Distributed Denial of Service attacks are an ever prevalent challenge for system administra-tors today to overcome. The attack, which is all about restricting legitimate users access to a service, such as a web-page. Can cost companies and governments millions of dollars if not properly managed. This study aims to explore if there is any difference in performance between some of the most modern iterations of popular server operating systems today. Those server operating systems are: Windows Server 2016, Ubuntu 16 and FreeBSD 11. And submitting them to one of the most popular DDoS attacks at the time of writing, a so called HTTP-Get request. The webservers used are some of the most widely used today, Apache and Microsoft IIS. Each server will be submitted to attacks, and compared between one another. Different de-fence methods will also be tested and examined. Tests include shorter tests that is repeated multiple times for data validity, and one longer test for every condition in order to control if the results are similar. During these tests, the operating systems will measure CPU/RAM utilization, and a control computer will measure Round Trip Time. Windows Server 2016 using IIS and FreeBSD 11 perform similarly resource wise, but Win-dows Server 2016 with IIS had a better Round Trip Time performance. Windows Server 2016 with Apache performs worst in all measurements, while Ubuntu 16 performs in the middle, but has the most stable performance.

Internetfenomenet ”Distributed Denial of Service”, förkortat DDoS, beskrivs ofta som ett av destörsta hoten mot Internet idag. Genom att utnyttja den grundläggande strukturen i kommunikationmellan nätverk och datorer kan kriminella blockera och stänga ute webbplatser och -tjänster frånanvändare, samtidigt som det är mycket svårt för offret och myndigheter att någonsin identifieraden eller de skyldiga. Enorma globala nätverk av ovetande människors infekterade datorer fjärrstyrstill att utföra angrepp mot alla sorters organisationer på Internet med olika motiv, som finansiella,politiska eller för ren vandalism. Syftet med det här arbetet är att göra en dagsaktuell kartläggningöver läget kring DDoS-angrepp och titta på statistik över de mest förekommande angreppstyperna,och se om den nyare publicerade forskningen kan svara på de pågående och framträdande trendersom kan ses. Sex forskningsartiklar väljs ut att jämföra med dessa trender för att se var merforskning krävs. Resultatet visar att forskningen kring försvar mot HTTP-GET-angrepp är bristande,samt att den framträdande trenden där angreppen allt oftare använder sig av olika angreppstypersamtidigt inte har undersökts. Mer öppen forskning bör riktas mot dessa bristande områden. / The Internet phenomenon ”Distributed Denial of Service”, in short DDoS, is often said to be one ofthe greatest threats to the Internet today. By abusing the foundation of inter-network and computercommunication, criminals can block and shut out websites and services from users while making itvery hard for the victim and the authorities to ever identify who was behind it. Enormous globalnetworks made up of unknowing peoples' infected computers can be remotely controlled to conductattacks against all sorts of organisations on the Internet with different motives, from financial orpolitic to sheer vandalism. The purpose of this study is to create an up-to-date mapping of thesituation of DdoS-attacks and look at statistics of the most prevalent attack types, and to check ifnewly published research can answer the current and emerging trends that can be seen. Six researcharticles are chosen to compare with these trends to see where more research is required. The resultsshow that the research around defense against HTTP-GET-attacks is lacking, and that the emergenttrend of DDoS-attacks that make use of more than one attack type at the same time has not beenexamined. More open research should be directed to these lacking areas.

No description available.