Ping

Ping is a fundamental network tool that checks whether a remote host is reachable and measures how long a packet takes to travel there and back. The name echoes sonar — the command "calls" a host and waits for its echo. It is usually the first step in diagnosing connection problems.

How ping works

Ping uses the ICMP (Internet Control Message Protocol). It sends the target an echo request packet, and the host, if it responds, returns an echo reply. By measuring the time difference between sending and receiving, the tool calculates latency (RTT — round-trip time) in milliseconds. The target can be given as an IP address or a domain name, which ping first resolves via DNS.

A typical result shows the number of packets sent, received and lost, plus timing statistics: minimum, average and maximum. A high percentage of lost packets points to an unstable link, while rising, jittery latency indicates congestion along the path. Each reply also shows a TTL value, from which you can estimate the number of hops.

Practical application

Administrators use ping to quickly confirm that a server is alive, whether there is packet loss and what the latency to a given location is. It is also the first test when diagnosing network trouble — if ping fails, the usual next step is tracert, which shows on which leg of the route the connection breaks.

Keep in mind that some servers and firewalls deliberately block ICMP, so a missing ping reply does not always mean a service failure. For continuous availability supervision, uptime monitoring is a better fit, since it tests not just ping but the application's actual response.

Ping is also handy for simple comparisons — for example checking which of several servers responds fastest from a given location, helping you pick the geographically closest infrastructure. Remember, though, that a ping result describes only network latency; it says nothing about the load on the application itself or how long the server takes to build a page.