"Submitted to:Submitted by:Under Supervision of:1 INDEX? Introduction…………………………………………….3? Proof-of-concept……………………………………….5? Definition……………………………………………...6? Architecture…………………………………………....8? Implementation using TCP…………….………………9? Problem solved using parasitic computing.…………...12? Issues in parasitic computing………….………………17? Advantages and disadvantages………………………..20? Conclusion…………………………………………….21? References……………………………………………..22 2 INTRODUCTION Parasitic computing has drawn a lot of attention in the Internet community since the firstdemonstration of its potential.A search of the term “parasitic computing” on Google returns 544results, with coverage on several major portal website such as CNN and BBC. Much attention was drawn to the concept of parasitic computing because it is capable ofmaking use of computational resources from all the computers connected to the Internet andcomplying with the transmission control protocol (TCP) [2]. Since most, if not all, of thecomputers on the Internet need to use the standard TCP protocol to communicate, the resourceavailable to a parasiting user appears unlimited and every computer could be exploited. Thisimposes a serious problem for the servers on the Internet, as they can possibly allocate all thecomputational capability to the parasiting users and thus affect the access effort from a normalclient similar to that in the denial of access attack. As the access of normal user is vital to theexistence of a commercial website, it is important, if not crucial; to design effective mechanismin the communication protocol stacks to identify the parasitic users among the many normalusers. Moreover, the lack of detection tools makes ordinary Internet users stressed and uneasy(“Is my computer parasited?”). In this project, we design three classes of parasitic detection schemes, namely theapplication layer detector, the transport layer detector, and the cross-layer detector. Upon thedetection of such parasitic users, the server can either blacklist from the specific IP address, orlower the priority to requests from the specific IP address. This helps to provide uncompromisedservice rates to the normal users. The rest of this report is organized as follows. In Section II, wereview the parasitic computing concept and reveal the difference between parasitic traffic andordinary traffic. Three detection methods are proposed. We implement and test the performanceof the newly proposed detection method through raw socket programming finally conclusions aredrawn. Reliable communication on the Internet is guaranteed by a standard set of protocols, usedby all computers1. Here we show that these protocols can be exploited to compute with the 3 "