When there is a steady increase in demand and it seems that conventional ways will not be able to meet the TTM then realize that time has come to change the game.... Time has come to have a new approach......
Recent approaches in Mobility networks
2G, 3G, 4G, LTE, A-LTE, 5G Ready, WIMAX and what not. As we go ahead in this journey of telecom we see that almost every year a new revision of 3GPP is coming in increasing the last mile throughput standards. India being a market of around 1 Billion users is not untouched by such developments and with the developments in the world India also wants to keep pace with the same. But the real challenges are that how do you constantly expand and evolve networks that are mammoth in size.
A typical tier 1 network of an Indian Telco consists of 50,000 + network elements of different kinds. While the Mobility technology changes there will be changes of course in the last mile RF and in the Packet core but a substantial change also occurs for the expansion of the entire transmission network.
Transmission networks are to be changed with respect to Technology ( from TDM to packet), Platform ( from traditional tunneled architecture to a full IP base), Capacity ( Penetration of nX10G or 100G networks in the access), Flavour ( Putting access DWDM for future growth) and Topology ( from Ring to a self restoring mesh). What I understand that replacing the last mile RF equipments is relatively easy as they are the last mile points however evolving the transmission network is a bit challenging.
Indian operators have wonderfully handled this challenge with respect to changing networks from TDM to packet and now subsequently changing to self healing ip mesh but the major aspect which is the capacity is the thorn in the rose. Increasing capacity needs CAPEX and Time both of which are crititical to the service margins and time to market. So now let us look at the two approaches that people take to handle this.
1. Make the network FAT... (Conventional Grey Hair Planning)
I have introduced this term of "Grey Hair" Planning because in India it is becomes very difficult to convince "Experienced" persons to do something new. So this kind of planning for the future is the planning that the experienced people do and will keep on doing unless the owners realize that they have been totally gone kaput.
Nothing intelligent in this. The concept is still the same that there is one particular or few particular core locations of GGSN, Billing, PCRF and all of the content has to come from the core passing these gateways. Of course now the amount of content is much higher and more in volume. So to meet these high demands they need to make the network FAT. 10G networks to 100G and then introduction of DWDM in the access. More expenses... More complications and more people to look after.
Now while this is really impressive and expensive we need to understand the major problems in this kind of growth. I am not saying that expansion in this fashion is not necessary. Yes it is. However, only thinking in this manner is not the key. This form of expansion has its challenges of time and especially in India where the network sizes are large the planning needs to be really careful. Most of the CAPEX is reserved in this but the experience for the users hardly increases.
2. Futuristic way of Planning
Most of the "Experienced" planners may not agree to this method so my humble request to them is please do not read this section and waste your time.
Most of the time we assume that critical services are going to be always in the core and not realizing that as and when there will be more demand for services they have to be closer to the users. The key to this is Virtualization and then localization in the edge. In this case expansion can take a back seat for some time and the operator can concentrate to distrubute services like Billing, PCRF and CDN in the edge so that content is deliverted from the Edge itself.
In this approach we introduce the concept of virtualization but making VNFs (Virtual network functions) at critical edge locations. This makes the PCRF and BSS more distributed. In this structure the content is now localized at the edge. The network is not planned to catter Million subscribers but is now distributed to cater Several sets of 1000 customers in actual sense.
Creating this infrastructure will take much less time and less capex than creating an expanded infrastructure.
But I know, my dear friends, this is unconventional. It will take more time to convince the grey hairs to start the work for this. However, time will come when they will give way. So have
patience.
Normally it takes more time to convince the "Experienced" people for this technology than to actually lay down this technology.....
Cheers,
Kalyan
Recent approaches in Mobility networks
2G, 3G, 4G, LTE, A-LTE, 5G Ready, WIMAX and what not. As we go ahead in this journey of telecom we see that almost every year a new revision of 3GPP is coming in increasing the last mile throughput standards. India being a market of around 1 Billion users is not untouched by such developments and with the developments in the world India also wants to keep pace with the same. But the real challenges are that how do you constantly expand and evolve networks that are mammoth in size.
A typical tier 1 network of an Indian Telco consists of 50,000 + network elements of different kinds. While the Mobility technology changes there will be changes of course in the last mile RF and in the Packet core but a substantial change also occurs for the expansion of the entire transmission network.
Transmission networks are to be changed with respect to Technology ( from TDM to packet), Platform ( from traditional tunneled architecture to a full IP base), Capacity ( Penetration of nX10G or 100G networks in the access), Flavour ( Putting access DWDM for future growth) and Topology ( from Ring to a self restoring mesh). What I understand that replacing the last mile RF equipments is relatively easy as they are the last mile points however evolving the transmission network is a bit challenging.
Indian operators have wonderfully handled this challenge with respect to changing networks from TDM to packet and now subsequently changing to self healing ip mesh but the major aspect which is the capacity is the thorn in the rose. Increasing capacity needs CAPEX and Time both of which are crititical to the service margins and time to market. So now let us look at the two approaches that people take to handle this.
1. Make the network FAT... (Conventional Grey Hair Planning)
I have introduced this term of "Grey Hair" Planning because in India it is becomes very difficult to convince "Experienced" persons to do something new. So this kind of planning for the future is the planning that the experienced people do and will keep on doing unless the owners realize that they have been totally gone kaput.
Nothing intelligent in this. The concept is still the same that there is one particular or few particular core locations of GGSN, Billing, PCRF and all of the content has to come from the core passing these gateways. Of course now the amount of content is much higher and more in volume. So to meet these high demands they need to make the network FAT. 10G networks to 100G and then introduction of DWDM in the access. More expenses... More complications and more people to look after.
 |
The conventional way of network expansion
|
2. Futuristic way of Planning
Most of the "Experienced" planners may not agree to this method so my humble request to them is please do not read this section and waste your time.
Most of the time we assume that critical services are going to be always in the core and not realizing that as and when there will be more demand for services they have to be closer to the users. The key to this is Virtualization and then localization in the edge. In this case expansion can take a back seat for some time and the operator can concentrate to distrubute services like Billing, PCRF and CDN in the edge so that content is deliverted from the Edge itself.
 |
| Virtualized network |
Creating this infrastructure will take much less time and less capex than creating an expanded infrastructure.
But I know, my dear friends, this is unconventional. It will take more time to convince the grey hairs to start the work for this. However, time will come when they will give way. So have
patience.
Normally it takes more time to convince the "Experienced" people for this technology than to actually lay down this technology.....
Cheers,
Kalyan
Thanks for your love for grey hairs....already VNF based Core networks are implemented globally but they are centralized appropriately...pushing those functions to edge will take few more years...
ReplyDeletetransport will be with multi layer protection
The key technology here is MEC (Mobile Edge Computing). This is a standardized technology and has all the 3GPP compliances in place with vCDN. Ideal for India. However this requires proper synergy between RF, CSS, Packet Core and Transmission teams. Looks difficult to attain in India. (Territorial Dominance )
ReplyDelete