Research carried out using NetBSD

Network orientated

• TCP successful outer networks (NASA Lewis)
• IPv6/IPsec (KAME)
• Institute for Media Communication (IMK) Value-added Solutions
• TCP Vegas pinch Live Experiments
• NEC Europe Network Laboratories
• NetBSD sets Internet2 Land Speed World Record
• A Scaleable Monitoring Platform for the Internet

Other

• The UVM Virtual Memory System
• Space Acceleration Measurement System II (SAMS-II)
• Massively-parallel And Real-time Storage
• New Callout and Timer Facilities
• Group Awareness successful Distributed Software Development
• Using Application-Driven Checkpointing Logic for Hot Spare High Availability

Network orientated

TCP successful outer networks (NASA Lewis)

NASA Lewis Research Center - Satellite Networks and Architectures Branch usage NetBSD almost exclusively successful their investigation of TCP for usage successful outer networks. They are presently examining respective projected modifications to the Transmission Control Protocol (TCP) protocol. Extensions are being tested successful an effort to amended outer communication. However, the extensions are besides being tested successful terrestrial environments. Among the extensions being tested are restransmission mechanisms based connected selective acknowledgements (e.g., FACK TCP) and TCP pinch larger first windows. They are presently testing the mechanisms' capacity benefits and their fairness to different traffic. They are besides moving pinch the Internet Engineering Task Force (IETF)'s TCP Over Satellite and TCP Implementations Working Groups.

IPv6/IPsec (KAME)

KAME project . A investigation group for implementing IPv6, IPsec and other caller TCP/IP related technologies into BSD UNIX kernels, nether BSD license. KAME stack sewage merged into NetBSD-current character successful June 1999.

Institute for Media Communication (IMK) Value-added Solutions

The Institute for Media Communication (IMK) Value-added Solutions department . The existent group of projects carries retired connected NetBSD:

TCP Vegas pinch Live Experiments

NetBSD TCP Vegas pinch Live Experiments . TCP Vegas is an extended slowstart TCP travel power from Lawrence Brakmo et al. astatine the University of Arizona. Prof. Peter B. Danzig and his group astatine usc.edu ported TCP Vegas to NetBSD 1.0.

NEC Europe Network Laboratories

NEC Europe Ltd. established the Network Laboratories successful Heidelberg, Germany successful 1997, arsenic NEC's 3rd investigation installation in Europe. Research and improvement functions are integrated into the aforesaid statement to shorten the clip to marketplace of cutting-edge web technologies. The laboratories place typical accent connected solutions gathering the needs of NEC Network's European customers.

The Heidelberg labs attraction connected software-oriented investigation and improvement for the adjacent procreation Internet. New connection architectures and protocols supporting multimedia and mobility complete the Internet, together with intelligent Internet services, are the halfway of our work. A mini marketplace investigation squad continuously analyses market trends and marketplace requirements to guarantee that R&D activities reside existent marketplace needs.

NetBSD is utilized successful the Network Laboratories for different task successful the fields of investigation and IPv6 network cognition among different free disposable BSD systems. The readiness of NetBSD for different platforms makes it an perfect campaigner for respective activities, like:

The free disposable sources, the tight coupling of the sources to the good documented 4.4 BSD and the usage of innovative techniques makes NetBSD an fantabulous software investigation platform.

NetBSD sets Internet2 Land Speed World Record

Researchers of the Swedish University Network ( SUNET ) person beaten the Internet2 Land Speed Record utilizing two Dell 2650 machines pinch azygous 2GHz CPUs moving NetBSD 2.0 Beta. SUNET has transferred astir 840 GigaBytes of information in little than 30 minutes, utilizing a azygous IPv4 TCP stream, betwixt a big astatine the Luleå University of Technology and a big connected to a Sprint PoP successful San Jose, CA, USA. The achieved velocity was 69.073 Petabit-meters/second. According to the investigation team, NetBSD was chosen "due to the scalability of the TCP code" .

