SAMSON Network Memory Server Project

Overview A network memory server (NMS) is a device that provides clients with access to a large amount of RAM via a fast network memory paging service, in analogy to the way in which a network file server provides clients with access to a large amount of disk storage. The ready availability of high-speed (>1Gb/sec), low-latency (<10us) networking hardware now makes it feasible to use commodity components to construct a network memory server that can offer a memory paging service two orders of magnitude faster than paging to and from local disks. Such a server could provide a seemingly "infinite" memory resource for memory-hungry applications on client systems, in such a way that the client applications can execute nearly as fast using the network memory server as they would if all the memory were local to the client. SAMSON stands for "Scalable Active Memory Server on a Network". This project, which has been funded by NSF grant NSF-CISE EIA-98-18342 (Prof. Tzi-cker Chiueh, PI, and eight co-PIs), has as its objective the construction of a prototype network memory server using a cluster of commodity PC-class systems. A block diagram of the system can be found here. News Flash (June, 2003)! Professor Stark has implemented the event-logging mechanism that was part of the original design. It is now possible to make some cool graphs showing detailed system performance. Also see here for a report on the performance of the "Visible Human" application running on the system.

Current Status (Summer 2003)

The system development is now at a stage where a variety of projects are possible. Some of these projects would involve kernel coding and some could be performed entirely at the application level.

• Investigate "active" paging strategies, which attempt to hide latency by predicting the pages that will be required by the client and delivering these pages "just-in-time". Measurements from the "Visible Human" application are encouraging here.
• Extend the "event stream" mechanism so that clients can communicate event information to servers over the network. The availability of such information may be useful in predicting the applications page requirements.
• Port additional interesting applications to the system and investigate their paging characteristics.
• Implement some of the remaining mechanisms for failure tolerance and resource recovery that were in the original design. In particular, the TCP links between client-side and server-side daemons remain to be implemented, and these are important to making the system operate in more of a "production" mode.
• Attempt to improve system performance by streamlining the latency-critical path and rewriting the current pageout code so that replies to pageout operations on the client can be handled asynchronously.
• Port the system to FreeBSD (a great project for somebody who really wants to get real hands-on kernel programming experience!).

Students seriously interested in working on any of these projects are invited to contact Professor Stark.

Project Members

Main developer (Spring 1999 - Summer 2003):

Prof. Stark

Former project member (Fall 2002):

Harikesavan Krishnan

Former developer (Summer 2000 to Fall 2001):

Eric Nuzzi

Spring 1999 Design Group (course CSE 679):

Prof. Stark, Kartik Gopalan, Tulika Mitra, Guizhen Yang, Ganesh Venkitachalam, Ashish Raniwala, Susan Frank.

Links to Detailed Information

• Project history.
• Design documentation (Spring, 1999). Somewhat outdated, but a reasonable description of "the big picture".
• Implementation notes (January, 2001). Notes on how the actual implementation compares to the 1999 design document.
• SAMSON source code CVS repository Up-to-the-minute source code (.sunysb.edu access only).

Other stuff that is now here (mostly useful for current and prospective project members):

• The Linux Kernel document
• Linux HOWTO's
• Alpha-related documents and binaries.
• Myrinet-related documents, binaries, and source code (.sunysb.edu access only).