Recent Posts

More Posts

Internet architecture and Internet-centered research being a global enterprise, I spend between four and seven weeks a year on the road, depending on which year, your definition of road and your definition of week, and a fair amount of time in teleconferences in various timezones in the time in between. One of the fixtures in my calendar is the thrice-annual meeting of the Internet Engineering Task Force (IETF), taking place right now in Chicago. I’ve only missed three such meetings in the past dozen years, and each time I do I attempt to take part via Internet as best I can. Here are my reflections about well it’s working this time around, how it’s improved, and how it could improve further. For in a world where those who steadfastly believe in borders and walls seem to be gaining the upper hand, it seems prudent to prepare to do the work of Internet architecture, engineering, and standardization without the benefit of free movement of the people doing it.

CONTINUE READING

On the shores of Lake Sarnen in central Switzerland, there’s a museli factory. (Of course there is.) It makes many different kinds of muesli for various markets. One of these is an organic chocolate-amaranth concoction that’s basically the only thing my daughter will eat for dinner this week. I happened to glance at the ingredients, and it occurred to me that there are basically three kinds of people in the world.

CONTINUE READING

Wasting time at Christmas by burning the site to the ground and starting over seems to be a tradition around here…

CONTINUE READING

Wow, that year went quickly, on which more later.

I’d wanted to try my hand at brewing for a while, but was put off it by the (accurate) fear than ninety percent of the work was washing bottles and cleaning pots. Then, last winter, as a newly-minted father of a baby with an age measured in weeks, life consisted mainly of sterilizing bottles and not sleeping. I made an offhand comment to the effect that if I was going to spend so much time boiling glass I might as well make beer. Ariane gave me a starter kit, and a year later I’m about seventy liters in and think I have a reasonable clue what I’m doing.

CONTINUE READING

I’m off to New York in a couple of weeks to present a paper at PAM (which I mentioned here, though sadly the flashy automated demo I was hoping to build was a bit optimistic). The question: “is it safe to turn on ECN on client machines by default, completing the end to end deployment of a simple fifteen year old protocol to give us a better way to signal network congestion than simply dropping packets on the floor?” The answer is: “define safe.” Our key findings:

CONTINUE READING

In German, there’s a word for an organization which takes its mission very seriously but is adorably incompetent at it: “Kaninchenzüchterverein” (lit. “rabbit-breeding club”). There’s another word for an organization which is bad at what it does because nobody cares: “Saftladen” (lit. “juice shop”).

CONTINUE READING

The issues identified in of part one of this post led to yet another search for solutions to the problem of making (especially passive) measurement repeatable. Of course, this has been done before, but I took as an initial principle that the social aspects of the problem must be solved socially, and worked from there. What emerged was a set of requirements and an architecture for a computing environment and set of associated administrative processes which allows analysis of network traffic data while minimizing risk to the privacy of the network’s end users as well as ensuring spatial and temporal repeatability of the experiment. For lack of a better name I decided to call an instance of a collection of data using this architecture an analysis vault.

CONTINUE READING

Part one of this post painted a somewhat bleak picture of the state of Internet measurement as a science. The dreariness will continue later this month in part two. And yet there seems to be quite a lot of measuring the Internet going on. It can’t all be that bad, can it?

CONTINUE READING

Mail is broken. This is nothing new. RFC 822, after all, wasn’t the beginning of Internet e-mail, merely an attempt to fix it, which admittedly worked reasonably well for a while. But even with all the brokenness of mail, I wasn’t expecting to dig into my Postfix logs today to find that USENIX couldn’t send me mail because the firm they’ve outsourced to was too lazy to create IN PTR records for their nodes in the cloud.

CONTINUE READING

In the back of the pantry at the house I grew up in in Memphis, there was always a stack of little plastic tubs of dried candied “fruits” of various colors (I say “colors” because the flavor was invariably “sugar”). My mother was never much of a baker, except at Christmas, when the baking would take two forms: fruitcake and stollen, both of which were filled with candied fruit. I’d try Mom’s fruitcake, the main ingredient of which seemed to be brandy, about once every five years to see if I was finally old enough to enjoy it. I never quite made it.

Stollen, on the other hand, was the main course of most breakfasts around Christmas. This was a bit odd in Memphis, doubly so because we didn’t have any particularly German ancestors; Mom just saw the recipe in a magazine sometime in the late 70s or early 80s and decided to make a tradition out of it. So I was thrilled when I moved to Switzerland and found out you could buy stollen in the grocery store at Christmastime. Almost as thrilled as I was disappointed when I found out that “real” Stollen is basically a marzipan delivery system.

CONTINUE READING

Projects

Selected Publications

We seek to generalize the idea of measurement within protocols, e.g., the way in which TCP relies on measurement to drive its end-to-end behavior. Rhetorically, we pose the question "what if the stack had been built with measurability and diagnostic support in mind?". We start from a set of principles for explicit measurability, and define primitives that, were they supported by the stack, would not only provide a solid foundation for protocol design going forward, but also reduce the cost and increase the accuracy of measuring the network.
ACM Computer Communication Review, April 2017

Explicit Congestion Notification (ECN) is an TCP/IP extension to signal network congestion without packet loss, which has barely seen deployment though it was standardized and implemented more than a decade ago. On-going activities in research and standardization aim to make the usage of ECN more beneficial.This measurement study provides an update on deployment status and newly assesses the marginal risk of enabling ECN negotiation by default on client end-systems. Additionally, we dig deeper into causes of connectivity and negotiation issues linked to ECN. We find that about five websites per thousand suffer additional connection setup latency when fallback per RFC 3168 is correctly implemented; we provide a patch for Linux to properly perform this fallback. Moreover, we detect and explore a number of cases in which ECN brokenness is clearly path-dependent, i.e. on middleboxes beyond the access or content provider network. Further analysis of these cases can guide their elimination, further reducing the risk of enabling ECN by default.
In PAM 2015

The Internet's transport layer - the SOCK_STREAM service from TCP and the SOCK_DGRAM service from UDP - has seen little evolution over the past three decades, despite wildly changing requirements. Indeed, the movement of the waist of the protocol stack hourglass from IP up the stack toward HTTP (over TLS) over TCP has combined with a proliferation of middleboxes that make stringent assumptions about the structure of the traffic they will pass to reduce protocol diversity over time. This ossification has reduced our ability to evolve transport protocols to meet these new application requirements. In this work, the authors describe aspects of this problem and propose a solution space and agenda for improving the situation.
In IEEE Internet Computing, Sep. 2014

Recent Publications

More Publications

  • Post Sockets - Towards an Evolvable Network Transport Interface
    Proc. IFIP/IEEE Networking Workshop on Future Internet Transport

    Details PDF

  • Principles for Measurability in Protocol Design
    ACM Computer Communication Review, April 2017

    Details PDF

  • Services Provided by IETF Transport Protocols and Congestion Control Mechanisms
    RFC 8095

    Details PDF IETF

  • UDP Bonding at Layer 3
    In ANRW 2016

    Details PDF

  • Report from the IAB Workshop on Stack Evolution
    RFC 7663

    Details PDF IETF

Recent Talks