Scaling Data In The Cloud
One of the great things about cloud deployments is the easy scaling. You might run your application on multiple machines, changing the number of machines up and down perhaps through a PAAS engine marshalling an IAAS provider.
This is all fine for processing, but what about data stores?
If these are hosted on a single server, this quickly becomes a bottleneck that limits the ability for the processing to deliver on the scaling promise.
In this talk we’ll look at Hazelcast in-memory data grid (IMDG), and how this can distribute and re-distribute data records evenly across multiple hosts. Now data can scale as easily as processing. Problem solved!
Well, not quite. There are a few irritating details to be aware of, so we’ll end this talk with a demo of Hazelcast data-grid running on some providers, to see how the problems manifest and how they are solve.