Srivathsan Srinivasagopalan Machine Learning | Cybersecurity | Applied Research

Router Arbiter Project: RAToolSet circa 1999–2000

An automatic router configuration tool for Bay Routers that works with any of the RADB servers around the world that runs RPSL. It analyzes the routing policies registered in the Internet Routing Registry (IRR) and produces router configuration files. This is an Open Source project and is part of the IRRToolSet. The project originated at USC Information Sciences Institute (1997–2001) before transitioning to ISC, where RIPE NCC proposed its continuation.

NetSNMP — Perl5 SNMP Extension Module circa 1999–2000

A Perl5 extension module for the Net-SNMP library. It provides a comprehensive, tri-lingual SNMP API supporting SNMPv3, SNMPv2c, and SNMPv1, along with interface access to the SMI MIB parse-tree for runtime retrieval of parsed MIB data. Its internals depend on the Net-SNMP toolkit library (formerly ucd-snmp).

The Perl SNMP 5.x module included with net-snmp 5.0+ behaves like a proper net-snmp application — it automatically reads configuration files, applies defaults, and parses MIB files. The SNMP module internals rely on the Net-SNMP toolkit library (previously known as ucd-snmp).

Spatial Modeling Environment circa 2004

An ecological modeling tool built by University of Maryland in collaboration with University of Vermont. SME is a modular, collaborative, and spatially explicit ecological-economic modeling tool, heavily associated with the Gund Institute for Ecological Economics. It links icon-based modeling interfaces (like Stella or Simile) with high-performance computing to simulate landscape-scale changes, such as hydrology and biodiversity, by integrating economic decisions with spatial ecological data.

Purpose

Build integrated models of human and ecological systems, enabling “what-if” analyses regarding environmental policy, land use, and landscape change.

Development

Originated from research led by Robert Costanza and others at the University of Vermont.

Core Functionality

Facilitates the creation of complex, modular models that can be run on distributed computing systems, handling large spatially explicit datasets.

Applications

Key projects included modeling the Patuxent River watershed, integrating agricultural practices with ecological services, and creating spatially explicit economic models.

Component Integration

Enables coupling of heterogeneous models — where one part of a model might be a differential equation and another is a rule-based spatial simulation.

My Contribution

Developed compatibility features and visualization capabilities for ArcGIS and terrestrial image formats.