IPv6 Basic : FAQ - Revision history

Onnowpurbo at 03:30, 20 May 2013

2013-05-20T03:30:40Z

← Older revision		Revision as of 03:30, 20 May 2013
Line 30:		Line 30:
	Barangkali tidak banyak orang yang tahu tentang Jim Fleming yang berfikir tentang IPv8 dan IPv16, rancangan mereka masih jauh dari penerimaan dan implementasi. Pada saat ini, 128 bit tampaknya masih menjadi pilihan terbaik jika kita melihat overhead header dan data transport.		Barangkali tidak banyak orang yang tahu tentang Jim Fleming yang berfikir tentang IPv8 dan IPv16, rancangan mereka masih jauh dari penerimaan dan implementasi. Pada saat ini, 128 bit tampaknya masih menjadi pilihan terbaik jika kita melihat overhead header dan data transport.

			Dengan memperhitungkan bahwa minimum nilai dari Maximum Transfer Unit (MTU) di IPv4 (576 oktet) dan di IPv6 (1280 oktet), panjang header dari IPv4 adalah 20 oktet (minimum, dan dapat naik ke 60 oktet dengan IPv4 options) dan di IPv6 adalah 40 oktet (tetap). Hal ini berarti 3.4 % dari minimum MTU di IPv4 dan 3.1 % dari minimum MTU di IPv6. Ini berarti header overhead hampir sama. Dengan lebih banyak bit address akan membutuhkan lebih banyak header dan oleh karenanya lebih banyak overhead. Juga memperhatikan, maximum MTU di sambungan normal (seperti Ethernet hari ini) adalah 1500 oktet (untuk kasus khusus bisa 9k oktet menggunakan Jumbo frame).

			Pada akhirnya, bukan disain yang baik jika 10-20% data yang dikirim pada paket lapisan ke 3 digunakan untuk address bukan untuk payload / muatan data sebenarnya.


	~~While~~, there are (possibly) some people (only know about Jim Fleming...) on the Internet who are thinking about IPv8 and IPv16, their design is far away from acceptance and implementation. In the meantime 128 bits was the best choice regarding header overhead and data transport. Consider the minimum Maximum Transfer Unit (MTU) in IPv4 (576 octets) and in IPv6 (1280 octets), the header length in IPv4 is 20 octets (minimum, can increase to 60 octets with IPv4 options) and in IPv6 is 40 octets (fixed). This is 3.4 % of minimum MTU in IPv4 and 3.1 % of minimum MTU in IPv6. This means the header overhead is almost equal. More bits for addresses would require bigger headers and therefore more overhead. Also, consider the maximum MTU on normal links (like Ethernet today): it's 1500 octets (in special cases: 9k octets using Jumbo frames). Ultimately, it wouldn't be a proper design if 10 ~~% or~~ 20 % ~~of transported~~ data ~~in a Layer-~~3 ~~packet were used for addresses and not for~~ payload.

Onnowpurbo at 03:24, 20 May 2013

2013-05-20T03:24:11Z

← Older revision		Revision as of 03:24, 20 May 2013
Line 2:		Line 2:


	~~2.4.1. Why is the name~~ IPv6 ~~and not~~ IPv5 ~~as successor for~~ IPv4?		==Mengapa namanya IPv6 bukan IPv5 sebagai kelanjutan dari IPv4?

	~~On any~~ IP header, ~~the first~~ 4 ~~bits are reserved for~~ protocol ~~version~~. ~~So theoretically a protocol number between~~ 0 ~~and~~ 15 ~~is possible:~~		Pada IP header , 4 bit pertama di alokasikan untuk versi protocol. Oleh karenanya, secara teoritis nomor protokol antara 0 s/d 15 adalah mungkin.

	4: ~~is already used for~~ IPv4		* 4: sudah digunakan untuk IPv4.

