[Cerowrt-devel] Google working on experimental 3.8 Linux kernel for Android

dpreed at reed.com dpreed at reed.com
Sat Mar 2 09:20:28 EST 2013

Hi Ketan -
It is possible for good architects to simplify rather than to ramify.  It takes clear understanding of the system as a whole, a unifying perspective, and a goal to make the system work extremely well and simply.
One of the key insights into how to do this was the choice of features included in the IP "layer" of the Internet stack.  That is - almost none.  And if you read history, the IP layer got simpler as features (like TOS) that had no sensible definition were de facto deprecated, by failure to be utilized for any useful purpose.
This process works for well-architected abstractions.
And it is why the original Internet team included people with expertise in radio networks, shared-medium LANs, etc., end-to-end cryptographic security and authentication, as well as people who understood the properties of voice conversation codecs, etc.  "Features" for those were not "omitted" - they were, instead, carefully thought through.
Since the goal of IP was to operate a completely technologically heterogeneous and application heterogeneous universal network, one where new technologies could be introduced without change,  those were the most serious issues.
Yet relative simplicity was achieved.
This contrasts with the process of 3GPP and the wireless industry.  Bad designs that specifically focus on one application (voip) without abstraction or generalization, or specifically bind-in properties of one specific "bearer" technology (circuits, scheduling algorithms), fail utterly to even fit new situations that come up as a matter of course.
This is why LTE (just deployed) is on version 12, and IP is on version 4 (where the first 3 never were deployed...), and moving to 6, which is almost exactly the same as 4.
And I claim (it's no longer a radical claim - we see the same successes in other areas of architecture), that the principle of well-thought-through simplicity is the *only* quality architectural approach.
However, there are opposing views.  The latest of these is the absolute obsession with finding "cross layer" optimizations, and making a religion of protocol cross-layer "features".  Which of course makes such architectures "optimal" in a sense, for one instant of time, one particular technology, and one Professor's "career".
However, nothing is more likely than change.  And "cross layer" ideas essentially blow up any potential for change.
So it's easy to predict that LTE is a bloody disaster in the making.  It's a dysfunctional system designed by a committee, driven by egos and equipment vendors' desire to "differentiate" merely to partition the market. (why else do no cellphones that support LTE work on any other operators' LTE network?  That's on purpose - it's a marketing requirement!
So a protocol that started out to use a very nice innovation (OFDM) is now a turkey.
It has a life, but not because of all those "features".  In most likelihood, all those "features" will eventually make it so toxic that it will be replaced quickly (not by another cellular operator-centric protocol but by something quite different that unifies fiber and wireless in "local" deployments, if I were to bet).
-----Original Message-----
From: "Ketan Kulkarni" <ketkulka at gmail.com>
Sent: Friday, March 1, 2013 10:39pm
To: dpreed at reed.com
Cc: "Jim Gettys" <jg at freedesktop.org>, "cerowrt-devel at lists.bufferbloat.net" <cerowrt-devel at lists.bufferbloat.net>, "bloat" <bloat at lists.bufferbloat.net>
Subject: Re: [Cerowrt-devel] Google working on experimental 3.8 Linux kernel for Android

Hi David,

While I tend to agree with most of the stuff, however the complexity and too many knobs in mobile networks do come with the added technology.

Consider from end-user perspective, getting a voice call while surfing/downloading on 2G/3G interrupts all the download and it is annoying. 
 So when LTE provides a spec to handle voip + internet both simultaneously, its great benefit to end user.
While roaming around in LTE and moving to 2G/3G network or vice-versa, the handoff occurs seamlessly, the internet traffic is not interrupted. This was not the case in previous mobile generations. End-user is more satisfied as it relates to the daily usage of mobile phones.
 Similarly going ahead we might very well have handoff from wifi to LTE - why not? 

Now for (non-technical) mobile users, these are good and simple to have features. But from networks perspective, where and how will this complexity be handled? definitely some nodes in the network will have to worry about LTE, UMTS, CDMA, eHRPD and what not.
 This gives some idea of how really complex the network looks like -
[http://www.trilliumposter.com/lte.php] http://www.trilliumposter.com/lte.php

From mobile ISP perspective, they invest heavy amount in getting channel license from governments. It takes years to cover up the amount invested. Moreover, consider e.g. facebook embedding ads as per user's interest areas. The ads revenue benefit facebook. However, the mobile ISPs who carry the same data and has potentially (more relevant) information of the subscriber cannot gain from this. 
 This is a real problem to solve for ISPs. Monetization is becoming more relevant to ISPs which again will definitely lead to more complexity in network.

