roussopoulos-to-or-not - 1 2 P2P or Not 2 P2P? Mema...

Info iconThis preview shows pages 1–2. Sign up to view the full content.

View Full Document Right Arrow Icon

Info iconThis preview has intentionally blurred sections. Sign up to view the full version.

View Full DocumentRight Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: 1 2 P2P or Not 2 P2P? Mema Roussopoulos Harvard University, Cambridge, MA TJ Giuli Stanford University, Stanford, CA Mary Baker HP Labs, Palo Alto, CA Petros Maniatis Intel Research, Berkeley, CA David S. H. Rosenthal Stanford University Libraries, CA Jeff Mogul HP Labs, Palo Alto, CA Abstract — In the hope of stimulating discussion, we present a heuristic decision tree that designers can use to judge how suitable a P2P solution might be for a particu- lar problem. It is based on characteristics of a wide range of P2P systems from the literature, both proposed and de- ployed. These include budget, resource relevance, trust, rate of system change, and criticality. 1. INTRODUCTION Academic research in peer-to-peer (P2P) systems has concentrated largely on algorithms to improve the effi- ciency [31], scalability [22], robustness [12], and secu- rity [33] of query routing in P2P systems, services such as indexing and search [20], dissemination [17], and ren- dezvous [27] [30] for applications running on top of these systems, or even many of the above [18]. While these improvements may be essential to enhancing the perfor- mance of some P2P applications, there has been little fo- cus on what makes a problem “P2P-worthy,” or on which other, previously ignored problems may benefit from the application of P2P techniques. What questions should a system designer ask to judge whether a P2P solution is appropriate for his particular problem? In this position paper, we hope to stimulate discussion by distilling the experience of a broad range of proposed and deployed P2P systems into a methodology for judg- ing how suitable a P2P architecture might be for a partic- ular problem. In Section 2, we identify some salient char- acteristics axes in typical distributed problems. In Sec- tion 3, we describe a spectrum of specific problems for which P2P solutions have been proposed. In Section 4, we propose an arrangement of problem characteristics into a heuristic decision tree. We walk through the tree explain- ing its choices and why we believe certain paths may lead to successful P2P solutions to important problems, while other paths may encounter difficulties. While any particu- lar set of characteristics axes or fixed decision graph may be inadequate for all purposes, we present the arrange- ment that has proved most useful in our work so far. 2. PROBLEM CHARACTERISTICS AXES In this section, we describe the characteristics we be- lieve are important in assessing the P2P-worthiness of dis- tributed problems. Paraphrasing the call for papers of this workshop, we identify as peer-to-peer those environments that satisfy the following three criteria: Self-organizing : Nodes organize themselves into a network through a discovery process. There is no global directory of peers or resources....
View Full Document

This note was uploaded on 12/08/2011 for the course CS 525 taught by Professor Gupta during the Spring '08 term at University of Illinois, Urbana Champaign.

Page1 / 6

roussopoulos-to-or-not - 1 2 P2P or Not 2 P2P? Mema...

This preview shows document pages 1 - 2. Sign up to view the full document.

View Full Document Right Arrow Icon
Ask a homework question - tutors are online