Profitap IOTA Use Case: Quickly isolating cloud-based application issues
Users at a remote office experience poor application performance with a cloud-hosted application.
The IT organization feels the server is under resourced. The provider says the problem is the client network. Neither side has proof.
The server ping roundtrip time seems to be ok, at least when engineers run occasional tests from the central office. However, this test only validates the network path between the client network and the cloud environment. They needed packet-level detail of the problem while it was happening. This was difficult to get because the problem was not always happening while engineers were onsite. They needed a way to simply and persistently capture traffic from the client side so the problem could be caught in the act.
The application was recently migrated to the cloud, so the network engineering team no longer had access to capture on the server side.
Once the problem was properly captured during a problem period, statistics like the network roundtrip time, server response time, TCP retransmission frequency, and other TCP outliers can be measured to isolate the true problem domain – whether client, network, or cloud-server
IOTA Made Things Easy
IT engineers were able to install the IOTA at the remote site by putting it inline between the client network and the edge router. This vantage point allowed them to see the activity of several clients, not just one. They could contrast client activity between problem periods and times with good performance.
After a few hours, clients reported that they again experienced the performance problem. Engineers were able to immediately access the IOTA using the web-based interface from the central office and begin troubleshooting. Within minutes they had access to the core details needed to isolate the problem domain.