VM Monitoring with SevOne - We Do It and You Can Too
Virtualized infrastructures and cloud deployments have become part of our everyday work lives. Now more than ever as IT managers, we are faced with decisions about operational efficiency and cost savings in our data centers as well as balancing the best service with the most amount of transparency to users. One out of every three conversations we have with our customers is about tools consolidation. Over the years, departments within different organizations have purchased products to serve specific needs and have accumulated an unwieldy amount of point solutions that are not integrated together. Further, teams don't communicate and are not able to look at the big picture of network, server, and applications together to make decisions.
Our service provider customers (and a few enterprise customers too) have turned to us to accomplish their cloud initiatives with our VMware Monitoring solution. Having the ability to look at the network layer in combination with the server health and its VM guests, all from the same product, has saved them deployment time and money. In order to share these tools consolidation and operations efficiency improvements with our other customers, we launched a new customer training series. Our first session was earlier this month and it focused specifically on this topic – VM Monitoring. We had over 50 customers participate which resulted in a very active Q&A session following the live presentation. A link to the webinar recording can be found by clicking the image below.The success of our first training session shows the high level of interest in this topic and it shows the interest in tools consolidation and having one performance platform for all monitoring needs. And consolidation works well for us - SevOne's internal infrastructure is almost entirely virtualized, and we use the SevOne PAS to monitor everything. Our developers have several VMs each that they develop code on and our Quality Assurance department has access to all versions of our software on VMs. The advantage of all-in-one monitoring and troubleshooting is that whenever we have questions about slowness on our network, we can click on the NetFlow button on a report and get an instant report on which VM is the bandwidth hog. Recently we had a problem where our Sales Engineers complained that one of our demo servers was slow. We quickly pulled a network utilization graph and drilled down into a NetFlow report and saw that a development VM running on the Amazon cloud was the top talker. However, doing a deeper dive with some of the other NetFlow reports, we were able to see that even though this was the largest percentage of our traffic, it was still very minimal. We called our service provider armed with this information and were able to troubleshoot that they had misconfigured our service to be 5 times less than the contracted rate.
This also shows that it's no longer enough to only look at the internal infrastructure since so many of our day-to-day applications are SaaS delivered. We use SalesForce, Google Mail, Rally and other SaaS solutions and it's often very handy to look at their response times at the same breadth and with the same performance application that we use to analyze our in-house network and services.