| Stevens, W. R. (1996) TCP/IP Illustrated, vol. 3, TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley, Reading, MA. |
....implementation will attach di erent degrees of importance to di erent objectives in di erent environments. There are several paradigms and associated protocols which might o er bases for adding network programming into database programming. The rst alternative is TCP data transmission protocol [Ste96, Ste94] But the user or programmer must be uent with a set of protocols or messages for various data communications if this paradigm is adopted for adding network programming into database programming. However, this can be avoided and we can make the protocols totally transparent in a high ....
W. Richard Stevens. TCP/IP Illustrated, Volume 3 , TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley, 1996.
....setup and four more for connection termination. This requirement is above three other segments needed to reliably send the data: one for sending the request, one for sending the reply along with an ACK for the request, and one to finally ACK the reply. One solution, described by Stevens in [Ste96] is called TCP for Transactions (T TCP) The T TCP extension mitigates the mentioned overhead by adopting an accelerated 23 open strategy. T TCP assigns a 32 bit incrementing connection count to newly opened connections in the form of a TCP option (CC) sent out with every segment. A host ....
....bursts to the network, which usually results from the ARQ process. STP also supports segment type overloading. Certain type combinations are allowed to exist in the same segment. The protocol uses this feature to support a fast connection start mechanism similar to the one proposed in T TCP [Ste96] In this case, the data is allowed to be transmitted along with the BGN segment in anticipation of a connection acceptance by the targeted host. This process is done by overloading the segment s type to enable both the BGN and the SD flags. Both the POLL and the END flags can also be enabled, ....
W. Stevens. TCP/IP Illustrated, Volume III; TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley, 1996.
....implementation of the accelerator, which decreases the web server load by more than a factor of three. Keywords: web server, load balancing, web booster. 1 Introduction Traffic to a web server domain is unpredictable and very peaky. Request rate peaks can be up to ten times the average rate [Ste96]. To avoid losing or alienating clients, web hosting sites are overbuilt to handle the peak load. While some of the sites are built as web server clusters that host multiple web domains, many of the sites require each domain to reside on dedicated servers for security or other reasons. This paper ....
....request has to be delivered to the user level HTTP server, the accelerator writes an unmodified request to a separate socket that is used by user level the HTTP server to read requests. The MTU of clients can vary greatly. For example, according to the study of the web server under real workload [Ste96], only 32 of all clients advertised MSS value of 1460, which corresponds to Ethernet MTU, and 60 of them advertised a value less than 1024. The packet caching technique only can be used within a range of the MTU s that are used for connection to the web booster or clients. If the size of the ....
W. R. Stevens. TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NTTP, and the UNIX Domain Protocols. Addison-Wesley, 1996.
....to those above, and does not serve to replace them, since it only detects measurement drops at the packet sender. For further discussion, including why it does not always indicate such, see x 10.5 and x 13.2. 3 The T TCP extension to TCP allows data to be sent prior to full establishment [Br94, St96]. None of the TCPs in our study used T TCP, however. 134 10.3.4 Trace truncation Related to packet filter drops but slightly different is the problem of trace truncation. Truncation occurs when the filter misses the packets belonging to either the beginning of a connection or the end. Both ....
....The fix is replacing a # test with a # test. Of these problems, we found that a number of the implementations in our study exhibited all of them, except we did not examine the RTO s used by the implementations and thus did not have an opportunity to observe the second problem. Stevens In [St96], Stevens devotes a chapter to an analysis of the behavior of a large number of TCP connections made to a World Wide Web server running Net 3 TCP. The analysis was based on a 24 hour tcpdump packet trace of 147,103 attempts by remote sites to connect to the Web server. He characterized the range ....
W.R. Stevens, TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison-Wesley, 1996.
....Mosaic web browser. Barford et al. 1] use HTTP proxies to track all documents referenced by unmodi ed Netscape Naviga tor clients. The third approach of gathering data, and the method used here, is to analyze packet traces taken from a subnet carrying HTTP trac. This method was used by Stevens [15] to analyze the trac arriving at a server, and by Mah [11] to model the client side of the HTTP trac. A further step is taken by Anja Feldman [7] when extracting full HTTP level as well as TCP level traces via packet monitoring. VI. Conclusions and future work We have presented an empirical ....
W. R. Stevens, TCP/IP Illustrated, Vol. 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, AddisonWesley, Reading, 1996.
....under test (circles) Http caused a burst well above the 64k CIR, but finished quickly. During these tests the time between a burst of data and the associated acknowledgment was usually between 170 and 350 ms, while the same tests on the LAN gave times between 1 and 8 ms. References [Stev94] [Stev96] discuss some of the more subtle effects of RTT on network performance such as its effect on TCP window size, timeout, and retransmission, but our simple packet trace analysis made it apparent that RTT was a critical network performance parameter for our client server applications. We also ....
