Lacks of IPv4

Let's tell about IPv4 (RFC-791) - the basic part of stack TCP/IP at present and basic network protocol.
This protocol is engaged in routing in networks, i.e. it directs a package of a way from the sender up to the addressee. The IP protocol sends datagrams. Everyone such datagram except for transferred data comprises also header. Here is format of IPv4 header:

Version	IHL	TOS	Total Length
Identification			Flags	Fragment Offset
Time to Live		Protocol	Header Checksum
Source Address
Destination Address
IP options ...

Usually the heading contains 20 octets, i.e. has length of 20 bytes, but this value may be changed. Its all does not simplify process of data transmission.
The most interesting to us on the given theme fields of header are fields of IP addresses (the address of the sender and the destination address), a field "TOS" - Type of Service (initially not used, and now become outdated) and a field "Protocol" (defining structure weeding "data", i.e. according to what protocol they were coded).
So, we shall list common issues of IPv4:

Deficiency of address space - quantity various The devices connected to the Internet grows exponentially. The size of address space 2 ³² is quickly exhausted;
Weak expansibility of the protocol - the insufficient size of heading IPv4 doesn't allow to place demanded quantity of additional parameters in it;
Problem of safety of communications - it is not stipulated any means for differentiation of access to the information placed in a network;
Absence of support of quality of service (QoS) - accommodation of the information about throughput, the delays and demanded for normal work of some network appendices is not supported;
The problems connected with the mechanism of a fragmentation - the size of the maximal block of data transmission on each concrete way is not defined;
Absence of the autoconfiguration ip addresses mechanism;
Problem of renumbering of machines;

Fields "IP addresses" contain IP addresses of the sender and the addressee. In IPv4 the IP address consists from 4 bytes and often it is represented in the form of 4 numbers of size 1 byte, divided by points. This formatting gives hardly more than 4 billion various addresses. In such scheme each computer in the Internet has the unique address. But at occurrence the Internet-resources of the address were distributed by huge blocks. So Massachusetts university has the block in 16 million addresses when the average provider has some thousand addresses. Such "wasting" of IP addresses has led to that their space has started to come to an end and be felt shortage. This day when the 32-bit space of addresses will end that will lead to a stop of development of the Internet. Though have been taken measures of the decision of this problem (for example technology of allocation of blocks on 2 addresses and technology of distribution of dynamical DHCP IP-addresses and NAT that allows to broadcast IP addresses from an external network in internal) but "ending" managed to be delayed only. At present rates of growth of the Internet, according to IPv4 Address Space Report ( bgp.potaroo.net/ipv4 ) free addresses will end by 2018 and 32-bit address space will end in 2040 (but there are much worse opinions).
The problem of address space IPv4 is considered the basic, but it not main and not single.
There are also problems connected with transportation of the data, routings consisting complexity and growth of routing tables. Though this problem also tried to solve introduction by classless of interdomain routing (CIDR) but it was not resolved, but IP-administration from it has become complicated. All difficultly to construct and order structure of addresses in 32-bit space...
Then, complexity of IP header processing. Essential fields' number which it contains does not facilitate a life to the processes that work with such headings. In fact some fields of heading have already lost the importance and others could be simplified. Also the variable length of header makes to recalculate constantly the control sum, and at high speeds it will strongly load the processor.
Also there are no some mechanisms necessary to modern measures in IPv4. No mechanisms of information safety and supporting of classes of service. There are no methods of enciphering of data, which now in practice helps very much. And such means should be realized at a network level to not slow applications. To provide support of classes of service is the router's owe to not to rise this problem on the application level that will lead to complication and instability of work.
Realizing these mechanisms on IPv4 in addition is wasting of time. We anyway will have to make changes in all TCP/IP stack, that in any case will lead to some complication of work after transition to the new standard. What for we need all defects of IPv4 if it is possible to realize all at once by means of IPv6?
As a result, it is possible to tell, that IPv4 has served successfully almost 30 years and could serve further, but it is not necessary to stay on the out-of-date protocol that contains incorrigible lacks at a network level when on change to it the excellent alternative: IPv6 for a long time is ready.