The goal of the Scalable Distributed Information Management System
(SDIMS) project is to develop a "distributed operating systems control
plane" that will serve as the backbone for a large-scale distributed
services. SDIMS aggregates information about large-scale
networked systems to provide detailed views of nearby information and
rare events and summary views of common, global information.
To serve as a basic building block, a SDIMS should have
four properties: scalability to many nodes and attributes, flexibility
to accommodate a broad range of applications, support administrative
autonomy and isolation, and robustness to node failures and
disconnections. To meet these goals, or SDIMS system (1)
uses techniques from Distributed Hash Table (DHT) literature to create
scalable aggregation trees, (2) provides flexibility through a simple
API that lets applications control propagation of reads and writes,
(3) provides autonomy and isolation through simple augmentations of
current DHT algorithms, and (4) is robust to node and network
reconfigurations through lazy reaggregation, on-demand reaggregation,
and tunable spatial replication.
Papers and Posters
- "Shruti: A Self-Tuning Hierarchical Aggregation System,"
P. Yalagandula, M. Dahlin, First IEEE International Conference on
Self-Adaptive and Self-Organizing Systems, July 2007 . pdf
- "STAR: Self-Tuning Aggregation for Scalable Monitoring,"
N. Jain, D. Kit, P. Mahajan, P. Yalagandula, M. Dahlin, Y. Zhang,
In review, Mar 2007 . preliminiary
draft (pdf) bibtex
- "PRISM: PRecision Integrated Scalable Monitoring," N. Jain,
D. Kit, D. Mahajan, M. Dahlin, Y. Zhang, In review, Feb 2007
- Online Aggregation over Trees . C. Greg Plaxton, Mitul
Tiwari, and Praveen Yalagandula.In Proceedings of the 21st
International Parallel and Distributed Processing Symposium
(IPDPS), March 2007. Conference version
[PDF] Extended version
Administrative Autonomy in Structured Overlays
Praveen Yalagandula and Mike Dahlin. .
Short version [PDF]
Extended version [PDF]
A Scalable Information Management Middleware
for Large Distributed Systems
Praveen Yalagandula. PhD Thesis
Shruti: Dynamically Adapting
Praveen Yalagandula and Mike Dahlin. Work-in-Progress
session. At OSDI 2004. Slides in [PPT] and [PDF]
A Scalable Distributed
Information Management System.
Praveen Yalagandula and Mike Dahlin. In Proceedings
of ACM SIGCOMM, August, 2004.
- SDIMS: A Scalable Distributed
Information Management System, Praveen Yalagandula. Ph.D. Proposal. February, 2004.
Designing a Scalable
Distributed Information Management System
Praveen Yalagandula and Mike Dahlin. Poster at SOSP 2003 [