Soulfood.dk

This talk will be an introduction to some of the storage technologies available today or in the near future in Solaris:

• In-kernel CIFS support
• Remote replication with AVS
• Iscsi and scsi (comstar) target support
• SamFS hierarchical storage management
• Highly scalable filesystems such as QFS and pNFS.

It is a slightly updated version of my talk from OpenSource Days. I don't have to hurry as much and have added a couple of additional slides.

Back at ITU for another day of conferencing. This was more of a community day than Friday. I had big plans for going to presentations over the day, but ended up spending most of my time catching up with people and talking about OpenSolaris at the Danish OpenSolaris user groups booth. We had a really great turnout and probably handed out a couple of hundred opensolaris kits.

My own talk went better than I'd feared but not nearly as well as I'd hoped. I'm used to talking about technical stuff and trying to be a lot less serious was very different, but got a fair number of laughs.
I'd love another chance to try this talk being more prepared. Not sure whether I should submit it to ApacheCon EU or not.

The one talk I did manage to catch was Dan Kleins "Security As If Your Life Depended On It (because it might!)". Yet another great talk by him. Very entertaining but also a very serious issue. If you missed the talk, I strongly suggest checking back for slides and video later on.

Overall, Open Source Days 2008 was a great success and I really enjoyed it. I'm alreadt looking forward to next year. I'd better start thinking about presentation ideas.

Today was the first day of Open Source Days. The new venue is at the IT-University of Copenhagen. The place is not very old and it shows, very nice auditoriums with all the right gear to make life easier for the speaker.
My own talk OpenSolaris Storage Platform went really well except that I had to rush it a bit towards the end and I had to skip a few points I wanted to make. It was a good turnout and I've had some great feedback. Most positive of all, some of the reactions were genuine interest from people who saw a use for the software. I've already promised to do the presentation once more and have been talking to the Sun campus ambassadors at DTU and DIKU about that.
I got hit by a surprise Interview (in danish) as well. Not what I'd been expecting, but it turned out reasonably well (I hope my numbers aren't too far off).

I got there too late in the morning to catch much of Steve McKintys OHAC talk, but I think I got most of that talk yesterday. Steve certainly knew what he was talking about and I was very happy to see High Availability clustering get some airtime as opposed to the usual HPC clusters.
I also managed to catch Dave Johnsons talk about Roller and Jens Nyrup along with Jorgen Olsen both from Sun presenting xVM server with a broken demo.

All in all a great day even if it was a bit rushed for me.
I'm looking forward to Saturday (even if I'm having very serious doubts about the talk I'm giving. It is going to be very, very different from any talks I've ever given before. I could use another month or two to rewrite it and practice a tiny bit).

Steve McKinty, Sun Microsystems will be in Denmark, because he is speaking at the Open Source Days 2008 conference and has offered to give an extra talk the day before the conference on Open High Availability Clustering. Steve has led The Solaris Clustering Geographic Edition team since 2005.

Open High Availability Cluster (OHAC) is the open-source code base of Solaris Cluster1, a high availability (HA) clustering solution from Sun Microsystems. The main difference between Open HA Cluster and Solaris Cluster is that Open HA Cluster doesn't provide an end-user product or complete distribution. Instead, it is an open source code base, along with build tools necessary to develop, build, and use the code.

The talk will take place Thursday October 2nd at the IT-University, Rued Langgaardsvej 7, 2300 Copenhagen S, in room 2A12 at 16.00

Please RSVP.

Abstract

The term cluster is often associated today with High Performance Computing (HPC), but it also has a key role to play in the area of Business Continuity and High Availability. There is a lot of commercial interest in developing innovative solutions to these problems.

This talk will start with a brief description of the fundamentals of Business Continuity and how a High Availability framework such as Solaris Cluster can be used in that area, when combined with data replication technologies.

It will describe in more detail some of the work being done through the Open HA Cluster (OHAC) and related communities, and then will look at the technical details and progress of some projects that have been recently started by community contributors.

Lastly I'll discuss other opportunities for joint work and contributions with the OHAC members.

There will be time for questions on any aspects of Open HA Cluster and Business Community.

Friday

• 10:30-11:20   Steve McKinty Disaster Recovery with Sun and Open High-Availability Cluster.
• 12:30-13:20   Mads Toftum OpenSolaris Storage Platform - Storage revolution in the making
• 13:30-14:20   David Johnson Once and future Roller
• 14:30-15:20   Jens Nyrup-Jensen Sun xVM Server

I get enough exposure to Sun Cluster at work, that any other little tips and tricks are well worth the time and I'm looking forward to Steve McKintys talk. My own talk will be pretty hard to skip, so that's obvious. Dave Johnsons roller talk is something I've been wanting to hear at ApacheCon, but things didn't work out so far. It's always good to keep up with what goes on in other ASF projects and seeing what a fellow member has to say. Last slot goes to hearing about xVM server, which is something I keep wanting to have a closer look at, but just never getting that far.

Saturday

• 09:15-10:40   Jan Wieck Slony-I A master to multiple slaves-replication
• 10:45-11:45   Magnus Hagander Hidden gems of PostgreSQL
• 13:00-13:55   Mads Toftum Developers guide to grumpy old Sysadmins
• 14:00-14:55   Andreas Ericsson Building successful enterprise network monitoring with Nagios
• 18:30-??:??   Daniel V. Klein Security As If Your Life Depended On It (because it might!)

Another round of good Postgres talks to start off the day, and given that we use nagios for monitoring.apache.org that's a given as well. My own Saturday talk will be rather different from my usual talks, so I'm looking forward to seeing how that works out. Dan Klein is always interesting and going by the description of this years keynote, we're in for another treat.
This year we've also got an Open source project booth for the Danish OpenSolaris user group where I'll probably be spending the rest of Saturday.