互联网工程任务组(IETF)于1999年8月发布了编号为2675的技术标准文档,题为“IPv6 Jumbograms”1。该标准由David Borman、Stephen Deering和Robert Hinden共同撰写,旨在定义一种允许有效载荷超过65,535字节的IPv6数据包传输机制,即“巨帧”1。这一技术规定仅适用于连接在链路最大传输单元(MTU)大于65,575字节的IPv6节点1。
在实现细节上,RFC 2675通过在逐跳选项头部(Hop-by-Hop Options header)中设置一个32位的“Jumbo Payload Length”字段来指定数据包的实际情况长度1。为了支持这种超大数据包,传输层协议也需进行相应调整:UDP协议将长度字段设为0以标识使用巨帧,而TCP协议则将最大段大小(MSS)值设为65,535来表示无限大1。
In August 1999, the Internet Engineering Task Force (IETF) published RFC 2675, titled "IPv6 Jumbograms," authored by David Borman, Stephen Deering, and Robert Hinden 1. This technical standard defines a mechanism for transmitting IPv6 packets with payloads larger than the conventional 65,535-byte limit 1. The document specifies that such "Jumbograms" are only applicable to IPv6 nodes connected to links where the Maximum Transmission Unit (MTU) exceeds 65,575 bytes 1.
The implementation of jumbograms relies on modifying existing protocol headers to accommodate the extended payload size. Specifically, the actual packet length is indicated via a 32-bit "Jumbo Payload Length" field located within the Hop-by-Hop Options header 1. To ensure compatibility with transport layer protocols, the RFC mandates specific adjustments: for UDP, the length field must be set to zero to signal the use of a jumbogram, while for TCP, the Maximum Segment Size (MSS) value should be configured as 65,535 to represent an infinite size in this context 1.
评论
还没有评论,欢迎留下第一条。