原文鏈接：https://datatracker.ietf.org/doc/html/rfc8445#appendix-B

附錄B. Design Motivations【設(shè)計意圖】

ICE contains a number of normative behaviors that may themselves be simple but derive from complicated or non-obvious thinking or use cases that merit further discussion.
ICE 包含許多規(guī)范行為，這些行為本身可能很簡單，但源自復(fù)雜或不明顯的想法或值得進一步討論的用例。

Since these design motivations are not necessary to understand for purposes of implementation, they are discussed here. This appendix is non-normative.
由于這些設(shè)計動機不是為了實現(xiàn)目的而必須理解的，因此在這里對其進行討論。 本附錄是非規(guī)范性的。

B.1. Pacing of STUN Transactions【STUN事務(wù)的速度】

STUN transactions used to gather candidates and to verify connectivity are paced out at an approximate rate of one new transaction every Ta milliseconds.
用于收集候選和驗證連接性的 STUN 事務(wù)以大約每 Ta 毫秒一個新事務(wù)的速率進行。

Each transaction, in turn, has a retransmission timer RTO that is a function of Ta as well.
反過來，每個事務(wù)都有一個重傳計時器 RTO，它也是 Ta 的函數(shù)。

Why are these transactions paced, and why are these formulas used?
為什么這些交易是有節(jié)奏的，為什么要使用這些公式？

Sending of these STUN requests will often have the effect of creating bindings on NAT devices between the client and the STUN servers.
發(fā)送這些 STUN 請求通常會在客戶端和 STUN 服務(wù)器之間的 NAT 設(shè)備上創(chuàng)建綁定。

Experience has shown that many NAT devices have upper limits on the rate at which they will create new bindings.
經(jīng)驗表明，許多 NAT 設(shè)備對它們創(chuàng)建新綁定的速率都有上限。

Discussions in the IETF ICE WG during the work on this specification concluded that once every 5 ms is well supported. This is why Ta has a lower bound of 5 ms.
IETF ICE WG 在本規(guī)范工作期間的討論得出結(jié)論，每 5 ms 一次得到很好的支持。 這就是 Ta 的下限為 5 ms 的原因。

Furthermore, transmission of these packets on the network makes use of bandwidth and needs to be rate limited by the ICE agent.
此外，這些數(shù)據(jù)包在網(wǎng)絡(luò)上的傳輸會占用帶寬，并且需要由 ICE 代理進行速率限制。

Deployments based on earlier draft versions of [RFC5245] tended to overload rate-constrained access links and perform poorly overall, in addition to negatively impacting the network.
基于 [RFC5245] 早期草案版本的部署往往會使速率受限的接入鏈路過載并且總體性能不佳，此外還會對網(wǎng)絡(luò)產(chǎn)生負面影響。

As a consequence, the pacing ensures that the NAT device does not get overloaded and that traffic is kept at a reasonable rate.
因此，調(diào)步可確保 NAT 設(shè)備不會過載，并將流量保持在合理的速率。

The definition of a “reasonable” rate is that STUN MUST NOT use more bandwidth than the RTP itself will use, once data starts flowing.
“合理”速率的定義是 STUN 使用的帶寬不得超過 RTP 本身使用的帶寬，一旦數(shù)據(jù)開始流動。

The formula for Ta is designed so that, if a STUN packet were sent every Ta seconds, it would consume the same amount of bandwidth as RTP packets, summed across all data streams.
Ta 的公式是這樣設(shè)計的，如果每 Ta 秒發(fā)送一個 STUN 數(shù)據(jù)包，它將消耗與 RTP 數(shù)據(jù)包相同數(shù)量的帶寬，所有數(shù)據(jù)流的總和。

Of course, STUN has retransmits, and the desire is to pace those as well.
當(dāng)然，STUN 有重傳，并且希望也能調(diào)整它們的速度。

For this reason, RTO is set such that the first retransmit on the first transaction happens just as the first STUN request on the last transaction occurs. Pictorially:
出于這個原因，設(shè)置 RTO 使得第一個事務(wù)的第一次重傳發(fā)生在最后一個事務(wù)的第一個 STUN 請求發(fā)生的同時。圖示：

