It will be decades before terrestrial networks will be able to provide adequate and equal bandwidth across the globe. For now, pockets of highly connected regions, such as Europe and the United States, will dominate in global telecoms.
Until then, geostationary satellites will provide communication to rural regions with little/no terrestrial infrastructure. However, what do you do if you need to provide large files to a large number of users with little or no infrastructure without incurring huge costs? A cache server is the answer.
A cache is where your computer would store regularly accessed files for a limited amount until a certain requirement is met. A computer has many caches for different purposes. An example is web browsers storing frequently accessed web pages. By storing a file close to the end-user, the access time is negligible.
Compared to the data transfer speeds of the memory and motherboards, your internet connection is the slowest by a factor of 10 or 100 times.
On a local level, your computer will store a record of all the websites that you have visited and download them. This is known as a cache.
Visiting a site for the first time, your browser will download the website data, content and the time-date stamp to the cache. When revisiting a site, the browser will check the time-date stamp in the cache and compare it with the time-stamp on the website. If the timestamp in the cache exceeds the host time-stamp, it will download the website data from the cache. If the host time-date stamp exceeds the cache time-date stamp, then the browser will download the website data from the host and replace the cache version.
This is important for all users and the network provider as it reduces bandwidth usage, server load and perceived delay. Without caching, the internet would be costly, slow and wouldn’t be as commonplace as it is today.
Before Web2.0 and content-driven websites, the website data would, most likely, download straight from the host server. This was not too much of a problem with a static website due to its vastly smaller size. But with content-based websites, this can cause problems.
As the internet has grown, it has become a large hierarchy of many cache servers to help ease the network load caused by content-based websites. Cache servers are usually operated by network providers and can operate on a national, regional and citywide basis, storing many of the most popular websites and files for their users.
For example, a user looks up a news site. The computer will look within the cache to see if it has a copy of the website with a time-date stamp that has exceeded the host’s time-date stamp. If not, it will go to the nearest cache server to their location - found by looking at their IP address. It will look at the time-date stamp on that cache server and make sure it has not expired. If it has, it will look towards the next nearest cache server, looking for a copy that it has not expired. It will continue looking at cache doing this until a copy has been found or, after a number of failed attempts, is redirected to the host server. Once a copy is found, it will check to make sure the content is still fresh. If not it will, download a copy from the host server. This is then copied through all the caches it passed until it reaches the original user.
The same method is applied to large content items, such as videos. Watching Netflix programmes, it is likely that you are actually watching a copy based on a local cache rather than from the Netflix's host servers.
In areas not served by terrestrial internet, installing a cache server is not only important, it is necessary. Without one, it would be too expensive to provide internet to that area.
Despite the price per Mb per month falling, satellite internet currently is the most expensive form of internet connection. However, in places that lack terrestrial internet, it is usually the only option for getting online. This is set to change in the coming years.
For companies and network providers, installing a content cache server and creating a content distribution network (CDN) may be beneficial for VSAT users. With the ability to create content cache servers at strategic places, you are able to send large files without bandwidth being overrun. This is because you will only send one copy of the file to that cache server. On a satellite connection, this can save a lot of money and bandwidth, especially if the connection is a gateway for many users.
When a user requests to download the file, the website will redirect the user, based upon IP address location, to the nearest cache server. This not only saves bandwidth and money but results in a better experience for the user. Prices for cache servers have dropped dramatically over the past five years, primarily due to fall in hardware and memory prices - making ROI much more achievable.
A company produces an operating system that is used by millions of people around the world. Every month, the company has to issue an update to its users to ensure their operating system remains secure and efficient.
The updates are not small, approximately 1000Mb in size. Over terrestrial, this is no problem as network providers will recognise this update will likely be downloaded by lots of users within a few days, so will store it in their caches.
But for VSAT users, it is trickier. It would cost more money to download to transfer 1000Mb over a satellite internet connection, compared to terrestrial internet. If your VSAT connection is the gateway for many users, downloading this tens/hundreds/thousands of times could be a costly endeavour.
So for companies/NGOs who have VSAT users, installing a cache server at the satellite receiver will mean that only one copy of the update will have to travel the satellite connection. Then, the cache server will store it and duplicate it to anyone on the local level wishing to download the update.
By utilising this method, it has reduced the cost of downloading the update by a large factor.
