Title: Welcome intro

Senior Manager, Cisco Systems
Chairman, High Tech Committee at American Chamber of Commerce Poland

Title: Update on BSD Certification

Abstract: The BSD Certification Group was founded in 2005 with a mission to create the standard for certifying system adminstrators of BSD systems. This talk will discuss what the Group has accomplished so far, what they are currently working on, and how you can participate in this community based effort. It will also discuss the BSDA certification exam and the upcoming BSDP certification exam.

Title: NoSQL

Abstract: Relational database systems are pervasive and well understood, but for some tasks the spreadsheet model is not sufficiently scalable and applicable. The talk will present alternatives to the mainstream data processing and show where it fits best.

Title: HAST – Highly Available storage for FreeBSD

Abstract: HAST (Highly Available STorage) is a software for synchronous data replication between two cluster nodes working in primary-secondary model.

Data is replicated on block-level (GEOM-level), thus it is independed on file system type and applications that are running on top of it.

When primary node fails, the secondary node should be able to fsck the file system (eg. UFS) and mount it without losing a single bit of data. In case of ZFS file system it is just a matter of importing a pool on secondary node. When primary node comes back, HAST will synchronize only the difference between
the nodes.

Title: The VFS/vnode interface in the FreeBSD kernel

Abstract: The presentation is focused on historical and qualitative discussion of the VFS/vnode and how the FreeBSD kernel implements it. First, a small historical overview about the VFS/vnode interface is presented. Then its actual representation in the FreeBSD kernel, covering its main components (mount and vnode structures, syncer, vnlru daemon, buffer cache, namecache, relationship with neighbor objects like filedescs, struct files and bios). Finally, a summary is given about vfs/vnode races during their life-cycle and how they are handled practically within real-life filesystems implementations, with an eye to future development and things not done due to lack of time (exports, stacked filesystems, etc).

Title: What ideas can FreeBSD borrow from AIX?

Abstract: AIX is IBM’s own proprietary UNIX-like operating system. But while it has many advanced features, it appears to be relatively unknown to the wider community, due to both the closed source nature and its requirement to run on proprietary hardware as well. This lecture intends to introduce some of the ideas and concepts present in AIX, that the author considers worthy to be introduced into the BSD world.

Title: FreeBSD on DaVinci DMSoC

Abstract: DaVinci is a family of ARM-based System-on-chips from Texas Instruments with powerful video and multimedia capabilities. This talk is focused on discussion about issues on porting FreeBSD/ARM onto DaVinci chips and challenges of using it’s multimedia capabilities in embedded BSD world.

Title: The new USB stack in FreeBSD

Abstract: The Universal Serial Bus or USB has become a very popular choice for connecting external devices to a PC, laptop, handheld or embedded device. Although found everywhere, not so many people fully understand USB. The new USB stack which I’ve designed for FreeBSD-8 and FreeBSD-9 is not like any other USB stack. In addition to being an USB stack it is also a multiprocessor safe USB stack. In my talk about USB I will try to explain some details about USB itself, the FreeBSD-8/9 USB API’s and general concepts I’ve built around with regard to achieving multiprocessor operation.

Title: mfsBSD

Abstract: mfsBSD is a toolset to create small-sized but full-featured mfsroot based distributions of FreeBSD that store all files in memory (mfs) and load from hard drive, usb storage device, optical media or via network. It can be used for a variety of purposes, including network boot recovery, diskless systems and remote replacement of operating systems. The latest version includes a script called zfsinstall for an automated ZFS-on-root installation of FreeBSD from a booted mfsBSD distribution.

Title: Network emulation using the virtualized network stack in FreeBSD.

Abstract: The capability of running multiple isolated network stack instances has been recently added to the FreeBSD kernel. In this talk we will discuss how a virtualized network stack can serve as a foundation for constructing a scalable network topology emulation platform. Each of the virtual nodes in an emulated network maintains a private view of networking state: link-layer, IPv4 and / or IPv6 addresss and routes, TCP machinery, IPFW, IPSEC etc., while being able to running an independent instance of existing unmodified applications and tools ranging from traffic generators or analyzers to routing protocol daemons or web servers. Virtual nodes can be interconnected into arbitrary topologies using FreeBSD’s netgraph subsystem, making the platform suitable for a wide range of network experimentation and testing scenarios.

Title: Development of high-speed FreeBSD-based solution for Internet traffic management

Abstract: In this presentation I would like to present information and results about the development of in-house Internet traffic management platform which can manipulate (i.e. analyze, route, shape etc.) large volumes of traffic using standard x86 server architecture. It is based on graph theory a la netgraph, but with much simpler graph objects which give rise to much more complex and custom solutions. The software was designed with speed and flexibility in mind and has single- and multi-processor versions. We will also briefly discuss a couple of case studies (e.g. ISP, distributed routing platform, interesting custom solutions etc.) and the kernel and driver modifications that were necessary in order to achieve performance near hardware limits (e.g. completely lock-free MP version).

Title: Using FreeBSD in a Commercial Setting

Abstract: Over the years, many organizations have integrated FreeBSD into their products and operations, using a number of different techniques and strategies. After presenting an overview of the various techniques being employed, the author will discuss:

• strengths and weaknesses of each technique
• best practices for FreeBSD integration
• tools to help automate some of the techniques
• resources for additional information and guidance

In addition, background information about how the FreeBSD project works today and how to effectively contribute to it will be presented.