Turning the Postal System into a Generic Digital Communication Mechanism

Report ID: TR-688-04
Author: Li, Kai / Martonosi, Margaret / Wang, Randolph Y.
Date: 2004-01-00
Pages: 20
Download Formats: |PDF|
Abstract:

Recent studies find that as the adoption rates of the Internet and broadband connections slow down in the U.S., the ``digital divide'' could be solidifying: those with modest incomes, rural residents, and minorities are among those who lag behind in Internet access. Most people living in developing regions, who represent an overwhelming majority of the world's population, also largely fall on the ``wrong'' side of the digital divide and are missing out on many important opportunities that the digital world has to offer.

Bridging this digital divide, especially by attempting to increase the accessibility of broadband connectivity, can be very challenging. The improvement of wide-area Internet bandwidth is constrained by factors such as how quickly we can dig ditches to bury fibers in the ground. The cost of furnishing ``last-mile'' wiring can be prohibitively high, and the progress has been excruciatingly slow.

In this proposal, we explore the use of digital storage media (such as DVDs or even hard disks) transported by the postal system as a general digital communication mechanism. While the idea of sending digital content via the postal system is not a new idea -- companies (such as AOL.com and netflix.com) have used this approach to deliver software and movies on a large scale for some time -- none of these existing attempts have turned storage devices delivered by the postal system into a generic communication channel that can cater to a wide array of applications. We shall call such a system a Postmanet.

Compared to traditional wide-area connectivity options, the Postmanet has several important advantages. These include wider reach, greater bandwidth potential, low cost, and better scalability. In terms of reach, the postal system is a truly global ``network'' that reaches a far greater percentage of the world's human population. In terms of its bandwidth, the Postmanet advantage stems from fundamental technology trends: the explosive growth of storage technology density implies that the amount of information that can fit into a fixed amount of volume, or the amount of information that can be shipped by the postal system for a fixed cost, increases exponentially at roughly the same rate, a rate that the wide-area Internet bandwidth growth is unlikely to be able to keep up with. Indeed, far from being a temporary fluke, the bandwidth gap between the Internet and the Postmanet is likely to widen as the storage density continues its rapid improvement.

We would like to use the Postmanet to extend and complement the existing Internet to support a wide array of bandwidth-intensive applications. These may include email with large attachments, web pointing to or embedded with large data objects, remote file system mirroring for sharing and/or backup, peer-to-peer file sharing, video ``almost on-demand,'' publish/subscribe systems for other types of content, and distance learning.

At a first glance, it may appear deceptively simple for an end user to manually burn and ship DVDs for whatever applications he desires. But with many applications, hundreds of communicating parties, and thousands of ``messages,'' for example, manual management would become infeasible. Therefore, systematic transparency support at the levels of network transport, programming models, routing, and applications is necessary. We shall see two recurring themes at these different levels of the system. One is the simultaneous exploitation of the low-latency low-bandwidth Internet and the high-latency high-bandwidth Postmanet so we can combine their latency and bandwidth advantages. The other is the exploitation of the abundant capacity and bandwidth of the Postmanet to improve its latency, reliability, and cost.