COMPLEX NETWORK ANALYSIS AND INTELLIGENT PLAFORM
Service provider networks possess a broad spectrum of information (network intelligence) relative to network users such as subscribers and their data flows. Attempts to collectively and collaboratively leverage this intelligence have, in the past, fallen short with many shortcomings not being well addressed. A key challenge is the scale of data from a network that encompasses millions of flows and terabytes of traffic. The potential gains associated with network intelligence is significant with key benefits addressing network performance, optimizing infrastructure, and detecting malicious traffic. Data Products LLC will develop efficient technologies to collect, manage, and analyze such distributed network complex BigData that is a challenge to many organizations including the scientific community. This new solution includes developing advance machine learning methods such as network analytic algorithms, and other computational intelligence techniques on massive amount of complex network data generated within datacenters for tangible business intelligence and insights as specified Phase I topic 1.a. call. We advocate a platform to address the challenges of monitoring and managing networks by leveraging advance machine learning methods such as network analytic algorithms, and other computational intelligence techniques on vast amount of complex network data for tangible business intelligence and insights. Types of complex data collected (as allowed by the enterprise network policy) will include users’ traffic logs, network management data, network flow data, PerfSONAR service type data (throughput; one-way delay and the resulting loss, trace-route), other types of active and passive network management measurement data, real time and historic structured and unstructured data, geo data, images etc.
- Research Organization:
- Data Products LLC
- Sponsoring Organization:
- USDOE Office of Science (SC), Advanced Scientific Computing Research (ASCR). Scientific Discovery through Advanced Computing (SciDAC)
- DOE Contract Number:
- DE-SC0018490
- OSTI ID:
- 1877991
- Type / Phase:
- SBIR (Phase II)
- Report Number(s):
- DOE18490
- Resource Relation:
- Related Information: 1. “Synergistic Challenges in Data-Intensive Science and Exascale Computing”, DOE ASCR Data Subcommittee Report 20132. A. Hanemann, J. Boote, E. Boyd, J. Durand, L. Kudarimoti, R. Łapacz, D. Swany, S. Trocha, and J. Zurawski. PerfSONAR: A service oriented architecture for multi-domain network monitoring. Service-Oriented Computing, 2005.3. Juve G, Deelman E (2011) Scientific workflows in the cloud. In: Grids. Clouds and Virtualization.4. Eli Dart, Mary Hester, Jason Zurawski, “Basic Energy Sciences Network Requirements Review - Final Report 2014”, ESnet Network Requirements Review, September 2014, LBNL 6998E5. Eli Dart, Mary Hester, Jason Zurawski, “Fusion Energy Sciences Network Requirements Review - Final Report 2014”, ESnet Network Requirements Review, August 2014, LBNL 6975E6. Eli Dart, Mary Hester, Jason Zurawski, Editors, “High Energy Physics and Nuclear Physics Network Requirements - Final Report”, ESnet Network Requirements Workshop, August 2013, LBNL 6642E7. Eli Dart, Brian Tierney, Editors, “Biological and Environmental Research Network Requirements Workshop, November 2012 - Final Report””, November 29, 2012, LBNL LBNL-6395E8. David Asner, Eli Dart, and Takanori Hara, “Belle-II Experiment Network Requirements”, October 2012, LBNL LBNL-6268E9. Eli Dart, Brian Tierney, editors, “Advanced Scientific Computing Research Network Requirements Review, October 2012 - Final Report”, ESnet Network Requirements Review, October 4, 2012, LBNL LBNL-6109E10. Quinlan, J.R. (1993); C4.5: Programs for Machine Learning, Morgan Kaufmann Publishers Inc., ISBN 1558602402.11. Deployments of Network Monitoring Software perfSONAR Hit 1,000. http://1.usa.gov/Qt94Nk.12. perfSONAR Deployment on ESnet. Brian Tierney. Presented at AIMS Workshop, CAIDA. 2011.13. F. Feather, D. Siewiorek, and R. Maxion. Fault detection in an Ethernet network using anomaly signature matching. In ACM SIGCOMM CCR, 1993.14. Yu, M., GRreenberg, A. G., Maltz, D. A., Rexford, J., Yuan, L., Kandula, S., A and Kim, C. Profiling network performance for multi-tier data center applications. In NSDI (2011).
- Country of Publication:
- United States
- Language:
- English
Similar Records
SDN-NGenIA, a Software Defined Next Generation Integrated Architecture for HEP and Data Intensive Science
Predictive Analytics-based U.S. Inland Waterways Voyage Planning Analysis Tool (VPAT)