[컴퓨터 통신] L4 Transport Layer_2

찌헨·2024년 9월 6일

CDA

목록 보기
5/6
post-thumbnail

No jam



Transport Layer Protocols

  • Simple Protocol
  • Stop-and-Wait Protocol
  • Go-Back-N Protocol
  • Selective-Repeat Protpcol
    -Bidirectional Protocols: Piggybacking

Simple Protocol

  • Simple and Connectionless protocol
  • Flow, Error control 없음!!
  • Receiver can immediately handle any frame it receives.

FSM for the simple protocol

  • Sender sends packets without thinking about the receiver

Stop-and-Wait Protocol

  • Connection-oriented protocol

  • Sliding window size: 1

  • Flow control is achieved

    • By forcing the sender to wait for an acknowledgment
    • 일반적인 Flow control 방법: Buffer
    • Stop-and-Wait의 buffer size: 1
  • Error control is achieved

    • By discarding corrupted packets and letting the sender resend unacknowledged packets when the timer expires
    • Error control을 위해 사용하는 번호: Sequence Number
    • Stop-and-Wait에서는 1 bit field를 sequence num 저장하기 위해 사용
      • m=1 → sequence numbers: modulo 2 (0 or 1)
      • the ack num always announces in modulo-2 the seq num of the next packet expected
  • All calculation in the Stop-and Wait protocol is in modulo 2
    FSM for the stop-and-wait protocol

  • Example

  • Inefficient

    • Lost Data

    • Lost Ack

    • Delayed Ack/Data

  • Bandwidth-delay product

    • Channel: pipe
    • Bandwidth-delay product: pipe capacity
    • e.g.) bandwidth: 1Mbps, delay: 20msec 일때 Bandwidth-delay product: (110^6)(20*10^-3) = 20,000 bit
  • Example

    • In a Stop-and-Wait system, the bandwidth of the line is 1Mbps, and 1 bit takes 20msec to make a round trip. What is the bandwidth-delay product? If the system sends one packet (1,000 bits in length), what is the utilization percentage of the link?
      1,000 bits / 20,000 bits = 5% (Too low!)
    • What is the maximal utilization percentage of the above link if we have a protocol that can send up to 15 packets before stopping and worrying about the acknowledgments?
      15,000 bits / 20,000 bits = 75%
  • So, sneding up to multiple packets are more efficient!




글로벌소프트웨어캠퍼스와 교보DTS가 함께 진행하는 챌린지입니다.

0개의 댓글