In this picture, there are three transactions that will be sent (for example, in the case of candidate gathering, there are three host candidate/STUN server pairs).
在這張圖中，將發(fā)送三個事務(wù)（例如，在候選收集的情況下，有三個主機候選人/STUN 服務(wù)器對）。

These are transactions A, B, and C. The retransmit timer is set so that the first retransmission on the first transaction (packet A2) is sent at time 3Ta.
這些是事務(wù)A、B和C。重傳計時器被設(shè)置為在時間3Ta發(fā)送第一個事務(wù)（分組A2）的第一個重傳。

Subsequent retransmits after the first will occur even less frequently than Ta milliseconds apart, since STUN uses an exponential backoff on its retransmissions.
在第一次重傳之后的后續(xù)重傳發(fā)生的頻率甚至低于 Ta 毫秒間隔，因為 STUN 在其重傳上使用指數(shù)退避。

This mechanism of a global minimum pacing interval of 5 ms is not generally applicable to transport protocols, but it is applicable to ICE based on the following reasoning.
這種全局最小來回間隔為 5 ms 的機制一般不適用于傳輸協(xié)議，但基于以下推理，它適用于 ICE。

• Start with the following rules that would be generally applicable to transport protocols:
  從以下通常適用于傳輸協(xié)議的規(guī)則開始

Let MaxBytes be the maximum number of bytes allowed to be outstanding in the network at startup, which SHOULD be 14600, as defined in Section 2 of [RFC6928].
讓 MaxBytes 為啟動時允許在網(wǎng)絡(luò)中未完成的最大字節(jié)數(shù)，應(yīng)為 14600，如 [RFC6928] 第 2 節(jié)中所定義。

Let HTO be the transaction timeout, which SHOULD be 2RTT if RTT is known or 500 ms otherwise. This is based on the RTO for STUN messages from [RFC5389] and the TCP initial RTO, which is 1 sec in [RFC6298].
讓 HTO 為事務(wù)超時，如果 RTT 已知，則應(yīng)為 2RTT，否則為 500 ms。這基于來自 [RFC5389] 的 STUN 消息的 RTO 和 TCP 初始 RTO，在 [RFC6298] 中為 1 秒。

Let MinPacing be the minimum pacing interval between transactions, which is 5 ms (see above).
讓 MinPacing 為事務(wù)之間的最小來回間隔，即 5 毫秒（見上文）。

• Observe that agents typically do not know the RTT for ICE transactions (connectivity checks in particular), meaning that HTO will almost always be 500 ms.
  觀察代理通常不知道 ICE 交易的 RTT（特別是連接檢查），這意味著 HTO 幾乎總是 500 毫秒。
• Observe that a MinPacing of 5 ms and HTO of 500 ms gives at most 100 packets/HTO, which for a typical ICE check of less than 120 bytes means a maximum of 12000 outstanding bytes in the network, which is less than the maximum expressed by rule 1.
  觀察到 5 ms 的 MinPacing 和 500 ms 的 HTO 最多提供 100 個數(shù)據(jù)包/HTO，對于小于 120 字節(jié)的典型 ICE 檢查，這意味著網(wǎng)絡(luò)中最多有 12000 個未完成的字節(jié)，這小于表示的最大值根據(jù)規(guī)則 1。
• Thus, for ICE, the rule set reduces to just the MinPacing rule, which is equivalent to having a global Ta value.
  因此，對于 ICE，規(guī)則集簡化為僅 MinPacing 規(guī)則，這相當(dāng)于具有全局 Ta 值。

B.2. Candidates with Multiple Bases【具有多個Base的候選】

Section 5.1.3 talks about eliminating candidates that have the same transport address and base. However, candidates with the same transport addresses but different bases are not redundant.
第 5.1.3 節(jié)討論了消除具有相同傳輸?shù)刂泛突返暮蜻x者。 然而，具有相同傳輸?shù)刂返煌鶖?shù)的候選不是冗余的。

