ALBERT

Description: This paper describes the concept for the End-to-End Data System that will support NASA's Living With a Star Space Environment Testbed missions. NASA has initiated the Living With a Star (LWS) Program to develop a better scientific understanding to address the aspects of the connected Sun-Earth system that affect life and society. A principal goal of the program is to bridge the gap.between science, engineering, and user application communities. The Space Environment Testbed (SET) Project is one element of LWS. The Project will enable future science, operational, and commercial objectives in space and atmospheric environments by improving engineering approaches to the accommodation and/or mitigation of the effects of solar variability on technological systems. The End-to-end data system allows investigators to access the SET control center, command their experiments, and receive data from their experiments back at their home facility, using the Internet. The logical functioning of major components of the end-to-end data system are described, including the GSFC Payload Operations Control Center (POCC), SET Payloads, the GSFC SET Simulation Lab, SET Experiment PI Facilities, and Host Systems. Host Spacecraft Operations Control Centers (SOCC) and the Host Spacecraft are essential links in the end-to-end data system, but are not directly under the control of the SET Project. Formal interfaces will be established between these entities and elements of the SET Project. The paper describes data flow through the system, from PI facilities connecting to the SET operations center via the Internet, communications to SET carriers and experiments via host systems, to telemetry returns to investigators from their flight experiments. It also outlines the techniques that will be used to meet mission requirements, while holding development and operational costs to a minimum. Additional information is included in the original extended abstract.

Description: In computer systems today, speed and responsiveness is often determined by network and storage subsystem performance. Faster, more scalable networking interfaces like Fibre Channel and Gigabit Ethernet provide the scaffolding from which higher performance computer systems implementations may be constructed, but new thinking is required about how machines interact with network-enabled storage devices. In this paper we describe how we implemented journaling in the Global File System (GFS), a shared-disk, cluster file system for Linux. Our previous three papers on GFS at the Mass Storage Symposium discussed our first three GFS implementations, their performance, and the lessons learned. Our fourth paper describes, appropriately enough, the evolution of GFS version 3 to version 4, which supports journaling and recovery from client failures. In addition, GFS scalability tests extending to 8 machines accessing 8 4-disk enclosures were conducted: these tests showed good scaling. We describe the GFS cluster infrastructure, which is necessary for proper recovery from machine and disk failures in a collection of machines sharing disks using GFS. Finally, we discuss the suitability of Linux for handling the big data requirements of supercomputing centers.