Consider a network of n nodes, in which the cost of sending messages to all
nodes (e.g., a flood) is O(n) and the cost of sending a message to a designated
node is O(pn). Let us denote by De the total number of the detected events, Q
the number of the queries, and Dq the number of the events which are returned
as answers for the Q queries. Furthermore, we assume that the data stored
can either be a list of events of a given type, or a summary of such events.
Table 3 shows the network-wide communication cost, as well as the hot-spot
energy usage for the three schemes. This analysis shows that the data-centric
storage scheme becomes more preferable as the size of the network increases,
or when many more events are generated than can be usefully queried. This
performance advantage increases when summaries are returned as answers.
Thus, the data-centric storage scheme is an attractive alternative as sensor
networks scale.
Table 3. [16].
Storage method Total energy consmption Hot-spot energy consmption
External DepN De
Local Qn + Dqpn Q + Dq
Data-centric Qpn + Depn + Dqpn Q + Dq
4.3 Mechanisms for Data-centric Storage
Our discussion has introduced data-centric storage as a concept, without describing
the specifics of a particular instance of a data-centric storage system.
In this section, we describe a system called a Geographic Hash Table (GHT)
[16, 17] and focus on its mechanistic underpinnings. We introduce GHT, a
geography-based routing protocol GPSR, its usage in GHT, the robustness of
GHT and its structured replication technique.
Pages:
457
458
459
460
461
462
463
464
465
466
467
468
469
470
471
472
473
474
475
476
477
478
479
480
481