Saturday, July 13, 2019

What is the best practise for monitoring latency, packet loss in a large enterprise system?

Hello, My workplace(a Govt. Org.) has a large IT network, with >100,000 Desktops, numerous routers/switches and about 10K local servers distributed across a large country (MPLS based).

While we do have a NoC with Solarwinds to monitor the entire network, it is manned by a vendor firm.

In the recent past (5-6 months), our firm has been noticing that customer complaints about network issues are rising, while the monitoring platform and the MIS provided atop the system are still painting a rosy picture.

Upon investigation, we realized that the NoC vendor had been acquired by one of our largest network providers last year.

This has understandably upset our management, since they weren't so much as intimated about the change. Nor were there public filings, as both firms are private, un-listed entities.

While management is separately undertaking an investigation into conflict of interest, and replacing the NOC vendor, we (an audit sub-group of the IT wing) have been asked to investigate deviations/mis-reporting by the vendor. We have a free hand to install software, monitor SNMP, etc., but not to touch solarwinds (A clause in the NOC contract allows them to inordinately delay providing us with data/access)

What would be the best way to go about monitoring the network, assuming that I can work with the end systems, but cannot touch the routers or solarwinds.

I apologise if this isnt the correct forum for such a question and request you to direct me to an appropriate place.



No comments:

Post a Comment