Citation Context

...th an emulated wide-area network. This methodology is particularly important with regard to the choice of reactive versus periodic recovery as described above. Existing studies of churn in DHTs (e.g. =-=[7, 8, 16, 19]) ha-=-ve used simulations that—unlike our emulated network—did not model the effects of network queuing, cross traffic, or message loss. In our experience, these effects are primary factors contributing...

Citation Context

... implementation of proximity neighbour selection [8]. We modified it to exploit heterogeneity (as described in the next section). Studies have shown that other structured overlay maintenance protocols=-=[21, 28]-=- also perform well under churn. Structured overlays map keys to overlay nodes. Overlay nodes are assigned nodeIds selected from a large identifier space and application objects are identified by keys ...

Citation Context

...very efficient distributed systems to lookup identifiers, and in the past it has been demonstrated that they can be made robust against node churn, random failures, and fluctuating network conditions =-=[6, 11]-=-. We devise strategies that make DHTs resilient to malicious nodes trying to poison lookups by providing inaccurate information. We achieve this by routing queries, not only to make them converge fast...

Citation Context

...st of the system. Furthermore, while DHT-based overlays provide a fast lookup() operation, the performance of such networks under churn (rapid series of join() and leave() operations) is questionable =-=[18]-=-. The second theory attempts to address the limitations of the first approach by introducing a dynamic mapping between nodes and content. This dynamic overlay network (as opposed to the largely static...

Citation Context

...in works like [6, 2]. However, within these bounds, performance can vary substantially as a function of different design decisions and configuration parameters. Hence simulation-based studies such as =-=[5, 8, 3]-=- often provide more realistic insights into the performance of DHTs. Relying on an understanding based on simulations alone is however not satisfactory either, since in this case, the DHT is treated a...

Citation Context

... its usefulness to the user, one commonly studied problem in the literature is the ability of P2P networks to stay connected in the face of random failures [1], [2], [6], [8], [10], [11], [12], [13], =-=[15]-=-, [17], [21], [19], [23]. It may be argued that compromised connectivity is one of the most fundamental byproducts of churn that directly affects routing efficiency and other metrics observed by the u...

Citation Context

...al building block in several proposed Internet-scale infrastructures [18, 20, 28, 33, 35]. In such DHT-based systems, availability governs the degree of churn, which impacts stability and performance =-=[21, 29]-=-. Figure 17 shows the setup of our simulation, which is similar to that of the study conducted by Li et al. [21]. Infrastructure servers could be deployed in a federated manner similar to LDNS — for e...