Full Text Available
Note: Clicking the button above will open the full text document at the original institutional repository in a new window.
In the past, network clock synchronization has been sufficient for the needs of traditional distributed systems, for such purposes as maintaining Network File Systems, enabling Internet mail services and supporting other applications that require a degree of clock synchronization. Increasingly real...
| Main Author: | |
|---|---|
| Other Authors: | |
| Format: | Thesis |
| Language: | English |
| Published: |
Department of Electrical Engineering
2016
|
| Subjects: | |
| Tags: |
No Tags, Be the first to tag this record!
|
| Summary: | In the past, network clock synchronization has been sufficient for the needs of traditional distributed systems, for such purposes as maintaining Network File Systems, enabling Internet mail services and supporting other applications that require a degree of clock synchronization. Increasingly real time systems arc requiring high degrees of time synchronization. Where this is required, the common approach up until now has been to distribute the clock to each processor by means of hardware (e.g. GPS and cesium clocks) or to distribute time by means of an additional dedicated timing network. Whilst this has proved successful for real time systems, the use of present day high speed networks with definable quality of service from the protocol layers has lead to the possibility of using the existing data network to distribute the time. This thesis demonstrates that by using system integration and implementation of commercial off the shelf (COTS) products it is possible to distribute and coordinate the time of the computer time clocks to microsecond range. Thus providing close enough synchronization to support real time systems whilst avoiding the additional time, infrastructure and money needed to build and maintain a specialized timing network. |
|---|