTube
> https://www.youtube.com/watch?v=Rb-UnHDw02o&t=1657s
> But I still could not really understand exactly what Dave was trying to 
> explain in it.
> 
> Then, another subject: [Bloat] Terminology for Laypeople, came up. I 
> thought I could finally understand what the bufferbloat is. But so far I 
> am not even close to the understanding. Now it seems to me that this 
> mailing list is only for Albert Einstein class scientists to discuss 
> something super difficult to understand. This is because no one seems to 
> be able to explain in a manner laypeople can understand. Or probably my 
> effort to understand it is not really enough yet. But I think I have 
> been really doing my best trying to understand what all of you are 
> trying to explain about.
> 
> I thought I could not be qualified to even drop a line to this mailing 
> list because I can not write nor explain in the manner others on this 
> mailinglist do. I have been hesitating to ask a question. But please 
> allow me to share with you what image came up in my mind after reading 
> all the Emails exchanged here since March.
> 
> As of now, I am wondering if the bufferbloat could be explained with 
> something like a sushi belt-conveyor, the one you may find at the sushi 
> restaurant equipped with automated belt-conveyor to bring plates with 
> sushi to customer's seat.
> 
> In my mind, I am seeing the plate as a packet and the conveyor as 
> network. For Europeans and Americans, it may be easier to picture the 
> KrispyKreme's donuts belt conveyor in mind instead of sushi one.  Donuts 
> are the packets and they go around the network on conveyor in my mind.
> 
> After watching Dave's YouTube video, it seems to me, the congestion of 
> the packets which Dave was explaining is equivalent to sushi plats on 
> conveyor stuck on the route and colliding each other on the conveyor 
> since the conveyor keeps bringing more packets one after another to the 
> collision point, then plats overflow from the conveyor and dropped on 
> the floor.
> 
> So now, my question is the picture I described above is close to what 
> bufferbloat is? Or I am still very far from understanding? If I am still 
> far from understanding, will you be able to explain it to me, the 
> laypeople, using the sushi or donuts conveyor? Is the problem the speed 
> adjustment of the conveyor? Or too many plates or donuts are placed on 
> the conveyor? If so, why the rate or speed of each factors can not be 
> adjusted? I even wonder if you could explain it using the door to door 
> package delivery service since you are talking about delivering packets.
> 
> By the way, thank you very much for your effort to make the explanation 
> of bufferbloat to laypeople!
> 
> john
> 
> 
> 
> 
> --------- Original Message ----------
> Subject:  Re: [Bloat] Terminology for Laypeople
> From:     Dave Collier-Brown <davecb.42@gmail.com>
> To:       bloat@lists.bufferbloat.net
> Cc:       
> Date:     Wed, 12 May 2021 17:51:16 -0400
> 
> "lag" is often understood by non-technical folks, as in "the lag between 
> the time you step on the gas and the time the car actually speeds up".
> Some folks who've been exposed to video enough will know about "lag and 
> jitter" (;-))
> 
> --dave 
> On 2021-05-12 11:50 a.m., Ingemar Johansson S via Bloat wrote:
>> Hi
> 
> Yes. "Idle latency"  and "Working latency" make sense.
> 
> Note however that if you think of idle latency as sparse ping, then 
> these sparse ping can give unreasonably high values over cellular access 
> (4G/5G). The reason is here mainly DRX which is a battery saving 
> function in mobile devices. More frequent pings like every 20ms over the 
> course of 100ms or so can give more correct values.
> 
> /Ingemar
> 
> 
> Message: 1
> Date: Tue, 11 May 2021 21:26:21 +0000
> From: Greg White <g.white@CableLabs.com>
> To: Jonathan Foulkes <jf@jonathanfoulkes.com> , "Livingood, Jason"
> 	<Jason_Livingood@comcast.com>
> Cc: bloat <bloat@lists.bufferbloat.net>
> Subject: Re: [Bloat] Terminology for Laypeople
> Message-ID: <0A5DF790-7A71-4B84-A20B-559A5E0CE65F@cablelabs.com>
> Content-Type: text/plain; charset="utf-8"
> 
> I recently heard Stuart Cheshire (sort of tongue-in-cheek) refer to "
> idle
> latency" as "the latency that users experience when they are not using 
> their
> internet connection" (or something along those lines).
> 
> I think terminology that reinforces that the baseline (unloaded) latency 
> is not
> always what users experience, and that latency under load is not 
> referring to
> some unusual corner-case situation, is good.  So, I like "idle latency" 
> and
> "working latency".
> 
> -Greg
> 
> 
> 
> 
>> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
> -- 
> David Collier-Brown,         | Always do right. This will gratify
> System Programmer and Author | some people and astonish the rest
> dave.collier-brown@indexexchange.com |              -- Mark Twain 
> -------------------------------text/plain-------------------------------
> 
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
> 
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Terminology for Laypeople

[Jonathan Morton]

> On 17 May, 2021, at 8:18 am, Simon Barber <simon@superduper.net> wrote:
> 
> How’s that?

