With more companies using Software as a Service (SaaS) applications such as salesforce.com that may be deemed business critical, it would be great to dynamically failover to another Internet connection, at the same site or a different site, based on the performance of these Internet applications.
Monitoring Business Critical Internet (Cloud/SaaS) Applications
Think of them as providing IP SLA [monitoring metrics] on steroids specifically for business critical internet-enabled applications. They are doing some pretty neat things by leveraging public BGP route servers and can even spot a carrier who may have route flapping going on. Of course, they are also testing packet loss, latency, and all that good stuff too.
If you could configure ThousandEyes on an Internet Edge router or Firewall like you can IP SLA, it would be pretty cool and easy to inject a backup route if salesforce.com performance was degraded. But, you cannot. However, here is where things get really cool. As every new company talks about these days, ThousandEyes has fully open RESTful APIs to their platform. This means all of stats you would see on their dashboard can be pulled down into an application to consume them and act on them.
Enter onePK
For those wondering how ThousandEyes deploys their solution, it is deployed as a VM or Linux package at each site that is accessing the Internet. These are called Private Agents. For example, if there was a corporate site and just 1 branch site each with their own Internet connection, you would deploy 2 private agents. Each one costs $99/month which isn’t bad if you start small to fully test the benefit of the solution.
If any end users that are current customers of ThousandEyes want to explore onePK integration, please let me know. I’d be more than happy to get involved somehow. If you’re local in the NYC Metro area, even better.
Related links on ThousandEyes:
MONITOR PUBLIC SAAS PROVIDERS WITH THOUSANDEYES
The Vision Of A ThousandEyes
THOUSANDEYES: A BETTER CLOUD-BASED APPLICATION PERFORMANCE MONITORING SOLUTION
Thanks,
Jason
Twitter: @jedelman8