Wednesday, December 29, 2010

Foundation End-of-Year Fund Raising Drive Final Plea

Thank you for all of your generous donations! We have raised $226,066 so far this year. We have three days left to meet our goal of $350,000. So, we are making one last plea for donations this year.

Why do we need to raise so much money? Here is a list of some of our accomplishments and where our money went this year:

Provided $100,000 in grants for projects that improve FreeBSD in the areas of:
- DTrace support
- High availability storage
- Enhanced SNMP reporting
- Virtualization and resource partitioning
- Embedded device support
- Networking stack improvements

Allocated $50,000 for equipment to enhance FreeBSD project infrastructure.

Sponsored 8 FreeBSD related conferences.

Funded 16 travel grants giving increased community and developer access to conferences.

Provided legal support to the FreeBSD project.

By meeting our fund-raising goal, we will be able to increase our project development funding, purchase more needed equipment for the Project's infrastructure, and support more FreeBSD related conferences.

Please help us continue to support the FreeBSD Project and community by making a donation to The FreeBSD Foundation.

To make a donation, please go to:

Thank you for your continued support of the FreeBSD Foundation.

Tuesday, December 21, 2010

Feed-Forward Clock Synchronization Algorithms Project

The FreeBSD Foundation is pleased to announce that Julien Ridoux and Darryl Veitch at the University of Melbourne have been awarded a grant to implement support of feed-forward clock synchronization algorithms.

The Network Time Protocol (NTP) is widely used for synchronization over the network and the ntpd daemon is the current reference synchronization algorithm. The system clock in FreeBSD is currently designed with ntpd in mind, leading to strong feedback coupling between the kernel and the synchronization daemon.

The RADclock is an example of an alternative class of synchronization algorithms based on feed-forward principles. This project will provide the core support for feed-forward algorithms, so that alternatives to ntpd can be developed and tested. The central motivation for this is the strong potential of such approaches for highly robust and accurate synchronization.

Beyond this, virtualization is one of the next major challenges faced by time keeping systems. The current feedback synchronization model is complex and introduces its own dynamics, an approach that is not suited to the requirements of virtualization. Feed-forward based synchronization offers a cleaner and simpler approach, which is capable of providing accurate time keeping over live migration of virtual machines.

This project will conclude in March 2011.

Friday, December 17, 2010

End-of-Year Newsletter

We are pleased to announce the publication of the FreeBSD Foundation's End-of-Year Newsletter which contains examples of how we have supported the FreeBSD Project and community this year.

It's not to late to make a donation to the Foundation for 2010. Thank you to everyone for their support and we wish you a happy holiday season and best wishes for the new year.

Saturday, December 4, 2010

Five new TCP Congestion Control Algorithms Project

The FreeBSD Foundation is pleased to announce that Swinburne University's Technology's Centre for Advanced Internet Architectures has been awarded a grant to implement five new TCP congestion control algorithms in FreeBSD.

Correctly functioning congestion control (CC) is crucial to the efficient operation of the Internet and IP networks in general. CC dynamically balances a flow's throughput against the inferred impact on the network, lowering throughput to protect the network as required.

The FreeBSD operating system's TCP stack currently utilizes the defacto standard NewReno loss-based CC algorithm, which has known problems coping with many aspects of modern data networks like lossy or large bandwidth/delay paths. There is significant and ongoing work both in the research community and industry to address CC related problems, with a particular focus on TCP because of its ubiquitous deployment and use.

Swinburne University of Technology's ongoing work with FreeBSD's TCP stack and congestion control implementation has progressively matured. This project aims to refine their prototypes and integrate them into FreeBSD.

The project will conclude in January 2011.

Addendum:

The five protocols are: