Distribution Systems With a Concentration

¶ … Distribution Systems with a Concentration on Cyber Foraging

Cyber foraging and the quantification of trust in context-aware Web Applications: A Performance Assessment of Caching Trust vs. Hoarding Content

Description

The use of cyber foraging as a mechanism for enabling Internet-aware mobile devices of all types to interconnect with data staging servers shows significant potential for creating composite applications in real time, in addition to updating content taxonomies defined by mobile device users. Content taxonomies could potentially include social networking, consumer-generated media feeds including blogs, and news feeds enabled through RSS. The definition and fine-tuning of content taxonomies on mobile devices is in its nascent stages today yet growing rapidly due to the sophistication of mobile devices; however a corresponding operating architecture with greater breadth and agility is needed. Caching trust, or the ability to sufficiently cache knowledge vs. hoarding content is a fundamental design decision in this architecture, and the design dichotomy addressed in these suggested research topics.

Source(s)

M. Satyanarayanan, Caching Trust Rather Than Content. M. Satyanarayanan. School of Computer Science, Carnegie Mellon University, Operating Systems Review. Volume 34, No. 4, October 2000

The author of this paper presents the concept of caching as it has typically applied to the performance of computing systems, mentioning the concept of how this concept decreases latency and makes network failures transparent at the user level of networks. The author contends that in mobile-based environments that would have the ability to deliver composite application components, entire applications and a wide variety of content-based applets, RSS and real-time newsfeeds through cyber foraging-based delivery mechanisms with data staging servers there is the fundamental decision of whether to hoard or cache content, and further, the author introduces the concept of caching trust.

In this case, it is the assumption that all data of potential interest can be hoarded on the mobile client [1, 2, 4]. In other words, such clients have to be prepared to cope with cache misses during normal use. If they are able to cache trust, then any untrusted site in the fixed infrastructure can be used to stage data for servicing cache misses -- one does not have to go back to a distant server, nor does one have to compromise security. (p. 1)."

Satyanarayanan goes on to mention the concepts of privacy being protected through the use of per-file private encryption keys to protect content on data staging servers and the use of a digital fingerprint to authenticate the content for use on the remote server. Satyanarayanan defines the security method as follows:

It is simple to extend this idea to privacy. In addition to a fingerprint, the client also hoards a per-file private encryption key. The server encrypts each file before staging it on the surrogate. To handle a cache miss at the remote site, the client fetches the data from the surrogate, decrypts it, verifies its fingerprint and then uses the data. The volume of cached keys can be reduced by using a single private encryption key for all files, at the price of total exposure if that key is broken. (p.2)."

The author concludes the article with mention of the Aura Project at Carnegie Mellon, an initiative which is aimed at defined distraction-free, ubiquitous computing and support for nomadic access using the Coda File System, which has recently been updated to support efficient update propagation over low-bandwidth networks [3]. Aura is a commonly referenced concept in much of the cyber foraging literature, and supports the concept of increasingly complex content taxonomies being synchronized with applications and data sources through the use of data staging servers.