IPv6 protocol was first defined in 1998. The main reason for developing a new internet protocol was based on lack of address; however this was not the only reason. Unfortunately, many people think of IPv6 only as enormous address space, but there are a lot of other advantages, for example:

• Group Addressing - possibility of communication in big groups of users. It is very important for large data transfer such as video stream or videoconference. Datastream is not delivered to every user (unicast) but to the whole group of users (multicast). IPv6 also enables communication with the nearest node (anycast).
• Security Function - authorizations and authentication function are implemented directly in the protocol and are mandatory.
• Auto Configuration - automatic configuration of network interfaces based on their physical address
• Flow Priority - protocol itself recognizes data streams which must be transmitted in real time, and the data must be processed with highest priority.
• Mobility - possibility of migration between many networks with the user’s home network address.

New protocol has also simplified header, which speeds up its processing on routers. Increasing of speed is also available because the fragmentation occurs on end stations not on routers. Mobile device support should be the main reason to implement this protocol into practice, which, in reality, is not true. Due to NAT technology, the address space is not exhausted and IPv6 protocol is mainly used only in academic networks. We approached universities in Slovakia to find out whether they use the new IPv6 protocol. The result of this research is described later on this paper. There are three ways how to connect to IPv6. We used some tests to compare these methods of connection into IPv6 network to show the need of IPv6. 1. SURVEY Protocol IPv4 is keeping its post because of the commercial sphere, so it is up to academic sphere to fight for its expansion. There is GÉANT2 network in Europe which interconnects national academic networks. GÉANT2 natively supports IPv6 protocol. Academic network in Slovakia is SANET (Slovak Academic Network). It also natively supports this protocol. We made a research on 15 Slovak university departments (Centers of ICT or Departments of Computer Science).We asked them whether they use IPv6 protocol and so what extend. We have obtained replies from 9 of these departments. University of Presov The support of IPv6 protocol is only on main SANET router. The University does not use this protocol. This university has 19 blocks of IPv4 C class addresses. University of SS Cyril and Methodius in Trnava Strictly uses only IPv4. Regarding to RIPE database, this University has only one C class range assigned. Academy of Arts in Banska Bystrica This university does not deal with this problem and has one C class range assigned. The Catholic University in Ruzomberok Faculty of Education has created VLAN networks, all backbone devices are IPv6 compatible but they do not use it. Regarding to RIPE this University has ten C classes assigned. Technical University of Kosice Formerly they had an IPv6 project as an effort of a few students. They routed IPv6 traffic from Cisco lab to the college. Later they requested IPv6 address range, but it was too small and incorrectly designed. Now they do not use IPv6 even though it is supported and configured on SANET router. This university has one B class address range assigned of IPv4, but no IPv6 address range. Most probably they had IPv6 address space when using 6bone. Comenius University in Bratislava This University does not use IPv6 protocol and they do not intend to implement it. In the past some departments tested IPv6 in labs, but they did not use it in practice. Comenius University has, as Technical University of Kosice, assigned one B class of IP4 address space. Matej Bel University in Banska Bystrica MBU uses only direct IPv4 addressing and all active network devices and end stations are configured for using IPv4 protocol. Technically they support IPv6, but the University does not use it. The MBU has 19 C classes assigned of IPv4, but also has IPv6 address range 2001:4118:0100::/48. University of Zilina It has one B class IPv4 address range assigned and also 2001:4118:300::/48 IPv6 address range. IPv6 services are natively accessible via IPv6. They use OSPF routing protocol in their network. Services accessible via IPv6 are DNS: nic.uniza.sk., proxy.uniza.sk. WWW: www.uniza.sk, nic.uniza.sk, EMAIL (pop3, imap): nic.uniza.sk, LDAP: nic.uniza.sk, FTP: nic.uniza.sk and are running on Linux and Windows 2003 servers. Slovak University of Agriculture in Nitra This university does not use IPv6 and has not done any research in this field. RIPE has twelve C classes assigned. Constantine the Philosopher University in Nitra All end hosts and servers are using direct IPv4 addressing. Nowadays the University has eight C classes of IPv4 addresses assigned. This range is for the University needs too small so the University is trying do get more IPv4 addresses. The rules are too strict and the process is too complicated. Most servers have UNIX operating system so the possibility to use IPv6 is open. The University have no IPv6 address space assigned. Thanks to auto configuration mechanism, all servers have link local IPv6 addresses (FE80::/10). All workstations use Windows XP, so after enabling IPv6 support with command ipv6 install all have link local address and also 6to4 (RFC3056) addresses from 2002::/16 range. 2. POSSIBILITIES OF USING IPV6 Most of our universities do not use IPv6 protocol. The answers for the question why are various. They are mostly saying “Why to change anything that works?”, “If you want to use IPv6, just enable it in your Windows and you will get 6to4 address.” We are looking for the method how to show people, that the best way for using IPv6 is its native form. There are three ways how to connect into an IPv6 network: using native IPv6 network and have IPv6 address space assigned from RIPE or using 6to4 or 6in4 mechanism. Both of them encapsulate IPv6 packets into IPv4 ones. The main difference between them is that in 6in4 it is necessary to establish en explicit tunnel between a host and a server. The tunnel is established and the configuration is done by a Tunnel Broker. In 6to4 there is no need to establish the tunnel. IPv6 prefix is derived from IPv4 address of a host. We tested the speed of using 6in4 and 6to4 addresses against IPv4 addresses. The test of native IPv6 was not done because our university has not assigned IPv6 address space. 2.1 Tunneling via Broker - 6in4 First of all, we must find a Broker. There are many brokers, e.g. Freenet6, Huricane Electric, SixXS, S26, SingNet. We decided to use Hexago which provides connectivity into a Freenet6 network. A tunnel registration is divided into three steps. First, we must fill a registration form www.hexago.com. Second, we need to install a TSP client - a virtual network adapter. Third, we need to run the TSP client. After this is done, we get 2001:5c0:8fff:fffe::44b7 IPv6 address. We made this procedure on two separate hosts and the second one had 2001:5c0:8fff:fffe::44b9 address. After this, we ran a ping test between those two hosts (Figure 1)

