Citation Context

...ns and protocols for debugging and verification. Like with standard logging subsystems, developers can integrate X-Trace into their applications. It is actually being used by the team developing DONA =-=[16]-=-, a content-based routing scheme for the Internet. 6 Discussion While X-Trace has many uses, it also has limitations. We discuss those here, as well as other interesting aspects. Evaluation The exampl...

Citation Context

...ogging [8], en∗ Work done while this author was an intern at Intel Labs Berkeley. cryption [1], filtering and intrusion detection [14], as well as a variety of more forward-looking research proposals =-=[21, 36, 39]-=-. In current networking equipment, however, high performance and programmability are often competing goals—if not mutually exclusive. On the one hand, high-end routers, because they rely on specialize...

Citation Context

... system, which has the responsibility of handling the matching between the publishers and subscribers. The rendezvous does not need to differ substantially from other topic-based pub/sub systems; cf. =-=[15, 23, 36]-=-. Whenever it identifies a publication that has both a publisher (or an up-to-date cache) and one or more active subscribers, it requests the topology system to construct a logical forwarding tree fro...

Citation Context

...ontent in the networking communication model. The fundamental principles of CCN have been investigated by a number of researchers over the last decade. Populard esigns are TRIAD [15], ROFL [11], DONA =-=[17]-=-, and more recently NDN [3, 16]. Key CCN features are: (1) content items have an identifier and not content hosts, (2) routing leverages content identifiers and not host identifiers for forwarding ope...

Citation Context

...of content to trust in the host that stores it, and then additionally requires secure mechanisms to identify, locate and communicate with that host. This is widely recognized as a significant problem =-=[11, 21]-=-. Because the “trust” the user gets in the content they depend upon is tied inextricably to the connection over which it was retrieved, that trust is transient – leaving no reliable traces on the cont...

Citation Context

...can refer to another not only by its name, but also by the cryptographic digest of its contents (forming effectively a self-certifying name [23, 25, 11, 8]), or by the identity (key) of its publisher =-=[8, 29, 22, 21]-=-). This allows individual signed pieces of content to effectively certify other pieces of content they (securely) refer to; each piece of content the user encounters acts as a potential piece of evide...

Citation Context

...-4/12/08 ...$15.00. 1. INTRODUCTION Naming content objects directly, instead of their respective end-hosts, gives the opportunity to identify content objects as they travel from source to destination =-=[16, 13, 12]-=-. In turn, given that the network transfers named objects (instead of unidentifiable data containers, i.e., IP packets), these objects can be cached in the network and be forwarded to subsequent users...

Citation Context

...idered a cache hit and the data-packet is forwarded to the requester. PSIRP [17], as another content-centric proposal, uses a different approach to data naming. It follows an approach similar to DONA =-=[13]-=-’s flat labels, using a combination of cryptographic measures; it uses a cryptographic fingerprint computed over the piece of content or some other similar, algorithmically generated identifier. The c...

Citation Context

...everal caches linked together, allowing each cache to also forward requests to its neighbors when needed. Common examples of such systems are hierarchical web and file system caches. Recently [1] [2] =-=[3]-=-, there have been proposals for multi-cache designs over networks of Internet-like scale and structure. Determining the performance of a specific multi-cache system is extremely difficult. Even for th...