When can an ICE agent have two candidates that have the same IP address and port but different bases? Consider the topology of Figure 11:
ICE 代理何時可以有兩個具有相同 IP 地址和端口但bases不同的候選？ 考慮圖 11 的拓撲：

圖 11: 具有不同Bases的相同候選Bases

In this case, the initiating agent is multihomed. It has one IP address, 10.0.1.100, on network C, which is a net 10 private network.
在這種情況下，啟動代理是多宿主的。它在網(wǎng)絡(luò) C 上有一個 IP 地址 10.0.1.100，該網(wǎng)絡(luò)是一個 net 10 專用網(wǎng)絡(luò)。

The responding agent is on this same network. The initiating agent is also connected to network A, which is 192.168/16, and has an IP address of 192.168.1.100.
響應(yīng)代理位于同一網(wǎng)絡(luò)上。發(fā)起代理還連接到網(wǎng)絡(luò) A，即 192.168/16，IP 地址為 192.168.1.100。

There is a NAT on this network, natting into network B, which is another net 10 private network, but it is not connected to network C. There is a STUN server on network B.
這個網(wǎng)絡(luò)上有一個NAT，natting到網(wǎng)絡(luò)B，這是另一個net 10私有網(wǎng)絡(luò)，但它沒有連接到網(wǎng)絡(luò)C。網(wǎng)絡(luò)B上有一個STUN服務(wù)器。

The initiating agent obtains a host candidate on its IP address on network C (10.0.1.100:2498) and a host candidate on its IP address on network A (192.168.1.100:3344).
發(fā)起代理在網(wǎng)絡(luò) C (10.0.1.100:2498) 上的 IP 地址上獲得一個候選主機，在網(wǎng)絡(luò) A (192.168.1.100:3344) 上的 IP 地址上獲得一個候選主機。

It performs a STUN query to its configured STUN server from 192.168.1.100:3344.
它從 192.168.1.100:3344 對其配置的 STUN 服務(wù)器執(zhí)行 STUN 查詢。

This query passes through the NAT, which happens to assign the binding 10.0.1.100:2498.
此查詢通過 NAT，它恰好分配了綁定 10.0.1.100:2498。

The STUN server reflects this in the STUN Binding response. Now, the initiating agent has obtained a server-reflexive candidate with a transport address that is identical to a host candidate (10.0.1.100:2498).
STUN 服務(wù)器在 STUN 綁定響應(yīng)中反映了這一點。現(xiàn)在，啟動代理已經(jīng)獲得了一個服務(wù)器自反候選，其傳輸?shù)刂放c主機候選 (10.0.1.100:2498) 相同。

However, the server-reflexive candidate has a base of 192.168.1.100:3344, and the host candidate has a base of 10.0.1.100:2498.
但是，服務(wù)器自反候選的基數(shù)為 192.168.1.100:3344，而主機候選的基數(shù)為 10.0.1.100:2498。

B.3. Purpose of the Related-Address and Related-Port Attributes【相關(guān)地址和相關(guān)端口屬性的用途】

The candidate attribute contains two values that are not used at all by ICE itself – related address and related port. Why are they present?
候選屬性包含 ICE 本身根本不使用的兩個值——相關(guān)地址和相關(guān)端口。 他們?yōu)槭裁丛趫?#xff1f;

There are two motivations for its inclusion. The first is diagnostic. It is very useful to know the relationship between the different types of candidates.
包含它有兩個目的。 首先是診斷。 了解不同類型候選之間的關(guān)系非常有用。

By including it, an ICE agent can know which relayed candidate is associated with which reflexive candidate, which in turn is associated with a specific host candidate.
通過包含它，ICE 代理可以知道哪個中繼候選人與哪個reflexive候選相關(guān)聯(lián)，而哪個reflexive候選又與特定的host候選相關(guān)聯(lián)。

When checks for one candidate succeed but not for others, this provides useful diagnostics on what is going on in the network.
當(dāng)對一個候選的檢查成功但對其他候選沒有成功時，這為網(wǎng)絡(luò)中發(fā)生的事情提供了有用的診斷。