It's a wall of text full of technical jargon.  It seems to be technically correct, but probably not very useful for the intended context.

 - Jonathan Morton

[Bloat] Terminology for Laypeople

[Livingood, Jason]

[-- Attachment #1: Type: text/plain, Size: 975 bytes --]

Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).

As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.

Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?

Thanks for any input,
Jason

[-- Attachment #2: Type: text/html, Size: 2855 bytes --]

Re: [Bloat] Terminology for Laypeople

[James R Cutler]

[-- Attachment #1: Type: text/plain, Size: 1615 bytes --]

Jason,

I find “idle” and “working” to be understandable both individually and as opposites in almost any field or endeavor. These terms also mesh with typical speed test instructions regarding testing under idle conditions without working internet applications.

	James

> On May 4, 2021, at 7:02 PM, Livingood, Jason via Bloat <bloat@lists.bufferbloat.net> wrote:
> 
> Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).
>  
> As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.
>  
> Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?
>  
> Thanks for any input,
> Jason
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net <mailto:Bloat@lists.bufferbloat.net>
> https://lists.bufferbloat.net/listinfo/bloat <https://lists.bufferbloat.net/listinfo/bloat>

[-- Attachment #2: Type: text/html, Size: 6333 bytes --]

Re: [Bloat] Terminology for Laypeople

[Erik Auerswald]

Hi,

I, too, think that "idle" and "working" seem to be
useful terms here.  An alternative to "idle" might
be "minimum," since this implies that the latency
is not always at the minimum.

Thanks,
Erik

On 05.05.21 02:14, James R Cutler wrote:
> Jason,
> 
> I find “idle” and “working” to be understandable both individually and as opposites in almost any field or endeavor. These terms also mesh with typical speed test instructions regarding testing under idle conditions without working internet applications.
> 
> 	James
> 
>> On May 4, 2021, at 7:02 PM, Livingood, Jason via Bloat <bloat@lists.bufferbloat.net> wrote:
>>
>> Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).
>>   
>> As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.
>>   
>> Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?
>>   
>> Thanks for any input,
>> Jason

Re: [Bloat] Terminology for Laypeople

[Dave Taht]

I too like idle and working.

On Wed, May 5, 2021 at 12:42 AM Erik Auerswald
<auerswal@unix-ag.uni-kl.de> wrote:
>
> Hi,
>
> I, too, think that "idle" and "working" seem to be
> useful terms here.  An alternative to "idle" might
> be "minimum," since this implies that the latency
> is not always at the minimum.
>
> Thanks,
> Erik
>
> On 05.05.21 02:14, James R Cutler wrote:
> > Jason,
> >
> > I find “idle” and “working” to be understandable both individually and as opposites in almost any field or endeavor. These terms also mesh with typical speed test instructions regarding testing under idle conditions without working internet applications.
> >
> >       James
> >
> >> On May 4, 2021, at 7:02 PM, Livingood, Jason via Bloat <bloat@lists.bufferbloat.net> wrote:
> >>
> >> Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).
> >>
> >> As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.
> >>
> >> Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?
> >>
> >> Thanks for any input,
> >> Jason
>
>
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 
Latest Podcast:
https://www.linkedin.com/feed/update/urn:li:activity:6791014284936785920/

Dave Täht CTO, TekLibre, LLC

Re: [Bloat] Terminology for Laypeople

[Matt Mathis]

I suggest moving further up the stack - any random Zoomer or gamer
understands that Application Lag is a bad thing.   Furthermore
statements of the form "Lag is most often caused by LUL or even BB"
are likely to be true without caveat.

Inverting the statement "BB causes LAG (or even LUL)" are less
generally true because the vast majority of queues in the internet are
drop tail, protected by adjacent managed queues (e.g. all of the
queues in switch fabrics within a chassi are protected my managed
queues at the input/output cards).    Statements of causality up the
stack are almost always vague and inaccurate or precise and too
complicated.

