Management and verification of timing quality is essential for enterprise time synchronization and, like many other critical properties of a complete solution, it is not simple to put together from parts. If you’ve got a large time critical deployment, there are few tools to help you quantify how well (or how badly) client time tracks reference time. Network samples at various locations provide some data but not nearly enough. For mission critical computing, IT staff and business units need to know how well applications are following reference time sources, on every network, at every site, across the enterprise.
FSMLabs’ TimeKeeper® Administration is designed as an enterprise level solution to time synchronization management. TK Administration builds on top of TimeKeeper’s 6.0 release to provide capabilities such as:
- Visualize sync quality of TimeKeeper servers, and TimeKeeper clients, from the same interface - live
- See live network overhead information between clients and their server(s) over multiple protocols and network paths
- At a glance, identify the best and the worst timing client time sync quality regardless of protocol, to quickly identify network issues and to validate performance
- Track PTP boundary clocks such Arista switches that are running TimeKeeper, and all their clients, so you can roll up the entire enterprise into one single interface
- Configure TimeKeeper instances, from alarm thresholds to timing security and cross check features to boundary clock configurations
Do you need to make sure that alarms go off if any clients have a sync quality that deviates by more than 10 microseconds, or even 1 microsecond? TimeKeeper Administration can do this for you, from one interface. Need to see which network segments are having erratic round trip times as it happens? That’s not a problem - you can compare live network delay data from specific networks as it happens. Erratic time sources can be visualized and compared to trustworthy inputs.
Here’s what TimeKeeper looks like when displaying a series of clock sources being tracked, whether on the client or server (click for a full size version):
And here’s a quick demonstration of network delay/PDV visualization, which happens live for the identified hosts/subnets:
Looking at client sync quality, either live or for some longer time window, is also simple: