EDB Postgres Advanced Server Review

As a replacement for Oracle RAC where there's need:

• More than four servers
• Dynamically deployable servers
• Split access between the transnational and reporting on the same data store.

• Major reduction in run cost, and a quantum leap in the flexibility of deployment models
• Elimination of replication of transaction from reporting systems
• Able to do rolling upgrades with zero production downtime.

• Variant server configuration: not requiring the same release on all.
• Effectively allowing us to be far more flexible from architecture to production, and the overall support model improves drastically.

• Lighter integration model for containerized deployment.

Effectively, there still remain some issues with a containerized deployment, so we have opted for a non-containerized deployment. This makes pack distribution a little more time-consuming.

More than five years.

This is an extremely robust, enterprise-grade solution that inherits all of the same reliability aspects of Ingress.

This is a very scalable environment both linear, horizontally and vertically.

We are one of the largest corporations in the world so our exectation and support given was extremely high. Would this be so with smaller company's I am not sure.

Oracle RAC and Informix. The single reason for moving off Oracle was the hard limit of four servers in a RAC and the degradation of performance at that. Meaning Oracle really is not scalable across servers, but requires massive increase in individual server specs.

Oracle and Informix are more costly deployments and support models, from a pure licensing perspective.

This was a straightforward one-to-one activity, and both Oracle and Informix database were fully able to be completely comfortable within a day.

In-house team with advice from the vendor.

Recovery of investment was less than four months.

I believe this is a negotiation issue with any organization that is based on the size of footprint looked at. Hence our experience would not be applicable. But negotiate, negotiate, and negotiate.

Yes numerous options, include HANA, MongoDB, MySQL, and MS SQL. None of which were sufficiently scalable or at our size, cost effective.

There will be a 10% pure play performance hit against ORacle and about 12% against Informix but the deployment and cost model more than compensate for the additional hardware requirements.