Categories
IETF Open Internet Standards Technology

Rough Guide to IETF 92: Internet Scalability & Performance

In this post I’ll shine a light on some of the Internet Engineering Task Force (IETF) and Internet Research Task Force (IRTF) efforts underway to explore and address more sophisticated ways to use available bandwidth, improve Internet performance, and otherwise efficiently get content to where it needs to be. These groups will all be meeting as part of the IETF 92 meeting in Dallas next week.

Following on from the recent IAB workshop on Stack Evolution in a Middlebox Internet (SEMI) that discussed the ossification of the transport layer and how to fix it for emerging applications, the Substrate Protocol for User Datagrams (spud) BoF will discuss requirements and strawman proposals for a protocol to expose selective and minimal metadata to the path. The intent is to restore control to the endpoint with regard to what data is exposed to the path so that middlebox functionality deemed useful can still be supported in the presence of encryption that otherwise renders such functionality impotent. The technical plenary meeting on Monday evening will include a brief readout from the SEMI workshop, and the Transport Area Open Meeting will include a longer presentation of the workshop discussion and outcomes.

Another outcome of the SEMI workshop will take place on Sunday evening in the form of an informal Bar BoF meeting to discuss measurement techniques and data sources that could help make better engineering decisions to work around some of the ossification in the protocol stack. The hope is that techniques similar to ‘happy eyeballs’ for IPv6 can be used to support deployment of new transport features and protocols.

Internet performance is to a large extent governed by the way transport protocols operate, and the tcpm WG will be meeting to discuss proposed new functionality to improve and enhance the working of TCP, the main transport protocol used on the Internet today.

Packet networks give rise to transient congestion by design and several groups are meeting to discuss different aspects of congestion control and avoidance (aqm, iccrg and rmcat). For regulators, being able to monitor the performance of networks, and the extent to which congestion or other factors are impacting consumers’ experience of the network is very important. The lmap working group is meeting in Dallas to advance their important work on standardizing a large-scale broadband performance measurement infrastructure.

Related Working Groups and BoFs at IETF 92

Follow Us

There’s a lot going on next week, and whether you plan to be there or join remotely, there’s much to monitor. To follow along as we dole out this series of Rough Guide to IETF blog posts, follow us on the Internet Technology Matters blogTwitterFacebookGoogle+, via RSS, or see http://dev.internetsociety.org/rough-guide-ietf92.

Categories
Technology

Reducing Internet Latency: The Long-term Challenge of Making the Internet Faster

Editor’s Note: This is a guest blog post by Andreas Petlund from Simula Research and the RITE Project. You can read more about the Internet Society’s work related to Internet Latency at https://dev.internetsociety.org/tags/latency.

The world is waking up to the need for consistent low latency on the Internet. Some people, like Stuart Cheshire of Apple, have tried for decades to make the technical world think about latency when they design systems and standards. Recently, efforts like the Bufferbloat and RITE (reducing Internet transport latency) projects have been working on some of the problems that increase Internet delays. We’ve also seen great initiatives like the Internet Society Workshop on Reducing Internet Latency. Now papers are being written that urge the network community to increase its efforts to realise a near-lightspeed Internet in order to release the potential that such stable low-latency communication will give the apps of the future <http://conferences.sigcomm.org/hotnets/2014/papers/hotnets-XIII-final111.pdf>.

I recently got a question from a journalist when interviewed about our new video explaining sources of Internet latency: when can we have near-zero delay in the Internet? I could glean from his tone that what he hoped for was something like “next year”. I could hear his disappointment over the phone when I answered that we can make good progress within the next years, but that the big changes that would give us the Internet of our dreams could take decades.

So why is the progress towards that goal so slow?

The root lies in the distributed structure of the Internet. If we could deploy a “new” Internet tomorrow, with every component under the same all-powerful control, we would have our low-latency net immediately. The technology is there and has been for a long time.

If we discount this clean-slate utopia however, the road to low-latency happiness has obstacles related to the political, economic and technical domains.

ISPs have businesses to run and customers to keep. They’re very nervous about making changes that may scare their customers away. Since providing network services is a low-margin business, very often, _any_ change proposed will meet strong resistance. Such fear will help maintain a status quo that keeps all the actors at the same level. If we can educate and encourage decision makers within such organisations, the chances of deployment will increase.

For the technical challenges, any solution that will drastically improve the situation has to be widely embraced by the community in order to succeed. Not only that, but it has to support incremental deployment. Some legacy equipment will lurk in the shadows waiting to break your beautifully designed algorithm for low latency communication. So to increase the chance of success, solutions should be standardised and designed with implicit incentives for people to adopt them, even though there may be a phase with sub-maximal benefits due to lack of widespread deployment. In the meantime, there are many ways to make smaller changes that have less impact, but that can still cut some milliseconds from Internet response times.

An important element for increasing the chance of success for Internet latency reduction is to raise public awareness. When the benefits are known, there should be a growing pressure on the influential players to help with the low-latency efforts. My feeling is that we’re about to reach a point where Internet latency is no longer a topic only for small groups of people with a special interest. We’re witnessing a growing interest, much due to the Bufferbloat project’s work. In RITE, we have just released an informational video aiming for raising public awareness about the topic. We’ve included educational material so that it can easily be used in IT 101 courses, allowing a new generation of technicians to be conscious of the latency aspect.

I’m an optimist about this. My hope is that the raised awareness will motivate a collective effort so that we’ll reach agreement on changes that will transform the Internet –without having to wait until 2040.

