When the blame game begins, it is time to pause and introspect. Your Cloud service may be delivering just what has been promised, but there may be other factors which may be reducing the quality of service actually received. The frustration that triggered off the blame game may be just what you need to inspire you to perform a thorough test on factors that impact your network, and do not facilitate the optimal delivery of Cloud services.
Here is a list of “factors” that you can check on before you tick off the Cloud service provider:
Are there latency and/or jitter in your network? It is a fact that all networks experience some amount of latency and jitter.
Latency can be simply defined as the additional time to respond introduced by system hardware / software / configurations. Latency is the measurement of the added time. The “time to respond” may be as small as a nanosecond or a millisecond, and as much as a few minutes. This delay may be introduced by a switch or a router or the Internet service.
Jitter is a variation in latency and is generally measured as a variation between multiple packets or streams. It is often expressed as an absolute value –| D(i) – D(i-1) | — where D is forwarding delay and i is the order in which frames were received.
At the application level, latency and jitter are translated into “response time”. Response time is the time required for an application to complete a transaction. This tends to be the most vital metric for enterprises that have a large number of applications in the Cloud, and huge volumes of data are uploaded and downloaded per transaction. End users are most aware of delays at this level, and most complaints about the quality of services obtained from Cloud vendors or applications, is received at this level of computing. Administrators will have to check on packet speeds and application speed to ascertain the nature of the problem before apportioning blame.
Site to site synchronization is measure of time delay between sending and receiving sites. Time delays are dependent upon the nature, direction, and volume of data flow.
However, a study of latency on the network alone will not tell the whole story. It is just one of the factors that may characterize device or network related delays. There may be other factors that cause delays and performance degradations. Network problems may be caused by:
- Highly fluctuating workloads that are not planned for
- Power lines running close to or in parallel with your network cables
- Crimped or bad cabling within a building
- Junk spitting from a bad network adaptor on the PC
- Poor router configuration
If none of these problems exist on your network, go ahead and take it up with your Cloud service provider.