1. Backhaul looks for candidates
As it was discussed in this article, 4G backhaul needs upgrades. The fourth generation and the increase of bandwidth requirements are the guilty ones. Telecom operators know that it is time to upgrade their mobile backhaul, in order to prepare it for the next generation services and their demands. Wireless incoming technologies, such as WiMAX and LTE (analysis here), are going to be too much for the present backhaul.
Without a proper backhaul optimization, neither your backbone nor your access can stay current. An old fashioned backhaul will be a gigantic bottleneck. Carriers are aware of this fact so they are managing some options to upgrade their Networks.
Which technologies could help carriers to deal with this problem? Lets see a little description of some possible candidates.
2. Candidate 1: Provider Transport Backbone (PBT)
Provider Transport Backbone, also known as PBB-TE (Traffic Engineered Provider Backbone Bridging), is a standard of the IEEE and the ITU-T (802.1 Ethernet working group 802.1Qay).
The objective of PBB-TE is to provide carrier class features over Ethernet networks. The standard implements some of the features of SDH/MPLS into Ethernet. In few words, it can be defined as a “connection oriented packet network“.
General characteristics of SDH, MPLS, PBB-TE
(To read more and see download options: Click “Read More…” below this line)
Traditionally, Ethernet has been a low cost technology, very efficient in terms of cost per bit. It has a lot of support from vendors around the world, thus is a “cheap” technology when it is compared with others. Ethernet has very cheap interfaces, it is highly deployed worldwide, it is very scalable, and it can be defined as an “IP native” protocol, feature that is very appreciated nowadays due to the fact that the mobile environment is turning into an all-IP world.
However, when one talks about Ethernet as a transport technology for the backhaul, it is clear that it lacks of several features (management, security, OAM, QoS), which are irreplaceable for operators in this part of the network. For this reason, telecom operators have been using Ethernet over SDH or Ethernet over MPLS instead of using Ethernet alone.
Ethernet’s advantages and disadvantages for the backhaul
In this situation, PBB-TE appears to complete those missing features of Ethernet and to transform it into an optimal transport technology.
To achieve this goal, the new features added on PBB-TE are the following ones:
- PBB-TE forgets about the self-learning techniques of Ethernet when one talks about the forwarding tables (Good bye Spanning Tree!). There are no broadcast floods over the network. Forwarding tables are now managed by special software, thus the management of the network goes better by eliminating broadcasting packets flooding the network.
- It provides link protection with pre-calculated paths, which are stored on the switches. The change between the original path and the protection path is very fast: the protection path is already designed and the switch does not have to discover it on the fly.
- It is said that up to 65-70% of the total cost of a telecom operator is the OPEX. As we have said, Ethernet technology is cheaper than others and it has a good cost-control. If Ethernet is a cost efficient technology, and the OPEX represents a huge percent over the total cost, then Ethernet can save a lot of money in this point.
PBB-TE effects on the Network cost
- Use of OAM packets over the user data link. A good OAM support can bring optimal control over the network resources, decreasing the OPEX of the carrier (again!).
- It supports different services types with different requirements of bandwidth, latency, etc. This is a very important point, because the transportation network of the backhaul must meet some requirements in this matter: the backhaul will be composed by different mobile generations. The backhaul will transport different services with different requirements of quality, of latency, etc. In order to support multiple generations, with different types of technologies among them (TDM, ATM, Ethernet), the possibility to manage these parameters is critical.
- PBB-TE allows Virtual LANs (VLANs) in a connection-oriented scenario. This feature is present in SDH protocol.
With these new features, Ethernet technology presents itself as a good candidate for the backhaul. Will MPLS compete with PBB-TE?
3. Candidate 2: Transport MPLS (T-MPLS)
MPLS (Multi Protocol Label Switching) is a well-known standard among telecom operators around the World. It is a highly deployed technology, which has been studied and used by operators during the last 10 years. Therefore, MPLS is considered a good and flexible technology in some areas of the network.
Some years ago (2006-2007), the ITU-T organization approved the T-MPLS (Transport MPLS) standard as an adaptation of MPLS, in order to use T-MPLS in the transport network. In this new standard, some complex features (Layer 3 features) have been removed, because they don’t seem to be necessary in the transport network. T-MPLS focuses on the L2 transport.
TMPLS is between Layer 3 and Layer 2
Furthermore, new features have been added in order to convert T-MPLS into a good transport technology to the backhaul. Some of these new features come from technologies like SDH (Synchronous Digital Hierarchy). The main difference between MPLS and T-MPLS is its real behavior: MPLS and T-MPLS both are packet switched technologies but MPLS-T acts like a circuit switched technology in some aspects.
The objective of T-MPLS is to mix the advantages of MPLS with some new features: end to end easy management, improved OAM, different client layers, end to end protection, good scalability and low costs.
The increasing requirements of the next generation backhaul (this article talks about them) have forced operators to find a proper technology to upgrade their networks. The use of IP and MPLS together can reach almost every requirement, but it seems to have problems with a very important one: the cost control à scalability with low costs.
4G Backhaul requirements
Here is where T-MPLS arrives. It brings the advantages of MPLS, it adds some new features to support circuit switched legacy, it is simpler (thus it is more cost effective) than IP/MPLS, and it has good management and OAM controls.
The best points of T-MPLS:
- It is a 2,5 Layer protocol: It has better intelligence than L2 protocols but it is not as complex as L3 protocols (thus cheaper). It aims to the transport of packets, not to the route of those. In this way, T-MPLS doesn’t change the philosophy of the transport technologies.
- Connection oriented packet switched protocol, supporting circuit switched legacy.
- OPEX and CAPEX lower than these of MPLS.
- Similar management controls than today technologies (for example SDH): no need for new control panel models.
- Good OAM and QoS, a carrier grade protocol.
4. And now…what happens?
After summarizing the general characteristics of PBB-TE and T-MPLS, one can question: Which one is better? Are there other options for the backhaul?
None is better than the other. The situation of each carrier will decide the technology that best fit in its network. From a general point of view, PBB-TE and MPLS both have similar features (with differences, of course), but they are based in different technologies, Ethernet and MPLS. The telecom operators will choose the candidate that best fit with their preferred technology, according to the rest of their network.
Answering the second question: Yes, there are other options for the backhaul. In fact, there are a lot of other options, variants, new points of view of these technologies, and so on. There are people that prefer Ethernet or MPLS in their classical form, there are some new schemas like MMBI (see this)… Several options are possible, none is better than others, they just fit better in one network or in other…Telecom operators must choose their best case based on their own situation.
– Download Options:
– Share this:
Candidates for the Backhaul: PBB-TE vs T-MPLS