R. Stevens, TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison-Wesley, 1996.
.... in packets transferred compared to TCP (obviously in cases where a large amount of data is being transferred, there will be more packets transmitted and thus a decrease in the percentage saving) Timing experiments have shown that there is a slightly longer time required for T TCP than for UDP [18], but this is a result of the speed of the computer, and not the network. As computers get more powerful, the performance of T TCP will approach that of UDP. 4.4 TCP Accelerated Open The TCP Accelerated Open (TAO) 3] is a mechanism introduced by T TCP that is designed to cut down the number of ....
....the first FIN is required to remain in the TIME WAIT state for twice the MSL once the connection is completely closed at both ends. This implies that the TIME WAIT state with the original TCP is 240 seconds, even though some implementations of TCP have the TIME WAIT set to 60 seconds. Stevens [18] shows how the TIME WAIT state for T TCP may be shortened to 12 seconds. CC options do have problems when used on networks with high speed connections. This is rarely a problem on older networks, but with FDDI and Gigabit Ethernets becoming more mainstream, the wrapping of the CC value will ....
[Article contains additional citation context not shown here]
Stevens W R, TCP/IP Illustrated, Volume 3, TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison-Wesley, 1996
.... Internet uses a unified set of protocols, in reality each protocol has been implemented by many different communities, often with significantly different features (and of course bugs) For example, the widely used Transmission Control Protocol (TCP) has undergone major evolutionary changes (see [Ste96] for a family tree showing the lineages associated with one evolutionary branch) A study of eleven different TCP implementations found distinguishing differences among nearly all of them [Pax97a] and major problems with several [PADR99] More recently, techniques for fingerprinting ....
Stevens, W. R. 1996. TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley.
....MSL positively replies the request with INI RSP. The 16 bit index V id in INI RSP is assigned for later reconnection. Typically five TCP segments are exchanged to establish a virtual connection. But we can combine INI REQ with SYN, since it is legal under TCP protocol with most TCP implementation[21] The INI REQ and INI RSP messages are used internally in MSL and will not be delivered to the upper layers. If the server system rejects this virtual connection, it simply closes the real TCP connection. The client system will know the request is rejected and all internal data structures will be ....
W. Richard Stevens. TCP/IP Illustrated, Volume 3, TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley, 1996.
....not yield much information about the paths of messages. This makes it difficult to use traffic analysis to determine who is communicating with whom. Onion Routing provides an anonymous socket connection through a proxy server. Since proxies are a well defined interface at the application layer [12, 11], and many protocols have been adapted to work with proxy servers in order to accommodate firewalls, Onion Routing can be easily used by many applications. Our prototype works with HTTP (World Wide Web) proxies. In addition, a proxy for TELNET has been implemented. Traffic analysis can be used to ....
W. R. Stevens. TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison--Wesley, Reading, Mass., 1996.
....the offered window advances a large amount (due to an incoming ack for a large amount of new data) and a fencepost error for determining whether the congestion window needs to be shrunk after a fast recovery sequence. They also discussed fixes for the problems. Stevens devotes a chapter in [St96] to an analysis of the behavior of a large number of TCP connections made to a World Wide Web server running Net 3 TCP. He characterized the range of options offered by the remote TCPs, finding tremendous variation (including many obviously incorrect values) the rate at which connection attempts ....
....2.3 uses a much lower initial RTO, around 300 msec, than the other implementations (Comer and Lin found the same for Solaris 2.1 [CL94] and takes much longer to adapt the RTO to higher, measured round trip times (RTTs) We discuss both of these latter problems further in x 8.6. Except for [St96], these previous studies used active techniques, in which an implementation s behavior is examined by controlling the packets it receives and determining its response. All of the studies involve manual analysis. Our emphasis is on off line analysis of packet traces, a passive technique, which is ....
W.R. Stevens, TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison-Wesley, 1996.
....mechanism between servers and clients) of the Web operates. This included both the mechanics of the components of the Web and the way in which people use the Web. General descriptions of the mechanics of Web components 3 and Web operation can be found in any number of reference books such as [120]. Below the application level are the networking protocols which allow transmission of HTTP messages and data, and at the lowest level are the host systems and physical network which are the Internet. Details of how the Web operates and how people use the Web have been addressed in studies such as ....
W. Stevens. TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley, 1996.
.... Internet uses a unified set of protocols, in reality each protocol has been implemented by many different communities, often with significantly different features (and of course bugs) For example, the widely used Transmission Control Protocol (TCP) has undergone major evolutionary changes (see [Ste96] for a family tree showing the lineages associated with one evolutionary branch) A study of eleven different TCP implementations found distinguishing differences among nearly all of them [Pax97b] and major problems with several [PADR99] More recently, techniques for fingerprinting ....
W. Stevens, TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison-Wesley, 1996.
No context found.
W. Richard Stevens, TCP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols, Addison Wesley, 1996.
No context found.
Stevens, W. R. (1996) TCP/IP Illustrated, vol. 3, TCP for Transactions, HTTP, NNTP, and the UNIX Domain Protocols. Addison-Wesley, Reading, MA.
No context found.
Stevens,W.R. TCP/IP Illustrated, Volume 3: TCP for Transactions, HTTP, NNTP, and the Unix Domain Protocols, 1996.
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC