We have been aware of the slowness that people have been experiencing when attempting to access their accounts. We actively monitor the load and performance of our servers and are notified once they have reached a threshold value. At that point, we attempt to identify the cause of the performance issue and take corrective action. We have actively been working on several servers recently that have been experiencing high load, which is the main cause of performance issues.
This load can come from 4 main sources: clients' installing and running programs which overutilize resources (CPU, RAM, Disk I/O), an large influx of traffic to multiple accounts on the server resulting in simultaneous requests for resources, denial of service attacks against accounts, and maintenance programs used by our administrators, mainly the backup process.
We are constantly working to correct the high load situation caused by the above 4 items, however there are times when we are unable to take corrective action. Such an instance is the impact of server backups on performance. There is a trade off between trying to provide the performance expected of the server and reducing the impact of destroyed or missing data.
As a
VPS is still a shared resource product, other clients on a server will impact the performance of your account. We work to bring the problematic accounts back under control so that their impact on other accounts are reduced. This can involve a degradation in the performance of those accounts. This is one of the few instances that we will take an active role in modifying configuration files and restarting/terminating processes within a
VPS. The other is if an account has been compromised or is being DOSed.
One performance issue which is outside of our control is connectivity between your computer and the server. Although ping time can indicate potential problems with either server responsiveness or connectivity path issues, the best method of identifying the cause of the long ping times is via a traceroute (tracert on window servers). Traceroutes show the path that your request takes to get to your server as well as the length of time it takes to get a response from the node in the path.
Most people assume that high values for the ping time indicates that there is a problem with the server. This may be the case. However, ping really only provides you with 1 solid piece of information...whether there is connectivity to your server. All of the other information provided, dropped packets and reponse time, can be used to infer whether there is a connectivity issue.
Traceroute on the other hand provides the route which your requests to the server are travelling, if there are any dropped requests, and the response time. Because you are able to obtain the information from every point in the path, you can identify if dropped packets or long response times are due to the server (last entry in the route) or by one of the nodes in the route.
When experiencing performance or connectivity issues, the first tool you should use is traceroute as it will help in isolating the problem, for both you and support.
Server Speed (SLOW?????)
Linear Mode