The second reason has to do with off-path Quality-of-Service (QoS) mechanisms.
第二個原因與非路徑服務(wù)質(zhì)量 (QoS) 機制有關(guān)。

When ICE is used in environments such as PacketCable 2.0, proxies will, in addition to performing normal SIP operations, inspect the SDP in SIP messages and extract the IP address and port for data traffic.
在 PacketCable 2.0 等環(huán)境中使用 ICE 時，代理除了執(zhí)行正常的 SIP 操作外，還會檢查 SIP 消息中的 SDP 并提取 IP 地址和端口以進行數(shù)據(jù)流量。

They can then interact, through policy servers, with access routers in the network, to establish guaranteed QoS for the data flows.
然后，它們可以通過策略服務(wù)器與網(wǎng)絡(luò)中的接入路由器交互，為數(shù)據(jù)流建立有保證的 QoS。

This QoS is provided by classifying the RTP traffic based on 5-tuple and then providing it a guaranteed rate, or marking its DSCP appropriately.
此 QoS 是通過基于 5 元組對 RTP 流量進行分類，然后為其提供保證速率或適當(dāng)標(biāo)記其 DSCP 來提供的。

When a residential NAT is present, and a relayed candidate gets selected for data, this relayed candidate will be a transport address on an actual TURN server.
當(dāng)一個住宅 NAT 存在，并且一個relayed候選被選中用于數(shù)據(jù)時，這個relayed候選將是實際 TURN 服務(wù)器上的傳輸?shù)刂贰?/p>

That address says nothing about the actual transport address in the access router that would be used to classify packets for QoS treatment.
該地址沒有說明接入路由器中用于對數(shù)據(jù)包進行分類以進行 QoS 處理的實際傳輸?shù)刂贰?/p>

Rather, the server-reflexive candidate towards the TURN server is needed.
相反，需要針對 TURN 服務(wù)器的server-reflexive候選。

By carrying the translation in the SDP, the proxy can use that transport address to request QoS from the access router.
通過在 SDP 中進行轉(zhuǎn)換，代理可以使用該傳輸?shù)刂废蚪尤肼酚善髡埱?QoS。

B.4. Importance of the STUN Username【STUN 用戶名的重要性】

ICE requires the usage of message integrity with STUN using its short-term credential functionality.
ICE 需要使用 STUN 的短期憑證功能來使用消息完整性。

The actual short-term credential is formed by exchanging username fragments in the candidate exchange.
實際的短期憑證是通過在候選交換中交換用戶名片段形成的。

The need for this mechanism goes beyond just security; it is actually required for correct operation of ICE in the first place.
對這種機制的需求不僅僅是安全性； 實際上首先需要正確操作 ICE。

Consider ICE agents L, R, and Z. L and R are within private enterprise 1, which is using 10.0.0.0/8. Z is within private enterprise 2, which is also using 10.0.0.0/8.
考慮 ICE 代理 L、R 和 Z。L 和 R 在使用 10.0.0.0/8 的私營企業(yè) 1 內(nèi)。 Z 在私有企業(yè) 2 內(nèi)，該企業(yè)也在使用 10.0.0.0/8。

As it turns out, R and Z both have IP address 10.0.1.1. L sends candidates to Z. Z responds to L with its host candidates.
事實證明，R 和 Z 的 IP 地址都是 10.0.1.1。 L 將候選發(fā)送給 Z。Z 用它的宿主候選者響應(yīng) L。

In this case, those candidates are 10.0.1.1:8866 and 10.0.1.1:8877.
在這種情況下，這些候選是 10.0.1.1:8866 和 10.0.1.1:8877。

As it turns out, R is in a session at that same time and is also using 10.0.1.1:8866 and 10.0.1.1:8877 as host candidates.
事實證明，R 同時在一個會話中，并且還使用 10.0.1.1:8866 和 10.0.1.1:8877 作為主機候選。