More accusation astir this grounds including the NetBSD configuration tin beryllium recovered at: http://proj.sunet.se/LSR2/

The website of the Internet2 Land Speed Record (I2-LSR) title is located at: http://lsr.internet2.edu/

A Scaleable Monitoring Platform for the Internet

SCAMPI is a two-and-a-half-year European task to create a scaleable monitoring level for the Internet. It also aims to beforehand the usage of monitoring devices for improving services and technology.

The task develops a web adapter, initially at 10 Gbps speeds, tailored to the needs of monitoring tools. This includes improvement of an unfastened and extensible monitoring architecture to support a unafraid and programmeable shared monitoring infrastructure. It will besides analyse the technical challenges of processing monitoring systems for 100 Gbps speeds and beyond.

The Liberouter COMBO6 Card is considered to beryllium adopted arsenic a base of the SCAMPI hardware monitoring adapter for speeds up to 10 Gbps.

Other

The UVM Virtual Memory System

UVM is simply a caller virtual representation strategy specifically designed to supply the I/O and IPC systems pinch a scope of flexible information activity mechanisms. Implemented successful the NetBSD operating system, UVM wholly replaces the Mach based 4.4BSD VM system. In summation to featuring elastic data activity mechanisms, UVM besides improves virtual memory capacity complete BSD VM successful accepted areas specified as forking and pageout. UVM is implemented wholly wrong the model of BSD and frankincense maintains each the features and modular parts of the accepted Unix situation that programmers person travel to expect.

Space Acceleration Measurement System II (SAMS-II)

SAMS-II Project - Space Acceleration Measurement System II

NASA will beryllium measuring the microgravity environment connected the International Space Station utilizing a distributed system, consisting of:

The first RTS was scheduled to beryllium delivered successful September 1998, for motorboat to the abstraction position successful Jan. 2000.

More accusation tin beryllium recovered successful immoderate mail to the port-i386 mailing list, which besides has immoderate much details connected the hardware.

Massively-parallel And Real-time Storage

Project Massively-parallel And Real-time Storage (MARS)

Cluster based architectures that employment inexpensive Personal Computers (PCs) interconnected by precocious speed commodity interconnect person been recognized arsenic a cost-effective measurement of building precocious capacity scalable Multimedia-On-Demand (MOD) retention servers. Typically, the PCs successful these architectures tally operating systems specified as UNIX that person traditionally been optimized for interactive computing and deficiency accelerated disk-to-network information paths and support for guaranteed CPU and retention access. In this work we study design, implementation and performance measurements of innovative enhancements to 4.4 BSD UNIX carried retired to rectify these limitations successful the discourse of our Massively-parallel And Real-time Storage (MARS) project. We person projected and implemented the following enhancements to a 4.4 BSD compliant nationalist domain NetBSD UNIX operating system:

Our experimental results show that these enhancements supply throughput improvements and QOS guarantees connected the information way from the disk to network.

There is simply a insubstantial describing the OS enhancements that appeared successful IEEE Multimedia98, and location are respective papers connected the "Project MARS-Scalable, Web Based Multimedia-On-Demand Servers and Services" of which this activity was a part:

http://www.ccrc.wustl.edu/~milind/MediaServers.html

New Callout and Timer Facilities

Research carried retired astatine the Washington University in St. Louis: Current BSD kernels (4.4BSD-Lite and derivatives) return clip proportional to the number of outstanding timers (``callouts'') to group aliases cancel timers. This implementation takes changeless clip to start, stop, and support timers starring to a highly scalable creation that tin support thousands of outstanding timers without much overhead. Unlike the existing implementation, it is guaranteed to fastener retired interrupts only for a small, bounded magnitude of time.

Group Awareness successful Distributed Software Development

Open-source package improvement projects are almost ever collaborative and distributed. Despite the difficulties imposed by distance, these projects have managed to nutrient large, complex, and successful systems. However, location is still small known astir really open-source teams negociate their collaboration. In this insubstantial we look at 1 facet of this issue: really distributed developers support group awareness. This paper intends astatine analyzing various Open Source projects (NetBSD, Apache, Subversion) and analyzing group building and communication.

Using Application-Driven Checkpointing Logic for Hot Spare High Availability

Antti Kantee is investigating the usage of application-driven checkpointing logic for basking spare high readiness successful this investigation work, which was successful portion presented astatine EuroBSDCon 2004.