From field engineers, I think many of them are bound by what they are "asked" to do. They have certain task to complete, with exactly what to test, and in what time. Not all field engineers will think beyond that. If their senior technical leads ask them to test packet drops they will test drops. Going ahead if they are asked to test latency they will do so with whatever available resources and knowledge they have. We can not expect an average engineer to think beyond a certain level and do what is not expected out of them from their seniors. Not all will think of the internet finally.

Even the network vendors like Cisco, ALU or Ericsson will test the latency and all the other stuff when there is real push from customers like VzW and ATT. Definitely the vendors can very well invest, research and come up with the latest methods and techniques of measurement and they are doing so to some extent.  Still there is some time to come to a really good picture here.

I think the tests like RRUL is definitely a good start and is more relevant to the ISPs VzW and ATT because its finally they own the network. If these companies are convinced, it will become little easy to push the vendors to do the right stuff. It then puts the correct target for field engineers to accomplish. 
 We have to accept the fact that no standard till now specifies what and how to test latencies. This is one of many of the reasons beating bloat has been a daunting task in the complex world.

FWIK, every ISP has a latency and jitter budget already with every node in their network and these are well communicated to vendors. Now how do one measure, under what scenarios is still not very clearly defined nor understood. So every one interpret these to their knowledge and understanding and a common consensus internally is achieved.
 Who looks at e2e latency? Who looks at the complex interactions of these nodes and their effect? - I really dont know.
Add to that the dynamic behavior of cell and radio access. All this really complicates the stuff at network.

While we have the right step with RRUL, but its still a long way to go for a better picture on mobile networks.

Sometimes "LTE = Long Term Employment" is an apt description.


On Fri, Mar 1, 2013 at 9:57 PM,  <[mailto:dpreed at reed.com] dpreed at reed.com> wrote:

I don't doubt that they test.  My point was different - there are too many knobs and too big a parameter space to test efectively.  And that's the point.
I realize that it's extremely fun to invent parameters in "standards organizations" like 3GPP.  Everybody has their own favorite knob, and a great rationale for some unusual, but critically "important" customer requirement that might come up some day.  Hell, Linux has a gazillion (yes, that's a technical term in mathematics!) parameters, almost none of which are touched.  This reflects the fact that nothing ever gets removed once added.  LTE is now going into release 12, and it's completely ramified into "solutions" to problems that will never be fixed in the field with those solutions.  It's great for European Publically Funded Academic-Industry research - lots for those "Professors" to claim they invented.
I've worked with telco contractors in the field.   They don't read manuals, and they don't read specs.  They have a job to do, and so much money to spend, and time's a wasting.  They don't even work for Verizon or ATT.  They follow "specs" handed down, and charge more if you tell them that the specs have changed.
This is not how brand-new systems get tuned.
It's a Clown Circus out there, and more parameters don't help.
This is why "more buffering is better" continues to be the law of the land - the spec is defined to be "no lost packets under load".   I'm sure that the primary measure under load for RRUL will be "no lost packets" by the time it gets to field engineers in the form of "specs" - because that's what they've *always* been told, and they will disregard any changes as "typos".
A system with more than two control parameters that interact in complex ways is ungovernable - and no control parameters in LTE are "orthogonal", much less "linear" in their interaction.
-----Original Message-----
From: "Jim Gettys" <[mailto:jg at freedesktop.org] jg at freedesktop.org>
Sent: Friday, March 1, 2013 11:09am
 To: "David P Reed" <[mailto:dpreed at reed.com] dpreed at reed.com>

Cc: "Ketan Kulkarni" <[mailto:ketkulka at gmail.com] ketkulka at gmail.com>, "[mailto:cerowrt-devel at lists.bufferbloat.net] cerowrt-devel at lists.bufferbloat.net" <[mailto:cerowrt-devel at lists.bufferbloat.net] cerowrt-devel at lists.bufferbloat.net>
 Subject: Re: [Cerowrt-devel] Google working on experimental 3.8 Linux kernel for Android

On Fri, Mar 1, 2013 at 10:40 AM,  <[mailto:dpreed at reed.com] dpreed at reed.com> wrote:

One wonders why all this complexity is necessary, and how likely it is to be "well tuned" by operators and their contract installers.
I'm willing to bet $1000 that all the testing that is done is "Can you hear me now" and a "speed test".  Not even something as simple and effective as RRUL.
Actually, at least some the the carriers do much more extensive testing; but not with the test tools we would like to see used (yet).
An example is AT&T, where in research, KK Ramakrishnan has a van with 20 or so laptops so he can go driving around and load up a cell in the middle of the night and get data.   And he's research; the operations guys do lots of testing I gather, but more at the radio level.
Next up, is to educate KK to run RRUL.
And in my own company, I've seen data, but it is too high level: e.g. performance of "web" video: e.g. siverlight, flash, youtube, etc.
A common disease that has complicated all this is the propensity for companies to use Windows XP internally for everything: since window scaling is turned off, you can't saturate a LTE link the way you might like to do with a single TCP connection.
- Jim

-----Original Message-----
From: "Ketan Kulkarni" <[mailto:ketkulka at gmail.com] ketkulka at gmail.com>
Sent: Friday, March 1, 2013 3:00am
 To: "Jim Gettys" <[mailto:jg at freedesktop.org] jg at freedesktop.org>
 Cc: "[mailto:cerowrt-devel at lists.bufferbloat.net] cerowrt-devel at lists.bufferbloat.net" <[mailto:cerowrt-devel at lists.bufferbloat.net] cerowrt-devel at lists.bufferbloat.net>
 Subject: Re: [Cerowrt-devel] Google working on experimental 3.8 Linux kernel for Android

On Fri, Mar 1, 2013 at 1:33 AM, Jim Gettys <[mailto:jg at freedesktop.org] jg at freedesktop.org> wrote:

I've got a bit more insight into LTE than I did in the past, courtesy of the last couple days.
To begin with, LTE runs with several classes of service (the call them bearers).  Your VOIP traffic goes into one of them.
And I think there is another as well that is for guaranteed bit rate traffic.  One transmit opportunity may have a bunch of chunks of data, and that data may be destined for more than one device (IIRC).  It's substantially different than WiFi.
Just thought to put more light on bearer stuff:

There are two ways bearers are setup: 
1. UE initiated - where User Equipment sets-up the "parameters" for bearer 
 2. Network initiated - where node like PCRF and PGW sets-up the "parameters". 
 Parameters include the Guaranteed bit-rates, maximum bit-rates. Something called QCI is associated with bearers. The QCI parameters are authorized at PCRF (policy control rule function) and there is certain mapping maintained at either PCRF or PGW between QCI values and DSCP and MBRs.
 These parameters enforcing is done at PGW (in such case it is termed as PCEF - policy and rule enforcement function). So PGWs depending on bearers can certainly modify dscp bits. Though these can be modified by other nodes in the network. 

There are two types of bearers: 1. Dedicated bearers - to carry traffic which need "special" treatment 2. Default or general pupose bearers - to carry all general purpose data.
So generally the voip, streaming videos are passed over dedicated bearers and apply (generally) higher GBRs, MBRs and correct dscp markings.
 And other non-latency sensitive traffic generally follows the default bearer.

Theoretical limit on maximum bearers is 11 though practically most of the deployments use upto 3 bearers max.

Note that these parameters may very well very based on the subscriber profiles. Premium/Corporate subscribers can well have more GBRs and MBRs.
 ISPs are generally very much sensitive to the correct markings at gateways for obvious reasons.

But most of what we think of as Internet stuff (web surfing, dns, etc) all gets dumped into a single best effort ("BE"), class.
The BE class is definitely badly bloated; I can't say how much because I don't really know yet; the test my colleague ran wasn't run long enough to be confident it filled the buffers).  But I will say worse than most cable modems I've seen.  I expect this will be true to different degrees on different hardware.  The other traffic classes haven't been tested yet for bufferbloat, though I suspect they will have it too.  I was told that those classes have much shorter queues, and when the grow, they dump the whole queues (because delivering late real time traffic is useless).  But trust *and* verify....  Verification hasn't been done for anything but BE traffic, and that hasn't been quantified.
But each device gets a "fair" shot at bandwidth in the cell (or sector of a cell; they run 3 radios in each cell), where fair is basically time based; if you are at the edge of a cell, you'll get a lot less bandwidth than someone near a tower; and this fairness is guaranteed by a scheduler than runs in the base station (called a b-nodeb, IIIRC).  So the base station guarantees some sort of "fairness" between devices (a place where Linux's wifi stack today fails utterly, since there is a single queue per device, rather than one per station).
Whether there are bloat problems at the link level in LTE due to error correction I don't know yet; but it wouldn't surprise me; I know there was in 3g.  The people I talked to this morning aren't familiar with the HARQ layer in the system.
The base stations are complicated beasts; they have both a linux system in them as well as a real time operating system based device inside  We don't know where the bottle neck(s) are yet.  I spent lunch upping their paranoia and getting them through some conceptual hurdles (e.g. multiple bottlenecks that may move, and the like).  They will try to get me some of the data so I can help them figure it out.  I don't know if the data flow goes through the linux system in the bnodeb or not, for example.
Most carriers are now trying to ensure that their backhauls from the base station are never congested, though that is another known source of problems.  And then there is the lack of AQM at peering point routers....  You'd think they might run WRED there, but many/most do not.
- Jim

