Post by account_disabled on Dec 20, 2023 6:31:57 GMT
Before delving into the specifics of the Google Cloud network, it is important to first understand these two concepts: bandwidth and latency. Bandwidth is the throughput capacity of the network, measured in Mbps; while latency is the delay or sum of all the delays that different routers add along the way to our web requests and responses. Figuratively speaking, bandwidth or flow rate can be represented as the capacity of the water pipe to allow a certain volume of water per second. Latency can be compared to the delay from the moment the water pipe is opened until it begins to flow out. Because of the short wait in establishing the connection between different routers, each “hop” along the path adds latency to the final requests and responses.
So, the further apart the visitor and the server hosting the Country Email List site are, the greater the latency. Additionally, the more fragmented the network, the higher the latency. We can imagine this using a tool called traceroute , or tracert on windows. We used it to inspect the routing delays of two requests, made from Europe, whose data is displayed in the screenshots below. In particular: one on weibo.com: Weibo.com Weibo.com and another on bbc.co.uk: BBC.co.uk BBC.co.uk As we expected, the number of hops to the website in China is almost 2 times higher than in Europe. It is added latency compared to a request to a website hosted in the UK. The three columns shown by tracert represent three roundtrips (RTTs).
Each row represents different routers or hops along the route. Often these tools provide URLs that help determine where that specific router is located. The round-trip time for routers in China/Hong Kong takes almost a third of a second. We used Pingdom's tools to load a website hosted in London from Pingdom's Australian location, to try and establish how much of a responsibility the network has for the overall load times of a website. Example of loading times Example of loading times This is the data for a small CSS file loaded in this test scenario. The Connect section plays the largest part in loading this resource, followed by SSL and Wait. All the time until, including Wait time, together, is also known as time to first byte ( TTFB ), which includes network latency.
So, the further apart the visitor and the server hosting the Country Email List site are, the greater the latency. Additionally, the more fragmented the network, the higher the latency. We can imagine this using a tool called traceroute , or tracert on windows. We used it to inspect the routing delays of two requests, made from Europe, whose data is displayed in the screenshots below. In particular: one on weibo.com: Weibo.com Weibo.com and another on bbc.co.uk: BBC.co.uk BBC.co.uk As we expected, the number of hops to the website in China is almost 2 times higher than in Europe. It is added latency compared to a request to a website hosted in the UK. The three columns shown by tracert represent three roundtrips (RTTs).
Each row represents different routers or hops along the route. Often these tools provide URLs that help determine where that specific router is located. The round-trip time for routers in China/Hong Kong takes almost a third of a second. We used Pingdom's tools to load a website hosted in London from Pingdom's Australian location, to try and establish how much of a responsibility the network has for the overall load times of a website. Example of loading times Example of loading times This is the data for a small CSS file loaded in this test scenario. The Connect section plays the largest part in loading this resource, followed by SSL and Wait. All the time until, including Wait time, together, is also known as time to first byte ( TTFB ), which includes network latency.