Jan 23, 17 / Aqu 23, 01 17:43 UTC

Fastenning Fibre-Optical Communications  

The fibre-optical communication is the fastest communication technology of all the types of the wired communication technologies where it is based on the electromagnetic signals. However, the capacity of the fibre-optical communication is the least through the wired communication. A balance between the high speed and the low capacity of the fibre-optical communication has to be achieved.

The capacity of the fibre-optical communication has to be improved. The best way to increase the capacity is to increase the wireless band of the fibre-optical signals. The advantages of the fibre-optical communication is that no risk of intereference exists. In addition, the fibre-optical communication is more secured than the other types of the wired communications because there is no vulnerability for an intruder to snoop the communication.

The fastenning of fibre-optical communications will be mainly achieved by performing compression on the transmitted data. One of the methods to perform the compression is to encapsulate the packets of the same destination within the same frame so that the transmitted frames will be very long and therefore this process will be done only in the stage of the fibre-optical communication.

Jan 24, 17 / Aqu 24, 01 11:16 UTC


In addition, the fibre-optical communication is more secured than the other types of the wired communications because there is no vulnerability for an intruder to snoop the communication.

Evidence suggests otherwise.

What needs to really be improved in the overall network is "last mile delivery" - a lot is still (a lot further out than a mile, in a lot of cases) over copper. In terms of capacity then I think it's something like 43TB/s over a single fibre, meaning a serious upgrade required for all the existing exchanges, switches, hubs, routers etc. and will result in significantly more responsive networking. Before compression. Techniques like "presto" can speed up browsing etc but only if the remote service supports. For actually moving chunks about the network, most use the likes of caching proxy servers and CDN's to move content closer to the user and reduce in distance transfer.

Jan 24, 17 / Aqu 24, 01 18:09 UTC

The idea here is that no interference occurs in the fibre-optics communications so that no interleaving process is required. The interleaving process increases the transferred data amount. As no extra sizes will be added to the transferred data over fibres, it is a chance to decrease the data amount by compression. The applying of the proper compression techniques will be the rule of the routers and switches.

Jan 25, 17 / Aqu 25, 01 09:06 UTC

Compression would only provide advantages if we was near the limits of what's physically possible with the fibre. Or trying to get more profit out of the same equipment that's been running for ten years that paid back investment costs within the first month and half of operations.

Most of the issues with the global network is the lack of fibre connections. Once they are solved, which is only possible via physical replacement which the places with copper left are unwilling to do, because they can still earn revenue from the copper - then the main bottlenecks is the capacities to the exchanges themselves. If you can be throwing 42TB/s (I'm sure I read of 240+, but can't find the paper) down a single strand, to assume 288 strand backbone cables then the exchanges would need in the region of 12096TB/s or 11.2PB/s in order to keep up with a single cable... Compare this to the available exchange technologies, which is not much bettering 100Gbps. Further problems are generated by ISP's downstream when they split that into a set of 10x 10Gbps exchanges to serve a residential area, and split it into 350 or 450 30Mbps pipes - Mostly because there's only 233 available, and they're selling the same bandwidth twice to two or more people. Then there's the "shaping techniques" many utilise to try and make sure it fits when people just try to take what they've actually paid for.

Sure, we possibly would be wise to make more effective use with the technologies in place - but networks are kinda chains, and chains are only as strong as their weakest links... We need all them weak links improved, or no software countermeasure is likely to be effective. And you need to stop accepting them messing with your traffic, data is sacred. Sure, through my VPN all my comms are packed, but the largest "speed increase" is gained purely because I don't doctor my traffic, prefering to deploy another node as opposed to use b/w that isn't there and I exit via a completely different backbone to most interwebs users...