This means that R is prepared to accept STUN messages on those ports, just as Z is.
這意味著 R 準備好在這些端口上接受 STUN 消息，就像 Z 一樣。

L will send a STUN request to 10.0.1.1:8866 and another to 10.0.1.1:8877.
L 將向 10.0.1.1:8866 發(fā)送 STUN 請求，向 10.0.1.1:8877 發(fā)送另一個請求。

However, these do not go to Z as expected. Instead, they go to R! If R just replied to them, L would believe it has connectivity to Z, when in fact it has connectivity to a completely different user, R.
然而，這些并沒有像預(yù)期的那樣到達 Z。相反，他們?nèi)！如果 R 只是回復(fù)他們，L 會認為它與 Z 有連接，而實際上它與完全不同的用戶 R 有連接。

To fix this, STUN short-term credential mechanisms are used. The username fragments are sufficiently random; thus it is highly unlikely that R would be using the same values as Z.
為了解決這個問題，使用了 STUN 短期憑證機制。用戶名片段足夠隨機；因此，R 極不可能使用與 Z 相同的值。

Consequently, R would reject the STUN request since the credentials were invalid.
因此，R 將拒絕 STUN 請求，因為憑證無效。

In essence, the STUN username fragments provide a form of transient host identifiers, bound to a particular session established as part of the candidate exchange.
本質(zhì)上，STUN 用戶名片段提供了一種臨時主機標(biāo)識符，綁定到作為候選交換的一部分而建立的特定會話。

An unfortunate consequence of the non-uniqueness of IP addresses is that, in the above example, R might not even be an ICE agent.
IP 地址不唯一性的一個不幸后果是，在上面的示例中，R 甚至可能不是 ICE 代理。

It could be any host, and the port to which the STUN packet is directed could be any ephemeral port on that host.
它可以是任何主機，STUN 數(shù)據(jù)包指向的端口可以是該主機上的任何臨時端口。

If there is an application listening on this socket for packets, and it is not prepared to handle malformed packets for whatever protocol is in use, the operation of that application could be affected.
如果有應(yīng)用程序在此套接字上偵聽數(shù)據(jù)包，并且它不準備為正在使用的任何協(xié)議處理格式錯誤的數(shù)據(jù)包，則該應(yīng)用程序的操作可能會受到影響。

Fortunately, since the ports exchanged are ephemeral and usually drawn from the dynamic or registered range, the odds are good that the port is not used to run a server on host R, but rather is the agent side of some protocol.
幸運的是，由于交換的端口是短暫的，并且通常來自動態(tài)或注冊范圍，因此該端口很有可能不用于在主機 R 上運行服務(wù)器，而是用于某些協(xié)議的代理端。

This decreases the probability of hitting an allocated port, due to the transient nature of port usage in this range.
由于此范圍內(nèi)端口使用的瞬態(tài)特性，這降低了命中分配端口的可能性。

However, the possibility of a problem does exist, and network deployers need to be prepared for it. Note that this is not a problem specific to ICE; stray packets can arrive at a port at any time for any type of protocol, especially ones on the public Internet.
但是，確實存在問題的可能性，網(wǎng)絡(luò)部署人員需要為此做好準備。 請注意，這不是 ICE 特有的問題； 對于任何類型的協(xié)議，尤其是公共 Internet 上的協(xié)議，雜散數(shù)據(jù)包可以隨時到達端口。

As such, this requirement is just restating a general design guideline for Internet applications – be prepared for unknown packets on any port.
因此，此要求只是重申了 Internet 應(yīng)用程序的一般設(shè)計指南——為任何端口上的未知數(shù)據(jù)包做好準備。

B.5. The Candidate Pair Priority Formula【候選對優(yōu)先級公式】

The priority for a candidate pair has an odd form. It is:
候選對的優(yōu)先級具有奇數(shù)形式：

pair priority = 2^32MIN(G,D) + 2MAX(G,D) + (G>D?1:0)

Why is this? When the candidate pairs are sorted based on this value, the resulting sorting has the MAX/MIN property.
為什么是這樣？ 當(dāng)候選對基于此值排序時，生成的排序具有 MAX/MIN 屬性。