I agree with your colleague that since people don't understand micro
bursts, they assume links that are underloaded in the average are
unloaded.  (But note that this issue is an artifact of self clocked
protocols, and may change as more paced CC rolls out).

So my vote would be [Working] Application Lag, just to move a little
further up the stack.

Thanks,
--MM--
The best way to predict the future is to create it.  - Alan Kay

We must not tolerate intolerance;
       however our response must be carefully measured:
            too strong would be hypocritical and risks spiraling out of control;
            too weak risks being mistaken for tacit approval.

On Tue, May 4, 2021 at 5:02 PM Livingood, Jason via Bloat
<bloat@lists.bufferbloat.net> wrote:
>
> Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).
>
>
>
> As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.
>
>
>
> Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?
>
>
>
> Thanks for any input,
>
> Jason
>
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Terminology for Laypeople

[Neil Davies]

[-- Attachment #1.1: Type: text/plain, Size: 1523 bytes --]

The Broadband forum in the QED initiative (https://www.broadband-forum.org/download/TR-452.1.pdf)  us “structural” to capture the “impairment” from technological / topology	issues.

Neil

> On 5 May 2021, at 01:02, Livingood, Jason via Bloat <bloat@lists.bufferbloat.net> wrote:
> 
> Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).
> 
> As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.
> 
> Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?
> 
> Thanks for any input,
> Jason
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net <mailto:Bloat@lists.bufferbloat.net>
> https://lists.bufferbloat.net/listinfo/bloat <https://lists.bufferbloat.net/listinfo/bloat>

[-- Attachment #1.2: Type: text/html, Size: 6266 bytes --]

[-- Attachment #2: Message signed with OpenPGP --]
[-- Type: application/pgp-signature, Size: 833 bytes --]

Re: [Bloat] Terminology for Laypeople

[Jason Iannone]

[-- Attachment #1: Type: text/plain, Size: 2141 bytes --]

It's not a short discussion but I start with a comparison of circuit and
packet switching, usually with an accompanying drawing. There's a physicist
joke in here about assuming a frictionless environment but for the intent
of this explanation, a circuit switched path is bufferless because circuit
switched networks are point to point and bits are transmitted at the same
rate that they are received. Packet switching introduces a mechanism for
nodes supporting multiple ingress, single egress transmission. In order to
support transient bursts, network nodes hold onto bits for a time while the
egress interface processes the node's ingress traffic. That hold time
equates to additional latency. Every node in a path may subject a flow's
traffic to buffering, increasing latency in transit based on its individual
load.

Jason

On Tue, May 4, 2021 at 8:02 PM Livingood, Jason via Bloat <
bloat@lists.bufferbloat.net> wrote:

> Like many of you I have been immersed in buffer bloat discussions for many
> years, almost entirely within the technical community. Now that I am
> starting to explain latency & latency under load to internal non-technical
> folks, I have noticed some people don’t really understand “traditional”
> latency vs. latency under load (LUL).
>
>
>
> As a result, I am planning to experiment in some upcoming briefings and
> call traditional latency “idle latency” – a measure of latency conducted on
> an otherwise idle connection. And then try calling LUL either “active
> latency” or perhaps “working latency” (suggested by an external colleague –
> can’t take credit for that one) – to try to communicate it is latency when
> the connection is experiencing normal usage.
>
>
>
> Have any of you here faced similar challenges explaining this to
> non-technical audiences? Have you had any success with alternative terms?
> What do you think of these?
>
>
>
> Thanks for any input,
>
> Jason
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
>

[-- Attachment #2: Type: text/html, Size: 3318 bytes --]

Re: [Bloat] Terminology for Laypeople

[David Lang]

[-- Attachment #1: Type: text/plain, Size: 2463 bytes --]

it's sometimesworth reminding technical folks that if you look at a small enough 
time slice, a network is either 0% or 100% utilized, so if the output is 100% 
utilized the instant a packet arrives, the device ither dropps the data or 
buffers it.

