1998 - 1999 CS Annual Report                                                                  Faculty
choices.gif (4488 bytes)

Thorsten von Eicken

Assistant Professor

PhD UC Berkeley, 1993

My research explores new opportunities at the intersection of
computer architecture, programming languages, and operating systems. My approach has been to identify
important trends in one area and leverage them to shift the boundaries between the components of computer systems. This follows an established tradition of experimental computer systems research widely popularized in
the RISC work which focused on the tradeoffs surrounding 

the compiler/architecture boundary. In the past, I have focused on high-performance communication in clusters of workstations. My group developed the U-Net user-level networking architecture to close the dramatic gap between the bit-rate of high-speed networks and the communication performance seen by applications. The key idea in U -Net is to virtualize the network interface, which allows each application on a multitasking computer to access the network directly without invoking the operating system. This effectively moves the protocol stack to the
user-level, where it can be coupled more tightly to the application, resulting in an order-of-magnitude reduction in communication overhead. In addition, the user-level protocols can be customized to the application, thereby enabling experimentation with new protocols that are, for example, tailored toward real-time multimedia stream transmission. 

The main ideas of U-Net have been incorporated into the VIA (Virtual Interface Architecture) industry stan
dard led by Compaq, Intel and Microsoft. At this point, commercial network interfaces designed for VIA are available. 

My group's recent research is premised on the conviction that advances in system security and safety could enable far more applications than further performance improvements. At the same time, Java has popularized the notion of safety at the language level and research projects such as
Proof Carrying Code made it clear that language-based protection technology was maturing to the point where it would force a re-evaluation of the boundaries between programming languages, operating systems, and architecture. 

As we began to investigate how to build an entire system on language-based protection, we
became convinced that a capability system is the best approach. Being able to revisit an operating system design approach essentially abandoned over a decade ago that still intrigues
many an operating system researcher is proving to be a fascinating journey. Two of the most commonly voiced reasons for the failure of capability systems are that they were too expensive to implement and that they were too difficult to use. We developed an operating system on Java, called the J-Kernel, to show how language technology can be used to implement capabilities at very low cost. While trying to extract performance from Java has been a painful undertaking, the J-Kernel does demonstrate that capabilities can be implemented very efficiently. We are now developing applications on the J-Kernel to gain experience using capabilities so we can make progress on simplifying their use. Today, the J-Kernel represents the most sophisticated attempt at building a capability system using language-based security. 

Going forward, our goal with the J-Kernel is not to replace desktop operating systems. Rather we are focusing on the emerging large numbers of networked embedded and mobile devices. We are using our experience with the J-Kernel to provide a run-time infrastructure with the network safety features required in that domain. Our goal is to show that language-based protection is the most
efficient avenue for building a run-time system for these devices. 

  • J-Kernel: A capability-base operating system for Java, secure internet programming.
    Springer Verlag, (1999) To appear. (with C. Chang, G. Czajkowski, C. Hawblitzel, D. Hu, and D. Spoonhower). 
  • Resource control for database extensions. Fifth USENIX Conference on Object Oriented
    Technologies and Systems (COOTS'99)
    (May 1999) To appear. (with G. Czajkowski, T. Mayr, and P. Seshadri). 
  • MRPC: A high performance RPC system for MPMD parallel computing. Software Practice and Experience 29, 1 (Jan. 1999) J. Wiley & Sons, Inc. (with C. Chang and G. Czajkowski). 
  • Evolution of the Virtual Interface Architecture Computer. IEEE Computer Society Press, (31)11, (Nov 1998) (with W. Vogels).  
  • JRes: A Resource Accounting Interface for Java. Proceedings of 1998 ACM OOPSLA Conference, (Oct 1998) (with G. Czajkowski). 
  • Security versus Performance Tradeoffs in RPC Implementations for Safe Language Systems. Proceedings of the 1998 ACM EuropeanWorkshop (Sept. 1998) (with C. Chang, G. Czajkowski, C. Hawblitzel, and D. Hu). 
  • Resource Management for Extensible Internet Servers. Proceedings of 1998 ACM SIGOPS European Workshop (Sept. 1998) (with G. Czajkowski, C. Chang, C. Hawblitzel, and D. Hu)
  • Retrospective on "Active Messages: A Mechanism for Integrating Computation and Communication". 25 Years of the International Symposium on Computer Architecture, Selected Papers, (G. Sohi, ed.) (1998) (with D. Culler, K. Schauser, and S. Goldstein).