This means that the pairs are first sorted based on decreasing value of the minimum of the two priorities.
這意味著首先根據(jù)兩個優(yōu)先級中最小值的遞減值對這些對進行排序。

For pairs that have the same value of the minimum priority, the maximum priority is used to sort amongst them.
對于具有相同最小優(yōu)先級值的對，使用最大優(yōu)先級在它們之間進行排序。

If the max and the min priorities are the same, the controlling agent’s priority is used as the tiebreaker in the last part of the expression.
如果最大和最小優(yōu)先級相同，則控制代理的優(yōu)先級用作表達式最后部分的決勝局。

The factor of 232 is used since the priority of a single candidate is always less than 232, resulting in the pair priority being a “concatenation” of the two component priorities.
使用 232 因子是因為單個候選的優(yōu)先級總是小于 232，導(dǎo)致對優(yōu)先級是兩個組件優(yōu)先級的“串聯(lián)”。

This creates the MAX/MIN sorting. MAX/MIN ensures that, for a particular ICE agent, a lower-priority candidate is never used until all higher-priority candidates have been tried.
這將創(chuàng)建 MAX/MIN 排序。 MAX/MIN 確保對于特定的 ICE 代理，在嘗試所有較高優(yōu)先級的候選人之前，永遠不會使用較低優(yōu)先級的候選人。

B.6. Why Are Keepalives Needed?【為什么需要 Keepalives】

Once data begins flowing on a candidate pair, it is still necessary to keep the bindings alive at intermediate NATs for the duration of the session.
一旦數(shù)據(jù)開始在候選對上流動，仍然需要在會話期間在中間 NAT 上保持綁定活動。

Normally, the data stream packets themselves (e.g., RTP) meet this objective. However, several cases merit further discussion.
通常，數(shù)據(jù)流包本身（例如，RTP）滿足這個目標(biāo)。然而，有幾個案例值得進一步討論。

Firstly, in some RTP usages, such as SIP, the data streams can be “put on hold”. This is accomplished by using the SDP “sendonly” or “inactive” attributes, as defined in RFC 3264.
首先，在某些 RTP 用途中，例如 SIP，數(shù)據(jù)流可以“暫停”。這是通過使用 RFC 3264 中定義的 SDP“sendonly”或“inactive”屬性來完成的。

RFC 3264 directs implementations to cease transmission of data in these cases. However, doing so may cause NAT bindings to time out, and data won’t be able to come off hold.
RFC 326 指示實現(xiàn)在這些情況下停止數(shù)據(jù)傳輸。但是，這樣做可能會導(dǎo)致 NAT 綁定超時，并且數(shù)據(jù)將無法暫停。

Secondly, some RTP payload formats, such as the payload format for text conversation [RFC4103], may send packets so infrequently that the interval exceeds the NAT binding timeouts.
其次，某些 RTP 有效負載格式，例如文本對話的有效負載格式 [RFC4103]，可能發(fā)送數(shù)據(jù)包的頻率太低，以至于間隔超過 NAT 綁定超時。

Thirdly, if silence suppression is in use, long periods of silence may cause data transmission to cease sufficiently long for NAT bindings to time out.
第三，如果使用靜音抑制，長時間的靜音可能會導(dǎo)致數(shù)據(jù)傳輸停止足夠長的時間，以使 NAT 綁定超時。

For these reasons, the data packets themselves cannot be relied upon.
由于這些原因，不能依賴數(shù)據(jù)包本身。

ICE defines a simple periodic keepalive utilizing STUN Binding Indications.
ICE 使用 STUN 綁定指示定義了一個簡單的周期性保活。

This makes its bandwidth requirements highly predictable and thus amenable to QoS reservations.
這使得它的帶寬需求高度可預(yù)測，因此可以接受 QoS 預(yù)留。

B.7. Why Prefer Peer-Reflexive Candidates?【為什么更推薦Peer-Reflexive候選】

Section 5.1.2 describes procedures for computing the priority of a candidate based on its type and local preferences.
第 5.1.2 節(jié)描述了根據(jù)候選人的類型和本地偏好計算候選人優(yōu)先級的過程。

