Thursday data center additional, Rackspace slows to a crawl.


Just reading about major performance impacting storage load issues on the Rackspace cloud in their DFW data center that have been dragging on (what is it with DFW anyways?).

This isn’t rocket science guys, find the host with the most traffic to the SAN, find the virtual on that host with the most traffic to the SAN, and throttle it or kill it. Repeat as required. This is a great example of what happens when you don’t build traffic monitoring into your cloud.

Vern

swiftwater telecom logo

Advertisements

2 responses to “Thursday data center additional, Rackspace slows to a crawl.

  1. I disagree on so many levels, I think the problems that Rackspace and Amazon are having are related to the inherent problem with the cloud. Regardless of your software or architecture when you allow several hundred (thousands?) of customers to share infrastructure you are going to have crossover problems. With the current technologies available there is no way to separate and meter 100% of the resources that customers use on the cloud and have it still be profitable.

    • Is it any less profitable than letting the whole thing go out of control and impact all of your paying customers plus whack your reputation? I don’t think so.

      Maintaining monitoring and control of the network usage of virtual machines is no more difficult or costly than monitoring a like number of physical machines. Running a cloud without it is inviting disaster.

      Vern

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s