Categories
IETF Open Internet Standards Technology

Rough Guide to IETF 91: Bandwidth, Scalability, and Internet Performance

In this post I’ll shine a light on some of the Internet Engineering Task Force (IETF) and Internet Research Task Force (IRTF) efforts underway to explore and address more sophisticated ways to use available bandwidth, improve Internet performance, and otherwise efficiently get content to where it needs to be. These groups will all be meeting as part of the IETF 91 meeting in Honolulu next week.

The Deterministic Networking (detnet) BoF will explore the feasibility of extending IEEE Audio Video Bridging to routed networks. The IEEE standard provides time synchronization and precise scheduling for zero congestion loss and finite latency for reserved layer-2 streams. The need to extend these QoS features to routed networks is now emerging for industrial and vehicular applications. The detnet problem statement contains more detail about the motivations for this work.

Internet performance is to a large extent governed by the way transport protocols operate, and the tcpm WG will be meeting to discuss proposed new functionality to improve and enhance the working of TCP, the main transport protocol used on the Internet today.

Packet networks give rise to transient congestion by design and several groups are meeting to discuss different aspects of congestion control and avoidance (aqm, iccrg and rmcat). For regulators, being able to monitor the performance of networks, and the extent to which congestion or other factors are impacting consumers’ experience of the network is very important. The lmap working group is meeting in Honolulu to advance their important work on standardizing a large-scale broadband performance measurement infrastructure.

Reducing Internet latency is important to us at the Internet Society and in the IRTF, the proposed research group on the subject of data centre latency control (dclcrg) will meet again at IETF 91. In recent years a number of techniques have been documented in the research literature on reducing latency for applications running in large data centres and this research group seeks to develop shared problem statements, solutions and other experimental tools.

Related Working Groups and BoFs at IETF 91

tcpm (TCP Maintenance and Minor Extensions) WG
Monday, 10 November 2014, 0900-1130 HST, Hibiscus
Agenda: https://datatracker.ietf.org/meeting/91/agenda/tcpm/
Documents: https://datatracker.ietf.org/wg/tcpm/
Charter: http://datatracker.ietf.org/wg/tcpm/charter/

aqm (Active Queue Management and Packet Scheduling) WG
Monday, 10 November 2014, 1300-1500 HST, Coral 2
Agenda: https://datatracker.ietf.org/meeting/91/agenda/aqm/
Documents: https://datatracker.ietf.org/wg/aqm/
Charter: http://datatracker.ietf.org/wg/aqm/charter/

detnet (Deterministic Networking) BoF
Monday, 10 November 2014, 1520-1720 HST, Coral 1
Agenda: https://datatracker.ietf.org/meeting/91/agenda/detnet/
Draft charter: https://bitbucket.org/pthubert/detnet/src/master/detnet%20charter.txt

dclcrg (Proposed Data Center Latency Control Research Group)
Tuesday, 11 November 2014, 1300-1500 HST, Hibiscus
Agenda: https://datatracker.ietf.org/meeting/91/agenda/dclcrg/

iccrg (Internet Congestion Control Research Group)
Tuesday, 11 November 2014, 1520-1720 HST, Coral 1
Agenda: https://datatracker.ietf.org/meeting/91/agenda/iccrg/
Documents: http://tools.ietf.org/group/irtf/trac/wiki/ICCRG
Charter: https://irtf.org/iccrg

rmcat (RTP Media Congestion Avoidance Techniques) WG
Wednesday, 12 November 2014, 0900-1130 HST, Coral 4
Agenda: https://datatracker.ietf.org/meeting/91/agenda/rmcat/
Documents: https://datatracker.ietf.org/wg/rmcat/
Charter: http://datatracker.ietf.org/wg/rmcat/charter/

lmap (Large-Scale Measurement of Broadband Performance) WG
Thursday, 13 November 2014, 0900-1130 HST, Kahili
Agenda: https://datatracker.ietf.org/meeting/91/agenda/lmap/
Documents: https://datatracker.ietf.org/wg/lmap/
Charter: http://datatracker.ietf.org/wg/lmap/charter/

Follow Us

There’s a lot going on next week, and whether you plan to be there or join remotely, there’s much to follow. To follow along as we dole out this series of Rough Guide to IETF blog posts, follow us on the Internet Technology Matters blog, Twitter, Facebook, Google+, via RSS, or see http://dev.internetsociety.org/rough-guide-ietf91.

 

Categories
Growing the Internet Technology

Why ‘Megafast’ Internet Often Isn’t (Video)

What’s the most important thing determining your satisfaction with your Internet connection? I’m sure a lot of people would say speed, and that’s not surprising as headline bandwidth figures have been the way many commercial Internet service providers have chosen to compete in the marketplace for subscribers for many years. 50Mbps must be better than 10Mbps, right? Superfast sounds really, erm, fast, right?

Well, in many cases, increased bandwidth won’t result in significant improvements to user experience and one of the reasons for that is related to latency, or delay. Measurements from Google show that upgrading your connection from 1Mbps to 2Mbps halves web page load times, but quickly thereafter we are into diminishing returns: upgrading from 5Mbps to 10Mbps results in a mere 5% improvement in page load times.

To help more Internet users with understanding this seeming conundrum, I’ve been working with the lovely folks over at the RITE project to develop a new video, released today, that seeks to explain the difference between bandwidth and delay, and the different ways that latency affects Internet performance.

Why not take a look at the video and then see how you do in the quiz? There are even more resources for teaching and other activities available at the RITE project website.

It’s great to see these educational efforts being launched, as they start to address some of the important actions identified during the Reducing Internet Latency workshop we held last year. There’s still lots more to be done though and, “getting smarter queue management techniques more widely deployed” remains a priority.

So check out the video today and share it with your contacts to help us all get educated about this challenge. You can also keep watching our ITM blog for more posts about tackling the scourge of network latency!

Enjoy the video!