Information Technology Reference
In-Depth Information
Figure 6. Effect of communication range
Figure 7. Effect of cache size
data objects are cached locally and shared in the
neighborhood. The figure also shows that even
when the mobile peers have large cache size (e.g.
more than 40% of its access range), CPIX and ACP
still outperform DGCoca and the performance
difference is fairly stable.
the data availability on other peers and the data
availability on the broadcast channel. While ACP
is superior to CPIX in most cases, it consumes
more energy. We believe that CPIX and ACP are
two representative cooperative cache management
schemes designed for highly dynamic mobile
peers in broadcast environments.
Recently, many routing protocols have been
designed to facilitate the multi-hop communica-
tion among mobile peers in the form of a MANET
(Mobile Ad-Hoc Networks). Researchers are also
working on P2P overlays for mobile peers. It will be
interesting to investigate whether it is feasible and
how to use multi-hop communication to further
improve the performance of collaborative caching
and prefetching in a highly dynamic environment.
We believe the challenges are keeping the mobile
p e e r s a u t o n o m o u s a n d l i m i t i n g t h e o v e r h e a d (t r a f -
fic) of information synchronization.
concluSion
In this chapter, we have reviewed the problem
of cooperative cache management in mobile
environments that support data broadcast. We
discussed the difference between access-driven
caching and proactive prefetching (from the
broadcast channel), and presented our view of
the challenges for cooperative caching and coop-
erative prefetching in such environments. After
reviewing some solutions based on group-based
cooperative cache management, we presented
two schemes in detail, namely CPIX and ACP.
CPIX is a cooperative caching scheme and ACP
is a cooperative prefetching scheme. They differ
from other schemes mainly in two ways. First,
they do not require a mobile peer to have stable
neighbors. This enables a mobile peer to make
caching (or prefetching) decision autonomously,
and does not impose high information synchro-
nization overhead. Second, they consider both
referenceS
Acharya, S., Alonso, R., et al. (1995). Broadcast
disks: Data management for asymmetric com-
munication environments. In Proceedings of
the ACM SIGMOD International Conference on
Management of Data .
Search WWH ::




Custom Search