First thoughts from liquidlabs

Liquidlabs have been around for over 2 years now and over that time we have been been solidly developing our distributed runtime platform to something we believe is truely next generational. The original agenda was to realise a self-healing AOP framework we developed about 5 years ago as a compute fabric; over time our realisation of our own distributed system and familiarity with the buzzwork compliant confusion of the moment (compute grids, data grids, cloud, OSGI - and now PAAS) pushed us to something that meets these agendas - and then beyond. While the goldrush/landgrab has been happening -(just have a look at the google groups 'cloud' list) - we are (quitely) crossing the finishing line.

Its certainly an interesting time where the small companies stand a chance against the giants like HP, IBM, Oracle...but...we have been focusing on what everyone is calling PAAS - platform as a service (VScape). There are quite a few good blogs on what it is and what it isnt - also the major overlap or layering that takes place over cloud. It seems PAAS is everyones comprehensible version of a cloud - not 'what is a cloud?'. Anyways enough of my rambling - our application Logscape executes over the VScape platform providing, adaptive healing, chargeback, autoscaling, ease of deployment (single click) - and a host of other goodies. Its currently deployed in a couple of banks and we test for scale using EC2.

The best thing about development for the platform is how easy it is - most distributed applications (SOA etc) take such a masssive amount of effort. Creating and maintaining multiple deployment environments (Dev, UAT, QA etc) is no easy task - and then when something goes wrong its 'all hands on deck' to figure out the one of the message queue config files is being picked up from an old jar - or that one of your machines had a network outtage at 1:30AM.