There seems to be a jinx which has cursed HPC:Factor that we have a year of stability and a year of total anarchy. After a year of stability in 2009, it's no surprise then that for our alternate yearly hex we've started 2010 with a fairly long outage.

Since March 2003, HPC:Factor has lived in sunny Florida under the watchful eye of John Rubier in a significantly subsidised hosting arrangement. At the beginning of the Autumn of last year we heard the unfortunate news that our data centre would be closing down around Q2 2010 and that we would have to move.

We were not expecting to have to tackle this problem until March time and I was planning on announcing the move, talking to the community and seeing what the consensus would be over what we should do next.

Unfortunately on Friday 22nd January a network reconfiguration at the data centre pulled our network connectivity and that was the end of that.

John Rubier has been burning the midnight oil, and has been able to get some administrative access up and running, however I didn't want him to waste time reconfiguring everything and hacking together a working version of the site in the full knowledge that we'll be doing it all again in a few weeks time. To that end I asked him to stop once we had administrative access to the servers and I spent 72 hours reconfiguring, rewriting and upgrading the site.

By the time you read this message, HPC:Factor will be in the UK on two servers and unfortunately a fairly uncertain future. Quite simply I don't have access to an OC line and there are limits on how far I can push this already overloaded line.

The changes that I quickly made to the back-end code and the interface on the site in September, although unpopular in some circles were engineered because of the expectation that something like this may happen. The rollout of HPC:Factor v2.5 was such that it eliminated well over 4.5 million server transactions in a monthly log cycle, saving bandwidth and CPU time.

The forced migration has also meant that we can move up to IIS 6, which supports fully dynamic GZip and Deflate compression and I have also migrated the database system from SQL Server 2005 to SQL Server 2008. The site is now ASP .net 3.5 enabled which provides a wider range of processing options over standard ASP 3.0.

We apologise for any inconvenience caused to anyone as a result of the outage and hope that you will bare with us while we and the site adjust to the changes.