C:\Documents and Settings\Petko>ping6 2001:5c0:8fff:fffe::44b9 Pinging 2001:5c0:8fff:fffe::44b9 from 2001:5c0:8fff:fffe::44b7 with 32 bytes of data: Reply from 2001:5c0:8fff:fffe::44b9: bytes=32 time=340ms Reply from 2001:5c0:8fff:fffe::44b9: bytes=32 time=372ms Reply from 2001:5c0:8fff:fffe::44b9: bytes=32 time=315ms Reply from 2001:5c0:8fff:fffe::44b9: bytes=32 time=340ms Ping statistics for 2001:5c0:8fff:fffe::44b9: Packets: Sent = 4, Received = 4, Lost = 0 (0% loss), Approximate round trip times in milli-seconds: Minimum = 315ms, Maximum = 372ms, Average = 341ms

Figure 1. Ping test between to 6in4 hosts 3.2 Automatic tunnel - 6to4 Assigning a 6to4 IPv6 address is very simple. All we have to do is to enable IPv6 support on the interface and have public IPv4 address. After this, the system is configured with IPv6 address from 2002::/16 range. We ran a test on workstation located in Constantine the Philosopher University in Nitra to a DNS server located at University of Zilina. Both universities are connected to SANET and Zilina has native IPv6 support. First, we made traceroute to count hops in IPv4 network and IPv6 network. The results can be seen in the Figures 2 and 3. When we used 6to4 address, there are twelve hops. The reason for this is that the packets do not stay in the SANET network. First we have to contact relay router with IPv4 anycast address 192.88.99.1 which, in our case, is situated in Portugal.

tracert nic.uniza.sk Tracing route to nic.uniza.sk [158.193.48.33] over a maximum of 30 hops: 1 <1 ms <1 ms <1 ms gw.ukf.sk [194.160.208.1] 2 2 ms 1 ms 1 ms Wgw-fo-c6k5.nr.sanet.sk [193.87.99.1] 3 5 ms 3 ms 4 ms ZU-Zilina.sanet2.sk [194.160.8.197] 4 8 ms 5 ms 5 ms sw-vd-a.net.utc.sk [158.193.7.241] 5 * * * Request timed out.

