Tag: cloud

Categories
timekeeper

Measuring NTPd in the Cloud II

 The problem with ntpd is the sudden performance degradations which can occur when conditions deviate from ‘ideal’. We have detailed these robustness issues of ntpd in prior work, including [17, 18]. Simply put, when path delay variability exceeds some threshold, which is a complex function of parameters, stability of the feedback control is lost, resulting in errors which can be large over small to very long periods. Recovery from such periods is also subject to long convergence times […]

Now to the results. As expected, and from the very
first migration, ntpd exhibits extremely large errors (from
-1 to 27 s!) for periods exceeding 15 minutes (see zoom
in middle plot) and needs at least another hour to converge
to a reasonable error level.

from  Virtualize Everything but Time, Broomhead ,Cremean,Ridoux,Veitch

TimeKeeper is a “from ground up” implementation of  clock synchronization over both PTP and NTP and does not share NTPd limits.

 

Categories
timekeeper

Managed services (and cloud) for financial industry

The nature of data security is constantly changing for the financial industry. With the ever-growing need for tighter IT infrastructure security and the increasing prevalence of BYOD-friendly workplaces, financial institutions are turning to managed services to help handle these changes. In fact, a TABB Group forecast predicts that by 2016, 50% of financial institutions will be opting for managed services to help account for their IT infrastructure management. Similarly, a survey by TSIA cites managed services revenue today as the fastest-growing in the service line (which includes professional services, support services, and field services). [Onramp]

Categories
timekeeper

Time Synchronization in the Cloud

Synchronizing clocks in the cloud, especially for virtual machines, is way beyond the capabilities of ordinary synchronization methods. Tests show that virtual machines in the cloud relying on NTPd can fall off the reference time by tens of minutes over a single day. Even bare metal cloud platforms are significantly worse than dedicated machines.  This creates an interesting dynamic because distributed applications are becoming more and more dependent on tight time synchronization and there are a large number of existing applications that need at least millisecond level synchronization.  In the cloud environment, the design limitations of the alternative time protocol, PTP, mean it cannot be relied upon to give provide better performance either. PTP was initially designed to servos and data acquisition devices on a single shared ethernet – weaknesses like dependence on multicast (broadcasting)  and top  down failover methods are a problem in the general enterprise, but really become impediments in the cloud.  That’s why we took a semantic approach, fixing time distribution above the level of the protocol. The low level, bit level, packet level, cannot deliver the sophisticated time analysis, fault recovery, and management needed for complex environments like the cloud.