David Lang

  On Thu, 6 May 2021, Jason Iannone wrote:

> It's not a short discussion but I start with a comparison of circuit and
> packet switching, usually with an accompanying drawing. There's a physicist
> joke in here about assuming a frictionless environment but for the intent
> of this explanation, a circuit switched path is bufferless because circuit
> switched networks are point to point and bits are transmitted at the same
> rate that they are received. Packet switching introduces a mechanism for
> nodes supporting multiple ingress, single egress transmission. In order to
> support transient bursts, network nodes hold onto bits for a time while the
> egress interface processes the node's ingress traffic. That hold time
> equates to additional latency. Every node in a path may subject a flow's
> traffic to buffering, increasing latency in transit based on its individual
> load.
>
> Jason
>
> On Tue, May 4, 2021 at 8:02 PM Livingood, Jason via Bloat <
> bloat@lists.bufferbloat.net> wrote:
>
>> Like many of you I have been immersed in buffer bloat discussions for many
>> years, almost entirely within the technical community. Now that I am
>> starting to explain latency & latency under load to internal non-technical
>> folks, I have noticed some people don’t really understand “traditional”
>> latency vs. latency under load (LUL).
>>
>>
>>
>> As a result, I am planning to experiment in some upcoming briefings and
>> call traditional latency “idle latency” – a measure of latency conducted on
>> an otherwise idle connection. And then try calling LUL either “active
>> latency” or perhaps “working latency” (suggested by an external colleague –
>> can’t take credit for that one) – to try to communicate it is latency when
>> the connection is experiencing normal usage.
>>
>>
>>
>> Have any of you here faced similar challenges explaining this to
>> non-technical audiences? Have you had any success with alternative terms?
>> What do you think of these?
>>
>>
>>
>> Thanks for any input,
>>
>> Jason
>> _______________________________________________
>> Bloat mailing list
>> Bloat@lists.bufferbloat.net
>> https://lists.bufferbloat.net/listinfo/bloat
>>
>

[-- Attachment #2: Type: text/plain, Size: 140 bytes --]

_______________________________________________
Bloat mailing list
Bloat@lists.bufferbloat.net
https://lists.bufferbloat.net/listinfo/bloat

Re: [Bloat] Terminology for Laypeople

[Dave Taht]

On Thu, May 6, 2021 at 6:41 AM David Lang <david@lang.hm> wrote:
>
> it's sometimesworth reminding technical folks that if you look at a small enough
> time slice, a network is either 0% or 100% utilized, so if the output is 100%
> utilized the instant a packet arrives, the device ither dropps the data or
> buffers it.

+1. Humans tend to think in terms of Mbit/sec, when a saner interval
to think about is bits/ms
or less. I tend to care about bits/20 ms as being the rightest number
for human perceptible latency.

At a ms level, well, we are so far from that. I'd put over here

http://flent-fremont.bufferbloat.net/~d/broadcom_aug9.pdf

What the "bandwidth" was for a typical web transaction with 50ms
latency nowadays.
It's zero.

The mental image I have of the latest home routers is of one of these:

https://www.telegraph.co.uk/news/picturegalleries/howaboutthat/11164032/Jet-powered-VW-Beetle-that-goes-like-a-rocket.html

impacted into the side of a mountain.

>
> David Lang
>
>   On Thu, 6 May 2021, Jason Iannone wrote:
>
> > It's not a short discussion but I start with a comparison of circuit and
> > packet switching, usually with an accompanying drawing. There's a physicist
> > joke in here about assuming a frictionless environment but for the intent
> > of this explanation, a circuit switched path is bufferless because circuit
> > switched networks are point to point and bits are transmitted at the same
> > rate that they are received. Packet switching introduces a mechanism for
> > nodes supporting multiple ingress, single egress transmission. In order to
> > support transient bursts, network nodes hold onto bits for a time while the
> > egress interface processes the node's ingress traffic. That hold time
> > equates to additional latency. Every node in a path may subject a flow's
> > traffic to buffering, increasing latency in transit based on its individual
> > load.
> >
> > Jason
> >
> > On Tue, May 4, 2021 at 8:02 PM Livingood, Jason via Bloat <
> > bloat@lists.bufferbloat.net> wrote:
> >
> >> Like many of you I have been immersed in buffer bloat discussions for many
> >> years, almost entirely within the technical community. Now that I am
> >> starting to explain latency & latency under load to internal non-technical
> >> folks, I have noticed some people don’t really understand “traditional”
> >> latency vs. latency under load (LUL).
> >>
> >>
> >>
> >> As a result, I am planning to experiment in some upcoming briefings and
> >> call traditional latency “idle latency” – a measure of latency conducted on
> >> an otherwise idle connection. And then try calling LUL either “active
> >> latency” or perhaps “working latency” (suggested by an external colleague –
> >> can’t take credit for that one) – to try to communicate it is latency when
> >> the connection is experiencing normal usage.
> >>
> >>
> >>
> >> Have any of you here faced similar challenges explaining this to
> >> non-technical audiences? Have you had any success with alternative terms?
> >> What do you think of these?
> >>
> >>
> >>
> >> Thanks for any input,
> >>
> >> Jason
> >> _______________________________________________
> >> Bloat mailing list
> >> Bloat@lists.bufferbloat.net
> >> https://lists.bufferbloat.net/listinfo/bloat
> >>
> >_______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net
> https://lists.bufferbloat.net/listinfo/bloat



