NLANR: the National Laboratory for Applied Network Research

Updated November 2026

The National Laboratory for Applied Network Research (NLANR) was a distributed, NSF-funded group that supported the US research networks of the late 1990s and early 2000s. Its teams gave the world iperf, a global web-cache hierarchy that fed into Squid, and years of public measurement data. Its funding ended on June 30, 2006.

About this site. nlanr.net today is an independent educational project about network performance and measurement. It has no connection to the former laboratory, its staff, its host institutions or the National Science Foundation.

Origins

NLANR began in early 1995 as a collaboration among five NSF-sponsored supercomputing centers: the Cornell Theory Center, the National Center for Atmospheric Research (NCAR), the National Center for Supercomputing Applications (NCSA) in Illinois, the Pittsburgh Supercomputing Center (PSC) and the San Diego Supercomputer Center (SDSC). They were also the first sites on the NSF’s new vBNS research backbone, and NLANR provided engineering, technical support and coordination for those connections. In 1996 NSF funded a proposal, led by Hans-Werner Braun at UC San Diego, to create a distributed NLANR.

The teams

From late 1997 NSF funded NLANR’s work as separate teams, each with its own cooperative agreement, to support the growing number of universities joining the research networks through NSF’s High Performance Connections program (more than 170 sites by 2001).

NLANR teams and their NSF awards
Team	Home	Work
Distributed Applications Support Team (DAST)	NCSA, University of Illinois	Helped researchers get performance from their applications; developed the original iperf and promoted tools such as Netlog. Award from November 1997.
Measurement and Operations Analysis Team (MOAT), later Measurement and Network Analysis (MNA)	UC San Diego / SDSC	Active and passive measurement of the research networks, working with UCSD’s CAIDA; by 2002 about 150 measurement devices. Awards from April 1998 and July 2002.
National Center for Network Engineering (NCNE), later Engineering Services	Carnegie Mellon University / PSC	Network engineers working directly with campus staff, workshops and a help desk for connecting sites. Award from January 1998.

Today’s iperf3 is a from-scratch redesign by ESnet and Lawrence Berkeley National Laboratory of the tool first written at NLANR/DAST. See our iperf guide.

Web caching and Squid

In 1995 NSF funded “A Distributed Testbed for National Information Provisioning”, led by kc claffy, to build a caching system across strategic US network locations; a follow-on award at UC San Diego (1996–1999) listed Duane Wessels as a co-investigator. The resulting NLANR cache hierarchy, later known as IRCache, ran from December 1995 and by 1998 involved about 1,100 caches in 40 countries. Wessels and claffy documented the Internet Cache Protocol in RFC 2186 and 2187, and the Squid proxy grew up alongside the project. See how web caching works and Squid.

The end of NLANR

NSF funding for the NLANR project expired on June 30, 2006, and NSF had no plans to continue it. From July 2006, CAIDA at UC San Diego took over operational stewardship of NLANR’s machines and data, to audit them and retire obsolete hardware.

More history

vBNS and research networksHow the NSF and MCI built the vBNS research backbone in 1995, how it grew from 155 Mbit/s to 2.4 Gbit/s, Internet2's Abilene network, and today's research networks.