Transparent distributed data management in large scale distributed systems

Soumeya Leila Hernane 1 Jens Gustedt 2
2 CAMUS - Compilation pour les Architectures MUlti-coeurS
Inria Nancy - Grand Est, ICube - Laboratoire des sciences de l'ingénieur, de l'informatique et de l'imagerie
Abstract : In this chapter, we deal with sharing resources transparency in large distributed systems. By using the Data Handover (DHO), together with a peer-to-peer system we provide an easy-to-use architecture to claim resources in dynamic environments: data resources are distributed over a set of peers that may appear and disappear. By means of DHO functions, users request the mapping of data into local memory (for reading or writing) without prior knowledge neither of the location of that data nor of the underlying structure nor of the mobility of peers. This abstraction level is ensured by three managers that interact within our three level architecture. Two algorithms, Exclusive Locks with Mobile Processes (ELMP) and Read-Write Locks with Mobile Processes (RW-LMP), are introduced on the lowest level of the architecture. They ensure data access consistency despite the dynamicity of the environment. Both algorithms satisfy Safety and Liveness properties. Experimental studies show good performance as well as the stability of our approach.
Complete list of metadatas

Cited literature [28 references]  Display  Hide  Download

https://hal.inria.fr/hal-01308989
Contributor : Jens Gustedt <>
Submitted on : Thursday, April 28, 2016 - 4:48:03 PM
Last modification on : Saturday, October 27, 2018 - 1:23:57 AM
Long-term archiving on : Tuesday, November 15, 2016 - 4:15:57 PM

File

chapter.pdf
Explicit agreement for this submission

Identifiers

  • HAL Id : hal-01308989, version 1

Citation

Soumeya Leila Hernane, Jens Gustedt. Transparent distributed data management in large scale distributed systems. Pervasive Computing, Academic Press, pp.153-194, 2016, 978-0-12-803663-1. ⟨hal-01308989⟩

Share

Metrics

Record views

240

Files downloads

135