That section requires that the type preference for peer-reflexive candidates always be higher than server reflexive.
該部分要求peer-reflexive候選的類型偏好始終高于server reflexive候選。

Why is that? The reason has to do with the security considerations in Section 19.
這是為什么？ 原因與第 19 節(jié)中的安全考慮有關(guān)。

It is much easier for an attacker to cause an ICE agent to use a false server-reflexive candidate rather than a false peer-reflexive candidate.
攻擊者更容易讓 ICE 代理使用虛假的server-reflexive候選，而不是虛假的對等自反候選者。

Consequently, attacks against address gathering with Binding requests are thwarted by ICE by preferring the peer-reflexive candidates.
因此，ICE 通過優(yōu)先選擇peer-reflexiv候選來阻止對使用綁定請求進行地址收集的攻擊。

B.8. Why Are Binding Indications Used for Keepalives?【為什么將綁定指示用于保活】

Data keepalives are described in Section 11. These keepalives make use of STUN when both endpoints are ICE capable.
數(shù)據(jù)保活在第 11 節(jié)中描述。當(dāng)兩個端點都支持 ICE 時，這些保活使用 STUN。

However, rather than using a Binding request transaction (which generates a response), the keepalives use an Indication.
但是，keepalive 不使用綁定請求事務(wù)（生成響應(yīng)），而是使用指示。

Why is that? The primary reason has to do with network QoS mechanisms.
這是為什么？ 主要原因與網(wǎng)絡(luò) QoS 機制有關(guān)。

Once data begins flowing, network elements will assume that the data stream has a fairly regular structure, making use of periodic packets at fixed intervals, with the possibility of jitter.
一旦數(shù)據(jù)開始流動，網(wǎng)絡(luò)元素將假定數(shù)據(jù)流具有相當(dāng)規(guī)則的結(jié)構(gòu)，以固定間隔使用周期性數(shù)據(jù)包，并可能出現(xiàn)抖動。

If an ICE agent is sending data packets, and then receives a Binding request, it would need to generate a response packet along with its data packets.
如果 ICE 代理正在發(fā)送數(shù)據(jù)包，然后接收到綁定請求，則需要生成響應(yīng)包及其數(shù)據(jù)包。

This will increase the actual bandwidth requirements for the 5-tuple carrying the data packets and introduce jitter in the delivery of those packets.
這將增加攜帶數(shù)據(jù)包的 5 元組的實際帶寬需求，并在這些數(shù)據(jù)包的傳遞中引入抖動。

Analysis has shown that this is a concern in certain Layer 2 access networks that use fairly tight packet schedulers for data.
分析表明，在某些第 2 層接入網(wǎng)絡(luò)中，這是一個問題，這些網(wǎng)絡(luò)使用相當(dāng)嚴格的數(shù)據(jù)包調(diào)度程序來處理數(shù)據(jù)。

Additionally, using a Binding Indication allows integrity to be disabled, which may result in better performance.
此外，使用綁定指示允許禁用完整性，這可能會帶來更好的性能。

This is useful for large-scale endpoints, such as Public Switched Telephone Network (PSTN) gateways and Session Border Controllers (SBCs).
這對于大型端點非常有用，例如公共交換電話網(wǎng)絡(luò) (PSTN) 網(wǎng)關(guān)和會話邊界控制器 (SBC)。

B.9. Selecting Candidate Type Preference【選擇候選人類型偏好】

One criterion for selecting type and local preference values is the use of a data intermediary, such as a TURN server, a tunnel service such as a VPN server, or NAT.
選擇類型和本地首選項值的一個標(biāo)準是使用數(shù)據(jù)中介，例如 TURN 服務(wù)器、隧道服務(wù)（例如 VPN 服務(wù)器）或 NAT。

With a data intermediary, if data is sent to that candidate, it will first transit the data intermediary before being received.
對于數(shù)據(jù)中介，如果數(shù)據(jù)被發(fā)送到該候選，它將在被接收之前首先通過數(shù)據(jù)中介。

