A stateless approach to connection-oriented protocols

Abstract

Traditional operating system interfaces and network protocol implementations force some system state to be kept on both sides of a connection. This state ties the connection to its endpoints, impedes transparent failover, permits denial-of-service attacks, and limits scalability. This article introduces a novel TCP-like transport protocol and a new interface to replace sockets that together enable all state to be kept on one endpoint, allowing the other endpoint, typically the server, to operate without any per-connection state. Called Trickles, this approach enables servers to scale well with increasing numbers of clients, consume fewer resources, and better resist denial-of-service attacks. Measurements on a full implementation in Linux indicate that Trickles achieves performance comparable to TCP/IP, interacts well with other flows, and scales well. Trickles also enables qualitatively different kinds of networked services. Services can be geographically replicated and contacted through an anycast primitive for improved availability and performance. Widely-deployed practices that currently have client-observable side effects, such as periodic server reboots, connection redirection, and failover, can be made transparent, and perform well, under Trickles. The protocol is secure against tampering and replay attacks, and the client interface is backward-compatible, requiring no changes to sockets-based client applications.

A stateless transport protocol in software defined networks

“…After a thorough analysis of the paper [6] we have implemented the original model of the Trickles protocol. This model was used for constructing the network model, which is shown in Fig.…”

“…The Trickles protocol was proposed in 2005 at Cornell University, USA [6].The main difference of Trickles protocol from TCP protocols is keeping of all of control parameters on the client side, while the backend does not store information about the transport connection. Transport protocol operating on this principle from server to client will be called asymmetric protocol or unallocated connection state.…”

“…In the slow start / congestion avoidance, Trickles server [6] associates with each packet its request number k, and decides whether to continue the flow or separate it using TCPCwnd (k) function.…”

“…It is evident that the class of asymmetric protocols can facilitate the operation of heavily loaded server that was shown in [6]. However, this work is not a comprehensive analysis of Trickles protocol performance, which is an important task in the development of any transport protocol.…”

Theory and Practice of Bloom Filters for Distributed Systems

“…Two Bloom filters of identical size and using the same family of hash functions are used to simplify the periodic purge operation [113]. The counting variant (CBF) is used in [114] to provide a lightweight route verification mechanism that enables a router to discover route failures and inconsistencies between advertised Internet routes and the actual paths taken by the data.…”