	5: ~~is reserved for the~~ Stream Protocol (STP, RFC 1819 / Internet Stream Protocol Version 2) ~~(which never really made it to the public)~~		* 5: di alokasikan untuk Stream Protocol (STP, RFC 1819 / Internet Stream Protocol Version 2) yang ternyata tidak di adpso dengan baik oleh komunitas di Internet.

	~~The next free number was~~ 6. ~~Hence~~ IPv6 ~~was born!~~		* 6: Masih kosong, oleh karenanya IPv6 lahir! :) ..
	2.4.2. ~~IPv6 addresses~~: ~~why such a high number of bits~~?

			==Mengapa IPv6 address demikian banyak?==

			Pada saat IPv4 dirancang, orang berfikir 32 bit cukup untuk dunia. Memang pada masa lalu 32 bit mencukup kebutuhan yang ada. Akan tetapi, dengan banyak gadget, handphone, mobil, mungkin juga lemari es, sakelar lampu dll yang semua akan membutuhkan IP address maka 32 bit jelas tidak akan mencukupi kebutuhan.

			Para perancang memilih 128 bit, 4 kali lebih panjang daripada IPv4 yang ada sekarang.

			Sebetulnya bit yang bisa digunakan lebih sedikit daripada yang terlihat. Ini terjadi karena teknik skema pendefinisian address, 64 bit digunakan untuk interface identifier. 64 bit yang lain digunakan untuk routing. Dengan asumsi level agregasi yang sekarang (/48, /32, ..), maka kemungkinan untuk kehabisan space masih ada. Akan tetapi tidak di waktu yang dekat.

			Untuk informasi tentang effisiensi alokasi address, ada baiknya membaca-baca :

			* RFC 1715 / The H Ratio for Address Assignment Efficiency
			* RFC 3194 / The Host-Density Ratio for Address Assignment Efficiency.

			==Mengapa jumlah bit yang ada terlalu sedikit?==

			Barangkali tidak banyak orang yang tahu tentang Jim Fleming yang berfikir tentang IPv8 dan IPv16, rancangan mereka masih jauh dari penerimaan dan implementasi. Pada saat ini, 128 bit tampaknya masih menjadi pilihan terbaik jika kita melihat overhead header dan data transport.

	During the design of IPv4, people thought that 32 bits were enough for the world. Looking back into the past, 32 bits were enough until now and will perhaps be enough for another few years. However, 32 bits are not enough to provide each network device with a global address in the future. Think about mobile phones, cars (including electronic devices on its CAN-bus), toasters, refrigerators, light switches, and so on...

	~~So designers have chosen 128 bits, 4 times more in length than in IPv4 today.~~

	The usable size is smaller than it may appear however. This is because in the currently defined address schema, 64 bits are used for interface identifiers. The other 64 bits are used for routing. Assuming the current strict levels of aggregation (/48, /32, ...), it is still possible to “run out” of space, but hopefully not in the near future.

	~~See also for more information RFC 1715 / The H Ratio for Address Assignment Efficiency and RFC 3194 / The Host-Density Ratio for Address Assignment Efficiency.~~
	~~2.4.3. IPv6 addresses: why so small a number of bits on a new design?~~

	While, there are (possibly) some people (only know about Jim Fleming...) on the Internet who are thinking about IPv8 and IPv16, their design is far away from acceptance and implementation. In the meantime 128 bits was the best choice regarding header overhead and data transport. Consider the minimum Maximum Transfer Unit (MTU) in IPv4 (576 octets) and in IPv6 (1280 octets), the header length in IPv4 is 20 octets (minimum, can increase to 60 octets with IPv4 options) and in IPv6 is 40 octets (fixed). This is 3.4 % of minimum MTU in IPv4 and 3.1 % of minimum MTU in IPv6. This means the header overhead is almost equal. More bits for addresses would require bigger headers and therefore more overhead. Also, consider the maximum MTU on normal links (like Ethernet today): it's 1500 octets (in special cases: 9k octets using Jumbo frames). Ultimately, it wouldn't be a proper design if 10 % or 20 % of transported data in a Layer-3 packet were used for addresses and not for payload.		While, there are (possibly) some people (only know about Jim Fleming...) on the Internet who are thinking about IPv8 and IPv16, their design is far away from acceptance and implementation. In the meantime 128 bits was the best choice regarding header overhead and data transport. Consider the minimum Maximum Transfer Unit (MTU) in IPv4 (576 octets) and in IPv6 (1280 octets), the header length in IPv4 is 20 octets (minimum, can increase to 60 octets with IPv4 options) and in IPv6 is 40 octets (fixed). This is 3.4 % of minimum MTU in IPv4 and 3.1 % of minimum MTU in IPv6. This means the header overhead is almost equal. More bits for addresses would require bigger headers and therefore more overhead. Also, consider the maximum MTU on normal links (like Ethernet today): it's 1500 octets (in special cases: 9k octets using Jumbo frames). Ultimately, it wouldn't be a proper design if 10 % or 20 % of transported data in a Layer-3 packet were used for addresses and not for payload.