One type of candidate that involves a data intermediary is the relayed candidate. Another type is the host candidate, which is obtained from a VPN interface.
涉及數(shù)據(jù)中介的一類候選者是relayed候選。 另一種類型是host候選，它是從 VPN 接口獲得的。

When data is transited through a data intermediary, it can have a positive or negative effect on the latency between transmission and reception.
當(dāng)數(shù)據(jù)通過數(shù)據(jù)中介傳輸時，它會對傳輸和接收之間的延遲產(chǎn)生積極或消極的影響。

It may or may not increase the packet losses, because of the additional router hops that may be taken.
它可能會或可能不會增加數(shù)據(jù)包丟失，因為可能會采用額外的路由器躍點。

It may increase the cost of providing service, since data will be routed in and right back out of a data intermediary run by a provider.
它可能會增加提供服務(wù)的成本，因為數(shù)據(jù)將被路由進出由提供商運行的數(shù)據(jù)中介。

If these concerns are important, the type preference for relayed candidates needs to be carefully chosen.
如果這些問題很重要，則需要仔細選擇relayed候選的類型偏好。

Another criterion for selecting preferences is the IP address family. ICE works with both IPv4 and IPv6.
選擇首選項的另一個標(biāo)準是 IP 地址系列。 ICE 適用于 IPv4 和 IPv6。

It provides a transition mechanism that allows dual-stack hosts to prefer connectivity over IPv6 but to fall back to IPv4 in case the v6 networks are disconnected.
它提供了一種轉(zhuǎn)換機制，允許雙棧主機優(yōu)先選擇連接而不是 IPv6，但在 v6 網(wǎng)絡(luò)斷開連接的情況下回退到 IPv4。

Implementation SHOULD follow the guidelines from [RFC8421] to avoid excessive delays in the connectivity-check phase if broken paths exist.
實現(xiàn)應(yīng)該遵循 [RFC8421] 的指導(dǎo)方針，以避免在存在損壞路徑的情況下在連接檢查階段出現(xiàn)過度延遲。

Another criterion for selecting preferences is topological awareness. This is beneficial for candidates that make use of intermediaries.
選擇偏好的另一個標(biāo)準是拓撲意識。這對使用中介的候選是有益的。

In those cases, if an ICE agent has preconfigured or dynamically discovered knowledge of the topological proximity of the intermediaries to itself, it can use that to assign higher local preferences to candidates obtained from closer intermediaries.
在這些情況下，如果 ICE 代理已經(jīng)預(yù)先配置或動態(tài)發(fā)現(xiàn)了中介與其自身的拓撲鄰近性的知識，它可以使用它來為從更接近的中介獲得的候選分配更高的本地偏好。

Another criterion for selecting preferences might be security or privacy.
選擇偏好的另一個標(biāo)準可能是安全性或隱私性。

If a user is a telecommuter, and therefore connected to a corporate network and a local home network, the user may prefer their voice traffic to be routed over the VPN or similar tunnel in order to keep it on the corporate network when communicating within the enterprise but may use the local network when communicating with users outside of the enterprise.
如果用戶是因為遠程辦公，才連接到公司網(wǎng)絡(luò)和本地家庭網(wǎng)絡(luò)，則用戶可能更喜歡通過 VPN 或類似隧道路由他們的語音流量，以便在企業(yè)內(nèi)部通信時將其保留在公司網(wǎng)絡(luò)上但在與企業(yè)外部的用戶通信時可能會使用本地網(wǎng)絡(luò)。

In such a case, a VPN address would have a higher local preference than any other address.
在這種情況下，VPN 地址將比任何其他地址具有更高的本地優(yōu)先級。

總結(jié)

以上是生活随笔為你收集整理的附录B. Design Motivations【设计意图】的全部內(nèi)容，希望文章能夠幫你解決所遇到的問題。

如果覺得生活随笔網(wǎng)站內(nèi)容還不錯，歡迎將生活随笔推薦給好友。