-- 
Latest Podcast:
https://www.linkedin.com/feed/update/urn:li:activity:6791014284936785920/

Dave Täht CTO, TekLibre, LLC

Re: [Bloat] Terminology for Laypeople

[Jonathan Foulkes]

[-- Attachment #1: Type: text/plain, Size: 1897 bytes --]

Hi Jason,

I’ve found that idle is a good descriptor for unloaded metrics, and for semi-technical audiences ‘working’ is a very good term. But for lay people, the term ‘loaded’ seems to work better, especially since we are talking about a metric that relates to capacity.

e.g.

When my truck is unloaded, my truck stops quickly, but when loaded, it takes longer to stop.

so now:

When my Internet line is unloaded, my latency is low, but when it is highly loaded (iCloud photo sync), the latency is very high.

Cheers,

Jonathan Foulkes


> On May 4, 2021, at 8:02 PM, Livingood, Jason via Bloat <bloat@lists.bufferbloat.net> wrote:
> 
> Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).
>  
> As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.
>  
> Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?
>  
> Thanks for any input,
> Jason
> _______________________________________________
> Bloat mailing list
> Bloat@lists.bufferbloat.net <mailto:Bloat@lists.bufferbloat.net>
> https://lists.bufferbloat.net/listinfo/bloat <https://lists.bufferbloat.net/listinfo/bloat>

[-- Attachment #2: Type: text/html, Size: 6818 bytes --]

Re: [Bloat] Terminology for Laypeople

[Greg White]

[-- Attachment #1: Type: text/plain, Size: 2639 bytes --]

I recently heard Stuart Cheshire (sort of tongue-in-cheek) refer to “idle latency” as “the latency that users experience when they are not using their internet connection” (or something along those lines).

I think terminology that reinforces that the baseline (unloaded) latency is not always what users experience, and that latency under load is not referring to some unusual corner-case situation, is good.  So, I like “idle latency” and “working latency”.

-Greg



From: Bloat <bloat-bounces@lists.bufferbloat.net> on behalf of Jonathan Foulkes <jf@jonathanfoulkes.com>
Date: Monday, May 10, 2021 at 2:10 PM
To: Jason Livingood <Jason_Livingood@comcast.com>
Cc: bloat <bloat@lists.bufferbloat.net>
Subject: Re: [Bloat] Terminology for Laypeople

Hi Jason,

I’ve found that idle is a good descriptor for unloaded metrics, and for semi-technical audiences ‘working’ is a very good term. But for lay people, the term ‘loaded’ seems to work better, especially since we are talking about a metric that relates to capacity.

e.g.

When my truck is unloaded, my truck stops quickly, but when loaded, it takes longer to stop.

so now:

When my Internet line is unloaded, my latency is low, but when it is highly loaded (iCloud photo sync), the latency is very high.

Cheers,

Jonathan Foulkes



On May 4, 2021, at 8:02 PM, Livingood, Jason via Bloat <bloat@lists.bufferbloat.net<mailto:bloat@lists.bufferbloat.net>> wrote:

Like many of you I have been immersed in buffer bloat discussions for many years, almost entirely within the technical community. Now that I am starting to explain latency & latency under load to internal non-technical folks, I have noticed some people don’t really understand “traditional” latency vs. latency under load (LUL).

As a result, I am planning to experiment in some upcoming briefings and call traditional latency “idle latency” – a measure of latency conducted on an otherwise idle connection. And then try calling LUL either “active latency” or perhaps “working latency” (suggested by an external colleague – can’t take credit for that one) – to try to communicate it is latency when the connection is experiencing normal usage.

Have any of you here faced similar challenges explaining this to non-technical audiences? Have you had any success with alternative terms? What do you think of these?

Thanks for any input,
Jason
_______________________________________________
Bloat mailing list
Bloat@lists.bufferbloat.net<mailto:Bloat@lists.bufferbloat.net>
https://lists.bufferbloat.net/listinfo/bloat


[-- Attachment #2: Type: text/html, Size: 7952 bytes --]