Onnowpurbo: New page: Sumber: http://tldp.org/HOWTO/Linux+IPv6-HOWTO/x447.html 2.4.1. Why is the name IPv6 and not IPv5 as successor for IPv4? On any IP header, the first 4 bits are reserved for protocol ver...

2013-04-28T08:15:40Z

New page: Sumber: http://tldp.org/HOWTO/Linux+IPv6-HOWTO/x447.html 2.4.1. Why is the name IPv6 and not IPv5 as successor for IPv4? On any IP header, the first 4 bits are reserved for protocol ver...

New page

Sumber: http://tldp.org/HOWTO/Linux+IPv6-HOWTO/x447.html

2.4.1. Why is the name IPv6 and not IPv5 as successor for IPv4?

On any IP header, the first 4 bits are reserved for protocol version. So theoretically a protocol number between 0 and 15 is possible:

4: is already used for IPv4

5: is reserved for the Stream Protocol (STP, RFC 1819 / Internet Stream Protocol Version 2) (which never really made it to the public)

The next free number was 6. Hence IPv6 was born!
2.4.2. IPv6 addresses: why such a high number of bits?

During the design of IPv4, people thought that 32 bits were enough for the world. Looking back into the past, 32 bits were enough until now and will perhaps be enough for another few years. However, 32 bits are not enough to provide each network device with a global address in the future. Think about mobile phones, cars (including electronic devices on its CAN-bus), toasters, refrigerators, light switches, and so on...

So designers have chosen 128 bits, 4 times more in length than in IPv4 today.

The usable size is smaller than it may appear however. This is because in the currently defined address schema, 64 bits are used for interface identifiers. The other 64 bits are used for routing. Assuming the current strict levels of aggregation (/48, /32, ...), it is still possible to “run out” of space, but hopefully not in the near future.

See also for more information RFC 1715 / The H Ratio for Address Assignment Efficiency and RFC 3194 / The Host-Density Ratio for Address Assignment Efficiency.
2.4.3. IPv6 addresses: why so small a number of bits on a new design?

While, there are (possibly) some people (only know about Jim Fleming...) on the Internet who are thinking about IPv8 and IPv16, their design is far away from acceptance and implementation. In the meantime 128 bits was the best choice regarding header overhead and data transport. Consider the minimum Maximum Transfer Unit (MTU) in IPv4 (576 octets) and in IPv6 (1280 octets), the header length in IPv4 is 20 octets (minimum, can increase to 60 octets with IPv4 options) and in IPv6 is 40 octets (fixed). This is 3.4 % of minimum MTU in IPv4 and 3.1 % of minimum MTU in IPv6. This means the header overhead is almost equal. More bits for addresses would require bigger headers and therefore more overhead. Also, consider the maximum MTU on normal links (like Ethernet today): it's 1500 octets (in special cases: 9k octets using Jumbo frames). Ultimately, it wouldn't be a proper design if 10 % or 20 % of transported data in a Layer-3 packet were used for addresses and not for payload.

==Referensi==

* http://tldp.org/HOWTO/Linux+IPv6-HOWTO/x447.html