Design and Concept of Ad-hoc Filesystem

Kinuko YASUDA; Tatsuya HAGINO

Design and Concept of Ad-hoc Filesystem

Kinuko YASUDA, Tatsuya HAGINO

Full Text Views

0

Cite this

Summary :

In this paper, we propose a new filesystem named Ad-hoc Filesystem which targets ad-hoc wireless networks. Ad-hoc Filesystem is a serverless filesystem which automatically generates temporary shared space among multiple mobile machines when they gather in a communicable range. The generated space is freely available for participating machines and can be used for any temporary work there. The design of Ad-hoc Filesystem is strongly motivated by the recent wireless network technologies and advances of ad-hoc network techniques. However, targeting such an unstable network forces us a completely different design from existing filesystems. We have designed an initial prototype of Ad-hoc Filesystem and have evaluated the simulation results. Our design is strongly based on an assumption such that people who wish to work together would form a stable group in a range, and replicating data on two different machines would be enough to provide a proper level of availability. Hence, Ad-hoc Filesystem distributes files among multiple machines and duplicates them, in order to keep the files available even if some of participating machines arbitrarily leave the communication range. The directory entries are kept as a soft-state mechanism maintained by exchanging broadcast packets. The simulation results show that using broadcast to maintain system state hardly affects the overall system performance, and optimizations such that delaying replies with broadcast and sensing other packets promiscuously during the delay reduce the overhead significantly. Also the result indicates that our strategy that keeps two replicas for each data is promising to provide availability. Although further investigation based on real implementation remains important work, we believe that the prototype has a contribution to realize the possibility of ad-hoc filesystem by synthesizing number of previous works in areas such as network filesystems, server replication and ad-hoc wireless networks.

Publication: IEICE TRANSACTIONS on Communications Vol.E84-B No.4 pp.920-929

Publication Date: 2001/04/01

Publicized

Online ISSN

DOI

Type of Manuscript: Special Section PAPER (Special Issue on Mobile Multimedia Communications)

Category

Cite this

Copy

Kinuko YASUDA, Tatsuya HAGINO, "Design and Concept of Ad-hoc Filesystem" in IEICE TRANSACTIONS on Communications, vol. E84-B, no. 4, pp. 920-929, April 2001, doi: .
Abstract: In this paper, we propose a new filesystem named Ad-hoc Filesystem which targets ad-hoc wireless networks. Ad-hoc Filesystem is a serverless filesystem which automatically generates temporary shared space among multiple mobile machines when they gather in a communicable range. The generated space is freely available for participating machines and can be used for any temporary work there. The design of Ad-hoc Filesystem is strongly motivated by the recent wireless network technologies and advances of ad-hoc network techniques. However, targeting such an unstable network forces us a completely different design from existing filesystems. We have designed an initial prototype of Ad-hoc Filesystem and have evaluated the simulation results. Our design is strongly based on an assumption such that people who wish to work together would form a stable group in a range, and replicating data on two different machines would be enough to provide a proper level of availability. Hence, Ad-hoc Filesystem distributes files among multiple machines and duplicates them, in order to keep the files available even if some of participating machines arbitrarily leave the communication range. The directory entries are kept as a soft-state mechanism maintained by exchanging broadcast packets. The simulation results show that using broadcast to maintain system state hardly affects the overall system performance, and optimizations such that delaying replies with broadcast and sensing other packets promiscuously during the delay reduce the overhead significantly. Also the result indicates that our strategy that keeps two replicas for each data is promising to provide availability. Although further investigation based on real implementation remains important work, we believe that the prototype has a contribution to realize the possibility of ad-hoc filesystem by synthesizing number of previous works in areas such as network filesystems, server replication and ad-hoc wireless networks.
URL: https://global.ieice.org/en_transactions/communications/10.1587/e84-b_4_920/_p

Copy

@ARTICLE{e84-b_4_920,
author={Kinuko YASUDA, Tatsuya HAGINO, },
journal={IEICE TRANSACTIONS on Communications},
title={Design and Concept of Ad-hoc Filesystem},
year={2001},
volume={E84-B},
number={4},
pages={920-929},
abstract={In this paper, we propose a new filesystem named Ad-hoc Filesystem which targets ad-hoc wireless networks. Ad-hoc Filesystem is a serverless filesystem which automatically generates temporary shared space among multiple mobile machines when they gather in a communicable range. The generated space is freely available for participating machines and can be used for any temporary work there. The design of Ad-hoc Filesystem is strongly motivated by the recent wireless network technologies and advances of ad-hoc network techniques. However, targeting such an unstable network forces us a completely different design from existing filesystems. We have designed an initial prototype of Ad-hoc Filesystem and have evaluated the simulation results. Our design is strongly based on an assumption such that people who wish to work together would form a stable group in a range, and replicating data on two different machines would be enough to provide a proper level of availability. Hence, Ad-hoc Filesystem distributes files among multiple machines and duplicates them, in order to keep the files available even if some of participating machines arbitrarily leave the communication range. The directory entries are kept as a soft-state mechanism maintained by exchanging broadcast packets. The simulation results show that using broadcast to maintain system state hardly affects the overall system performance, and optimizations such that delaying replies with broadcast and sensing other packets promiscuously during the delay reduce the overhead significantly. Also the result indicates that our strategy that keeps two replicas for each data is promising to provide availability. Although further investigation based on real implementation remains important work, we believe that the prototype has a contribution to realize the possibility of ad-hoc filesystem by synthesizing number of previous works in areas such as network filesystems, server replication and ad-hoc wireless networks.},
keywords={},
doi={},
ISSN={},
month={April},}

Copy

TY - JOUR
TI - Design and Concept of Ad-hoc Filesystem
T2 - IEICE TRANSACTIONS on Communications
SP - 920
EP - 929
AU - Kinuko YASUDA
AU - Tatsuya HAGINO
PY - 2001
DO -
JO - IEICE TRANSACTIONS on Communications
SN -
VL - E84-B
IS - 4
JA - IEICE TRANSACTIONS on Communications
Y1 - April 2001
AB - In this paper, we propose a new filesystem named Ad-hoc Filesystem which targets ad-hoc wireless networks. Ad-hoc Filesystem is a serverless filesystem which automatically generates temporary shared space among multiple mobile machines when they gather in a communicable range. The generated space is freely available for participating machines and can be used for any temporary work there. The design of Ad-hoc Filesystem is strongly motivated by the recent wireless network technologies and advances of ad-hoc network techniques. However, targeting such an unstable network forces us a completely different design from existing filesystems. We have designed an initial prototype of Ad-hoc Filesystem and have evaluated the simulation results. Our design is strongly based on an assumption such that people who wish to work together would form a stable group in a range, and replicating data on two different machines would be enough to provide a proper level of availability. Hence, Ad-hoc Filesystem distributes files among multiple machines and duplicates them, in order to keep the files available even if some of participating machines arbitrarily leave the communication range. The directory entries are kept as a soft-state mechanism maintained by exchanging broadcast packets. The simulation results show that using broadcast to maintain system state hardly affects the overall system performance, and optimizations such that delaying replies with broadcast and sensing other packets promiscuously during the delay reduce the overhead significantly. Also the result indicates that our strategy that keeps two replicas for each data is promising to provide availability. Although further investigation based on real implementation remains important work, we believe that the prototype has a contribution to realize the possibility of ad-hoc filesystem by synthesizing number of previous works in areas such as network filesystems, server replication and ad-hoc wireless networks.
ER -