How We Did It: Cloud Databases And Storage

Filed in Product & Development by J.R. Arredondo | August 29, 2012 4:00 pm

One of the advantages of the container-based virtualization architecture[1] we use in Rackspace’s Cloud Databases[2] service is the improved IO performance for your database. This performance improvement is primarily due to two reasons: the elimination of the hypervisor as a “middleman” in the path that goes from the database engine to the storage; and the flexibility in IO priority allocation that Cloud Databases gives to each one of the different containers of the system.

This is really important for databases because IO can be a significant factor in performance, especially if your database performs lots of reads and writes, reads large amounts of data (SELECTs) or performs batch UPDATEs or DELETEs transactions.

In this short video we talk a little more about this:

Endnotes:
  1. container-based virtualization architecture: http://c1776742.cdn.cloudfiles.rackspacecloud.com/downloads/pdfs/Rackspace-Cloud-Databases-and-Container-based-Virtualization.pdf
  2. Rackspace’s Cloud Databases: http://www.rackspace.com/cloud/public/databases/

Source URL: http://www.rackspace.com/blog/how-we-did-it-cloud-databases-and-storage/