Figure 2. Tracing route in IPv4 network

tracert6 nic.uniza.sk Tracing route to nic.uniza.sk [2001:4118:300:48::33] from 2002:c2a0:d2ce::c2a0:d2ce over a maximum of 30 hops: 1 58 ms 60 ms 58 ms ::192.88.99.1 2 58 ms 62 ms 67 ms ROUTER1.IPv6.GE.Lisboa.fccn.pt   [2001:690:800:1::1] 3 78 ms 124 ms 60 ms fccn.rt1.lis.pt.geant2.net   [2001:798:24:10aa::1] 4 62 ms 62 ms 58 ms 2001:798:cc:1701:2401::1 5 58 ms 59 ms 59 ms so-7-2-0.rt1.gen.ch.geant2.net   [2001:798:cc:1201:1701::1] 6 72 ms 68 ms 68 ms so-7-2-0.rt1.fra.de.geant2.net   [2001:798:cc:1201:1401::2] 7 93 ms 74 ms 74 ms so-6-0-0.rt1.pra.cz.geant2.net   [2001:798:cc:1301:1401::1] 8 83 ms 79 ms 79 ms cz.sk1.sk.geant.net   [2001:798:20cc:1301:2701::2] 9 81 ms 79 ms 79 ms sanet-gw.sk1.sk.geant.net   [2001:798:2027:10aa::2] 10 83 ms 82 ms 84 ms ZU-Zilina.sanet2.sk [2001:4118::c5] 11 83 ms 85 ms 96 ms sw-vd-a5.net.utc.sk   [2001:4118:300::7708:215] 12 81 ms 81 ms 83 ms nic.uniza.sk [2001:4118:300:48::33] Trace complete.

Figure 3. Tracing route in IPv6 network 3.CONCLUSION Although Slovak Academic Network supports native IPv6 and routes IPv6 packets, the usage of this protocol is very low. Only two universities have IPv6 addresses assigned, yet only one on them uses it in practice. The reason most probably lies in no need for IPv6 because most of them have enough IPv4 addresses and do not want to change something that works. Some universities use private IP addresses and NAT technology for end users. NAT decreases end-to-end communication which is one of the basic principles of IP. On the Constantine the Philosopher University we are ready to request IPv6 address space and implement it on all servers and most workstations at the university. We are looking for applications that will convince users to switch to IPv6. There is a problem with Windows XP as IPv6 is disabled by default in this operating system, but this problem can be solved. Windows Vista is a great support for IPv6 implementation IPv6 is enabled in this operation system by default and preferred. We also tested the other two ways of connecting to IPv6 network which is via Broker and via IPv4 compatible IPv6 address. The connection via Broker works well only for testing. The problem is that any traffic is done via the tunnel even if there is a shorter path than the one from other side of the tunnel. A problem with 6to4 address is, that the host sometimes does not connect to the nearest relay server. We can enforce it to use a different relay server, but we cannot expect a user to use a command prompt and type commands. We hope to obtain IPv6 address range in a near future. REFERENCES Deering, S., Hinden R., 1998, Internet Protocol, Version 6 (IPv6) Specification. RFC 2460 Huitema, C., 1997, IPv6: The New Internet Protocol, (Second edition). Prentice Hall, New Jersey, USA Miller, P.E., Miller M. A., 2000, Implementing IPv6: Supporting the Next Generation Internet Protocols, Faster City: IDG Books Worldwide Satrapa, P., 2007, IPv6 v MS Windows Vista. 2007. [online]. Satrapa, P., 2002, IPv6. Vydavate?stvo Neocortex spol. s. r. o., Praha, Czech Republic Svec, Peter: IPv6 in Slovak Academic Network. In: IADIS Multi-Conference on Computer Science and Information Systems 2007. - Lisabon: Universidade de Lisboa, 2007. - ISBN 978-972-8924-40-9. - S. 120-123.