Linux Scalability Project Status report for November and December, 1999

Linux Scalability Project

Status report for November and December, 1999

Center for Information Technology Integration
School of Information at the University of Michigan

The primary goal of this research is to improve the scalability and robustness of the Linux operating system to support greater network server workloads more reliably. We are specifically interested in single-system scalability, performance, and reliability of network server infrastructure products running on Linux, such as LDAP directory servers, IMAP electronic mail servers, and web servers, among others.

Summary

We're working with vendors such as Sun and IBM on their Linux scalability issues. We've provided several significant modifications to the Linux kernel. We've submitted papers for the Usenix Technical Conference to occur in the summer of 2000. We're continuing to reach out to potential sponsors. Work continues on long-term projects.

Milestones

Niels and Chuck have submitted a paper abstract to the Usenix Technical Conference on their work with enhancements to poll() and kernel file descriptor management. The paper will also analyze the performance trade-offs between poll() and using POSIX RT signals to support asynchronous I/O.
The madvise() system call is close to being incorporated in the current Linux development kernel. Chuck continues to work with kernel developers on improving the patch before it is incorporated.
Steve continues to explore generic solutions to thundering herd issues in the Linux TCP stack. His accept() paper has been submitted to the Usenix Technical Conference for the summer of 2000.
Chuck submitted his paper on malloc() performance to the Usenix Technical Conference for the summer of 2000.
See the NFSv4 project site for project status on Andy's NFSv4 implementation.
IBM has contacted the Linux Scalability Project regarding work on thread performance in Java virtual Machines. We hope to cement a research relationship with IBM Austin via their CAS graduate research program within the next two months.
Dell Computer is a new sponsor of the Linux Scalability Project. Dell has granted the LSP a US$20,000 grant via the Dell STAR higher education grant program to continue our work in Linux scalability issues.
Peter wrote a paper describing the Linux Scalability Project for the upcoming NLUUG conference in Amsterdam, entitled "Linux en Open Source." He spent a week there in November presenting the paper and discussing funding with potential sponsors, as well as discussing the project with principal Linux developers. The paper is available as CITI Tech Report 99-4.
A directory server and certificate management system is being deployed at CITI to enable research on public key infrastructure and related technologies. Based on Netscape products, this infrastructure will help CITI pursue related areas of research, such as:
- managing large clusters of Linux batch computers via GRID
- transitioning the U-M security infrastructure to PKI-friendly technologies
- integrating CITI's SmartCard work into an operational PKI
We are continuing to pursue research relationships with Red Hat Software, Silicon Graphics Incorporated, and the Stichting NLNet Foundation.

Challenges

The "gift economy" is one of the basic tenets of open source software development. Andrew Leonard has published an article [http://www.salon.com/tech/feature/1999/12/23/linux_ipos/] in Salon magazine examining some of the effects of personal fortunes on the gift economy.

If you have comments or suggestions, email linux-scalability @ citi.umich.edu