On Thu, Feb 28, 2013 at 2:08 PM, Dave Taht <[mailto:dave.taht at gmail.com] dave.taht at gmail.com> wrote:

On Thu, Feb 28, 2013 at 1:57 PM,  <[mailto:dpreed at reed.com] dpreed at reed.com> wrote:

Doesn't fq_codel need an estimate of link capacity?
No, it just measures delay. Since so far as I know the outgoing portion of LTE is not soft-rate limited, but sensitive to the actual available link bandwidth, fq_codel should work pretty good (if the underlying interfaces weren't horribly overbuffired) in that direction.
I'm looking forward to some measurements of actual buffering at the device driver/device levels.
I don't know how inbound to the handset is managed via LTE.
Still quite a few assumptions left to smash in the above.
in the home router case....
When there are artificial rate limits in play (in, for example, a cable modem/CMTS, hooked up via gigE yet rate limiting to 24up/4mbit down), then a rate limiter (tbf,htb,hfsc) needs to be applied locally to move that rate limiter/queue management into the local device, se we can manage it better.
I'd like to be rid of the need to use htb and come up with a rate limiter that could be adjusted dynamically from a daemon in userspace, probing for short all bandwidth fluctuations while monitoring the load. It needent send that much data very often, to come up with a stable result....
You've described one soft-rate sensing scheme (piggybacking on TCP), and I've thought up a few others, that could feed back from a daemon some samples into a a soft(er) rate limiter that would keep control of the queues in the home router. I am thinking it's going to take way too long to fix the CPE and far easier to fix the home router via this method, and certainly it's too painful and inaccurate to merely measure the bandwidth once, then set a hard rate, when
So far as I know the gargoyle project was experimenting with this approach.
A problem is in places that connect more than one device to the cable modem... then you end up with those needing to communicate their perception of the actual bandwidth beyond the link.

Where will it get that from the 4G or 3G uplink?

-----Original Message-----
From: "Maciej Soltysiak" <[mailto:maciej at soltysiak.com] maciej at soltysiak.com>
Sent: Thursday, February 28, 2013 1:03pm
 To: [mailto:cerowrt-devel at lists.bufferbloat.net] cerowrt-devel at lists.bufferbloat.net
Subject: [Cerowrt-devel] Google working on experimental 3.8 Linux kernel for Android

Looks like Google's experimenting with 3.8 for Android: [https://android.googlesource.com/kernel/common/+/experimental/android-3.8] https://android.googlesource.com/kernel/common/+/experimental/android-3.8
Sounds great if this means they will utilize fq_codel, TFO, BQL, etc.
Anyway my nexus 7 says it has 3.1.10 and this 3.8 will probably go to Android 5.0 so I hope Nexus 7 will get it too some day or at least 3.3+
Phoronix coverage: [http://www.phoronix.com/scan.php?page=news_item&px=MTMxMzc] http://www.phoronix.com/scan.php?page=news_item&px=MTMxMzc
Their 3.8 changelog: [https://android.googlesource.com/kernel/common/+log/experimental/android-3.8] https://android.googlesource.com/kernel/common/+log/experimental/android-3.8
 Cerowrt-devel mailing list
[mailto:Cerowrt-devel at lists.bufferbloat.net] Cerowrt-devel at lists.bufferbloat.net
[https://lists.bufferbloat.net/listinfo/cerowrt-devel] https://lists.bufferbloat.net/listinfo/cerowrt-devel

Dave Täht

Fixing bufferbloat with cerowrt: [http://www.teklibre.com/cerowrt/subscribe.html] http://www.teklibre.com/cerowrt/subscribe.html 
 Cerowrt-devel mailing list
[mailto:Cerowrt-devel at lists.bufferbloat.net] Cerowrt-devel at lists.bufferbloat.net
[https://lists.bufferbloat.net/listinfo/cerowrt-devel] https://lists.bufferbloat.net/listinfo/cerowrt-devel

 Cerowrt-devel mailing list
[mailto:Cerowrt-devel at lists.bufferbloat.net] Cerowrt-devel at lists.bufferbloat.net
[https://lists.bufferbloat.net/listinfo/cerowrt-devel] https://lists.bufferbloat.net/listinfo/cerowrt-devel

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <https://lists.bufferbloat.net/pipermail/cerowrt-devel/attachments/20130302/b1c64e1a/attachment-0002.html>

More information about the Cerowrt-devel mailing list