The TIME-WAIT state in TCP and its effect on busy servers

  title={The TIME-WAIT state in TCP and its effect on busy servers},
  author={Theodore Faber and Joseph D. Touch and Wei Yue},
  journal={IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320)},
  pages={1573-1583 vol.3}
  • T. Faber, J. Touch, Wei Yue
  • Published 21 March 1999
  • Computer Science
  • IEEE INFOCOM '99. Conference on Computer Communications. Proceedings. Eighteenth Annual Joint Conference of the IEEE Computer and Communications Societies. The Future is Now (Cat. No.99CH36320)
Hosts providing important network services such as HTTP and FTP incur a per-connection memory load from TCP that can adversely affect their connection rate and throughput. [] Key Method These approaches control server memory load better with growing client load than per-transaction techniques such as persistent HTTP connections. These approaches also interoperate with persistent connections to take advantage of their other benefits. This paper describes the causes of the memory loading, called TIME-WAIT…

Figures and Tables from this paper

TCP Connection Management Mechanisms for Improving Internet Server Performance

  • Amol ShuklaTim Brecht
  • Computer Science
    2006 1st IEEE Workshop on Hot Topics in Web Systems and Technologies
  • 2006
This paper investigates TCP connection management mechanisms in order to understand the behaviour and improve the performance of Internet servers during overload conditions such as flash crowds and demonstrates that connection establishment mechanisms that eliminate the TCP-level retransmission of connection attempts by clients can increase server throughput and reduce client response times under overload.

Connection-less TCP

This paper implements methods for activating and deactivating TCP connections that allow us to maintain a large store of open TCP connections without a large amount of storage overhead.

Prefetching the means for document transfer: a new approach for reducing Web latency

  • E. CohenHaim Kaplan
  • Computer Science
    Proceedings IEEE INFOCOM 2000. Conference on Computer Communications. Nineteenth Annual Joint Conference of the IEEE Computer and Communications Societies (Cat. No.00CH37064)
  • 2000
This work proposes scalable deployment solutions to control the potential overhead to proxies and particularly to Web servers, and proposes simple techniques that address these factors: pre-resolving host-names (pre-performing DNS lookup), pre-connecting (prefetching TCP connections prior to issuance of HTTP request), and pre-warming.

TCP Implementation Enhancements for Improving Webserver Performance

Lack of scalability with respect to high TCP connection rates reduces the throughput of Web servers by up to 25% and imposes a memory overhead of up to 32 MB on the kernel.

Using bundles for Web content delivery

Improving Multiple File Transfers Using

FTP over SCTP avoids the identified overheads in the current FTP protocol without introducing complexity at the application, while still remaining “TCP-friendly”.

Minimal TCP/IP implementation with proxy support

A generic mechanism for offloading the TCP/IP stack in a small device using an intermediate agent known as a proxy, which works at the transport level and keeps some of the end to end semantics of TCP.

Connection and performance model driven optimization of pageview response time

  • Dinesh KumarD. OlshefskiLi Zhang
  • Computer Science
    2009 IEEE International Symposium on Modeling, Analysis & Simulation of Computer and Telecommunication Systems
  • 2009
This work presents Connection and Performance Model Driven Optimization (CP-MDO), a novel approach for providing optimal QoS as defined by a cost objective based on client perceived pageview response time and pageview drop rate.

N for the price of 1: bundling web objects for more efficient content delivery

If embedded objects on a Web page are delivered as a single bundle, the response time experienced by the clients is as good as or better than that provided by currently deployed mechanisms, and the approach reduces the load on the network and servers.

Improving file transfers using SCTP multistreaming

  • S. LadhaP. Amer
  • Computer Science
    IEEE International Conference on Performance, Computing, and Communications, 2004
  • 2004
FTP over SCTP avoids the identified overheads in the current FTP protocol without introducing complexity at the application, while still remaining "TCP-friendly".



Network performance effects of HTTP/1.1, CSS1, and PNG

The investigation of the effect of persistent connections, pipelining and link level document compression on client and server HTTP implementations confirms that HTTP/1.1 is meeting its major design goals and further performance and network savings enabled by the improved caching facilities provided by the HTTP/ 1.1 protocol are investigated.

Efficient demultiplexing of incoming TCP packets

A simple analytic approach is used to examine how different PCB lookup schemes perform with OLTP traffic, and one scheme is shown to work an order of magnitude better forOLTP traffic than the one-PCB cache approach while still maintaining good performance for packet-train traffic.

TCP Extensions for High Performance

This memo presents a set of TCP extensions to improve performance over large bandwidth*delay product paths and to provide reliable operation over very high-speed paths. It defines new TCP options for

TCP/IP Illustrated, Volume 1: The Protocols

TCP/IP Illustrated, Volume 1 is a complete and detailed guide to the entire TCP/IP protocol suite - with an important difference from other books on the subject. Rather than just describing what the

DOD standard transmission control protocol

Abstract : The Transmission Control Protocol (TCP) is intended for use as a highly reliable host-to-host protocol between hosts in packet-switched computer communication networks, and especially in

Internet Protocol

Along with TCP, IP represents the heart of the Internet protocols and has two primary responsibilities: providing connectionless, best-effort delivery of datagrams through an internetwork; and providing fragmentation and reassembly of data links to support data links with different maximum transmission unit (MTU) sizes.

The Apache HTTP Server Project

This collaborative software development effort has created a robust, feature-rich HTTP server software package that currently dominates the public Internet market and is more often attributed to performance than price.

TIME-WAIT Assassination Hazards in TCP

This note describes some theoretically-possible failure modes for TCP connections and discusses possible remedies. In particular, one very simple fix is identified.

Myrinet: A Gigabit-per-Second Local Area Network

The Myrinet local area network employs the same technology used for packet communication and switching within massively parallel processors, but with the highest performance per unit cost of any current LAN.

Hypertext Transport Protocol HTTP/1.1

Machine tools with a horizontally sliding tool-holding ram in which a force is applied to the anterior end of the ram by a bar which moves with the ram, an upwardly directed force for balancing the weight of the Ram and of the tooling which it carries is described.