25 October 2011
Background article
SURFnet has always been a driver of network innovation, inspired by the needs of its clients: the higher education and research sector. Mobile devices are playing an increasingly important role in that sector. Unrestricted, fast, uninterrupted access to the Internet is hugely important in new approaches to working and learning. By
entering into alliances with organisations such as KPN – the leading telecommunications and ICT service provider in the Netherlands – SURFnet can continue to drive innovation.
Consumerization
The fact that people want to work on their mobile devices anytime and anywhere – a phenomenon known as consumerization – is something of a headache for knowledge institutions. They must not only choose between the many new options now available but must also consider the management and other costs involved. Students, staff and visitors generally have three separate devices on them for which they require continuous online access. Each device serves a different purpose: one is good for watching videos, the other for writing or editing documents, and the third for making phone calls and keeping up with friends and family. All of these devices require a huge amount of bandwidth. Knowledge institutions must therefore increasingly offer wireless, high-bandwidth connections, and take end-user mobility into account. The demand varies: week X might see a large group of people working together at location Y, but that picture could be entirely different in month Z. Network managers face challenges both inside and outside of their institution’s buildings. Members of SURFnet’s target group want trouble-free connectivity while commuting between their homes and the campus or moving between campus buildings, when conducting research out in the field, and even while they are abroad. After all, research and innovation depend on being in contact with peers and on having continuous access to reference works, images, databases, and one’s own documents.
Bottleneck
For almost 25 years, SURFnet has been a leader in building network infrastructure aimed at improving the quality of Dutch education and research. Its main priorities are to remove restrictions in bandwidth and network access. In the past, network bottlenecks were found in two places: at the core of the network and in the access layer. Rolling out a fibre-optic network has eliminated virtually all of the problems in the core, and the introduction of eduroam has eased access to the wireless network and removed a bottleneck in the access layer. Consumerization has increased the demand for mobile bandwidth, however, creating yet another bottleneck in the mobile access layer. WiFi networks inside buildings are becoming overloaded and there is a constant scramble to increase and relocate capacity. Knowledge institutions are doing their best to solve these problems, but progress has been limited owing to a shortage of qualified staff and (thanks to budget cuts) the restricted funds available to contract the necessary external expertise. Outside the buildings, coverage via the mobile networks (3G/UMTS) is often very patchy, with no broadband access at all. In addition, telecom providers have thrown up barriers to network utilisation, especially abroad. Users must keep a sharp eye on their connectivity or pay an enormous bill (for data traffic) at the end of the month. All these factors combined are a huge impediment to innovation in education and research. Mobility is wonderful, but in terms of connectivity and reliability it represents a huge step backwards.
Driving innovation
SURFnet believes that progress is too slow and is also moving in the wrong direction. It has therefore decided to break through the current status quo in mobile communication and to focus on driving progress and innovation in that area. It cannot do this alone, however; it has neither the necessary resources nor the ambition to roll out its own mobile network in the Netherlands or on campuses. There are enough mobile telecom providers in the Netherlands to ensure competitiveness, and the institutions' campus networks are and will remain their own responsibility. However, by working closely and joining forces with educational and research institutions, SURFnet and mobile telecom providers can together guarantee a mobile network that offers unrestricted broadband access to the Internet.
IPv6
Another bottleneck in Internet innovation is the impending exhaustion of IPv4 addresses. The Internet will consequently be making the transition to the new Internet protocol, IPv6, in the next few years. SURFnet has long encouraged knowledge institutions to adopt IPv6, and it is itself a member of the IPv6 Taskforce along with many mobile and other telecom providers. In terms of ensuring Internet innovation and growth in the near future, the introduction of IPv6 is just as important as developing the mobile network. Without IPv6, users will be unable to access new Internet parties from the existing networks. The network-wide introduction of IPv6 is a labour-intensive process that requires an accurate approach. ICT departments must not simply offer users a “quick fix”; that would place network transparency and reliability at risk, precisely what SURFnet wishes to avoid. An unreliable, non-transparent Internet will impede innovation in education and research.
Systematic introduction of IPv6
The introduction of IPv6 will need to take place over several years. The small number of unallocated IPv4 addresses is insufficient to cover the period prior to the introduction of IPv6, at least if that introduction is to take place effectively and according to schedule. Roughly speaking, there are two options: to make a rapid but less effective transition to IPv6, or to use NAT technology so that fewer public IPv4 addresses are needed. Both options are unsatisfactory, however: they would not provide the necessary network transparency, are not future-proof, and would lead to a disproportionate rise in the network management burden. Without extra IPv4 addresses, it will ultimately come down to choosing between the lesser of two evils.
In the early days of the Internet – before the Regional Internet Registry was introduced – SURFnet was assigned a large number of IPv4 addresses. The knowledge institutions connected to SURFnet are making use of these addresses, and there are still enough IPv4 addresses to supply the target group for the foreseeable future. SURFnet now aims to join forces with KPN by using some of its unassigned IPv4 addresses to ensure the effective introduction of IPv6, setting an example for the entire Internet community. That way SURFnet can help that community, as well as the education and research community, to move into the future.
Research alliance between SURFnet and KPN
At a meeting earlier this year, SURFnet and KPN found that they had a number of interests in common in the areas described above. That has resulted in a research alliance between SURFnet and KPN in which both parties hope to generate knowledge and gain experience working with the new “fourth generation” (4G) mobile networks, as well as give an extra boost to the introduction of IPv6 in the Netherlands. All this will serve the higher purpose of promoting new ways of working and learning in higher education and research.
Integrating WiFi and mobile networks, based on eduroam
The research alliance between SURFnet and KPN will become apparent before the end of this year in a 4G/LTE pilot project, to be carried out at a Dutch campus. The pilot project is meant to investigate the value of an LTE network for the research and education sector. It will involve (1) rolling out a campus-wide LTE network on the 2.6 Ghz band, and (2) testing the smoothest possible connection between the campus WiFi network (backed by the SURFnet network) and the mobile network, based on eduroam credentials. The aim is to allow users to move freely, securely and without disruption in both domains.
The project will be the first 4G test in the Netherlands in a large-scale production environment, and the first in the world aimed at seamless integration between the WiFi and mobile networks. For SURFnet, it is important that any information gained through the pilot project is made available to the target group and the Internet community. Equally important, however, is to use that knowledge to dismantle barriers to wireless access on behalf of the target group.
Looking ahead
By collaborating with the authorities, businesses, partners, providers, and knowledge institutions, SURFnet will continue to drive innovation in the years ahead in order to remove barriers in the ICT infrastructure that are detrimental to the quality of education and research. Its research alliance with KPN is a good example of what SURFnet can achieve and will undoubtedly lead to more instances of such collaboration. SURFnet continues to challenge the market.