blank.gif (43 bytes)

Church Of The
Swimming Elephant

Search:
What is... Packets/MTU?
Packet/MTU Theory.
3-30-01

Page 3

By John Holstein, Cotse Help Desk Coordinator

Throughput vs. Packet Size

Using this formula:

( MSS + header ) * 8 bits/byte
------------------------------ = latency (per hop)
1,544,000 bits/sec.

Assume we need to transfer 1 MByte file:

1MByte = 1024 KB = 1,048,576 bytes.

If MTU = 1500, then: (1460+40) * 8 / 1,544,000 = 7.772 ms delay per hop

1 MByte / MSS = 1,048,576 bytes / 1460 = 718.2, or effectively 719 packets to transfer 1 MByte.

Then, to transfer 1Mbyte: 719 packets * 7.772 ms delay per hop = 5588.068 ms, or 5.588 seconds per hop.

If we are transferring our 1 MByte file over 10 hops, it will take us 55.88 sec.

If MTU = 576, then: (536+40) * 8 / 1,544,000 = 2.924 ms delay per hop.

1 MByte / MSS = 1,048,576 bytes / 536 = 1956.3, or effectively 1957 packets to transfer 1 MByte.

Then, to transfer 1 MByte: 1957 packets * 2.924 ms delay per hop = 5722.268 ms, or 5.722 seconds per hop.

If we are transferring our 1 MByte file over the same 10 hops, it will take us 57.22 sec.

The difference comes from the fact that when using larger packets the overhead is smaller. To transfer 1 MByte, if using MTU of 1500 there are 719 * 40 = 28,760 bytes of overhead, while if using MTU of 576 1957 * 40 = 78,280 bytes, additional 49,520 bytes of headers transferred each MByte. For our 10-hop transfer, the additional overhead accounts for 1.34 seconds difference in transfer time for every MByte. This difference is a bit higher in practice, considering TCP options and the fact that modern TCP/IP implementations tend to use larger headers (additional 12 bytes header space for Timestamps for example).

Summary

It's logical to assume bigger packets are better, because of all the following factors:

  • network - reduced number of headers, as illustrated above
  • routers - less routing decisions
  • clients - less protocol processing and device interrupts

    If throughput is not the goal, smaller packets may be a better selection since they take less time to travel throughout the network. That effect might be preferred in some applications and online gaming, at the expense of throughput.

    Ultimately, packet size should be decided based on the type of the desired result, considering the underlying network as well, to avoid negative factors such as fragmentation of packets. Still one has to realize the fact that larger packets will still transmit more useful data than smaller packets, and that there is no single "best" solution for all applications.

    Appendix

    Everything you need to know about Networks

    Registry Changes to Windows 9x & ME in Relation to TCP/IP

    Description of the Internet Protocol Packet Size Setting


    Comments? Questions? Bugs? John Holstein

    Return to the Help Desk


  • Cotse.Net

    Protect yourself from cyberstalkers, identity thieves, and those who would snoop on you.
    Stop spam from invading your inbox without losing the mail you want. We give you more control over your e-mail than any other service.
    Block popups, ads, and malicious scripts while you surf the net through our anonymous proxies.
    Participate in Usenet, host your web files, easily send anonymous messages, and more, much more.
    All private, all encrypted, all secure, all in an easy to use service, and all for only $5.95 a month!

    Service Details

     
    .
    www.cotse.com
    Have you gone to church today?
    .
    All pages ©1999, 2000, 2001, 2002, 2003 Church of the Swimming Elephant unless otherwise stated
    Church of the Swimming Elephant©1999, 2000, 2001, 2002, 2003 Cotse.com.
    Cotse.com is a wholly owned subsidiary of Packetderm, LLC.

    Packetderm, LLC
    210 Park Ave #308
    Worcester, MA 01609