 |
Taiga: Internet-scale computing
The concept of a program is changing from a local, self-contained object into an Internet-scale, pervasive, self-organizing, omnipresent entity. We have build a prototype framework that addresses the programming issues that will make this transition smooth and manageable. There are a number of current trends that point to the future of programs and programming. These include web services, grid computing, peer-to-peer computing, autonomic software [8], the open source movement, and faster, more-reliable networking. These trends are converging. Programs are written using a multitude of web services some of which use other web services. These web services are written by different, often anonymous programmers and change with little or no notice. Grid-based applications such as today's database systems run on any available machine and find new computational resources as needed. Data is shared across the Internet both through web services and through peer-to-peer connections. Instant messaging, SETI, Gnutella [6], Napster, and electric sheep [7] are examples of current Internet-scale applications, with more to come.
Dealing with this new reality will require us to change the way we think about programs and programming. We can't continue to think of a program as a self-contained entity with only local effects that we control. Instead, we need to work in terms of a global system where we do not control most of the components or interactions. We need a means for programming in a world where software systems will be built mainly from components designed, developed, maintained, and modified by different people, on machines that are not under the control of the software developers or designers, and where the components themselves evolve outside of the control of the software system. It is this set of problems that our research tries to address. Our goal is to ensure: Developing an Internet-scale application should be no more difficult in the future than developing a standalone application is today.
Project status: Active
Project Home Page: http://www.cs.brown.edu/people/spr/research/taiga.html
Research Areas
| Software Engineering |
People
| Steven Reiss |
Publications
Reiss, S. P. A component model for Internet-scale applications. In Proceedings of the 20th IEEE/ACM International Conference on Automated Software Engineering (ASE 2005) (Nov 2005), pp. 34-43. [ pdf ]
Reiss, S. P. Evolving evolution. In Proceedings of the 8th International Workshop on the Principles of Software Evolution (Sep 2005), pp. 136-139. [ pdf ]
| Page Owner: Webmaster | Last Modified: Mon Oct 23 14:57:09 2006 |