Overview | Tutorial Descriptions

Tuesday, February 26, 2008 Tuesday Morning Half-Day Tutorials T1 Clustered and Parallel Storage System Technologies Brent Welch and Marc Unangst, Panasas 9:00 a.m.–12:30 p.m. To meet the demands of increasingly data-hungry cluster applications, cluster-based parallel storage technologies are now capable of delivering performance scaling from 10s to 100s of GB/sec. This tutorial will examine current state-of-the-art high-performance file systems and the underlying technologies employed to deliver scalable performance across a range of scientific and industrial applications. The tutorial has two main sections. The first section describes the architecture of clustered, parallel storage systems and then compares several open-source and commercial systems based on this framework. Of particular interest are the newly emerging Object Storage Device (OSD) and Parallel NFS (pNFS) standards. The second half of the tutorial is about performance, including what benchmarking tools are available, how to use them to evaluate a storage system correctly, and how to optimize application I/O patterns to exploit the strengths and weaknesses of clustered, parallel storage systems. T2 Storage Class Memory, Technology, and Use Richard Freitas, Winfried Wilcke, and Bülent Kurdi, IBM Almaden Research Center 9:00 a.m.–12:30 p.m. The dream of replacing rotating mechanical storage, i.e., the disk drive, with solid-state, nonvolatile random access memory is finally becoming a reality. There are several technologies (>10) under active research and development. They are collectively called Storage Class Memory (SCM). The advent of this technology will likely have a significant impact on the design of future storage systems. This tutorial will describe several of the SCM technologies and how the availability of an SCM technology will impact the design of storage controllers, storage systems, etc. Storage-Class Memory (SCM) combines the benefits of solid-state memory—such as high performance and robustness—with the low cost of conventional hard-disk magnetic storage. Such a device would require a solid-state nonvolatile memory technology that could be manufactured at an extremely high effective areal density, using some combination of sublithographic patterning techniques, multiple bits per cell, and multiple layers of devices. We review the candidate solid-state non-olatile memory technologies that could potentially be used to construct such a Storage-Class Memory. Tuesday Afternoon Half-Day Tutorials T3 Parallel I/O for High-Performance Computing Rob Ross and Rob Latham, Argonne National Laboratory 1:30 p.m.–5:00 p.m. Today's scientific applications demand that high performance I/O be part of their operating environment. These applications access datasets of many gigabytes or terabytes, checkpoint frequently, and create large volumes of visualization data. Such applications desire interfaces to the I/O system that provide not only high performance but also meaningful abstractions that make interacting with the I/O system convenient. We will first discuss I/O in scientific applications and how the I/O software stack, from parallel file systems at the lowest layer, to intermediate layers such as MPI-IO, and finally high-level I/O libraries such as HDF-5, provides the capabilities needed by high-performance computing applications. Next we will highlight the role of parallel file systems (PFSs), first covering general concepts and then examining three examples: GPFS, Lustre, and PVFS. We will examine the upper layers of the I/O stack, covering POSIX I/O, MPI-IO, Parallel netCDF, and HDF5, and will discuss interface features and how application calls translate into PFS operations. Finally, looking at a case study, we will consider I/O best practice. T4 Cryptographic Methods for Protecting Storage Systems Christian Cachin, IBM Zurich Research Laboratory 1:30 p.m.–5:00 p.m. This tutorial presents cryptographic methods for storage protection, with a focus on recently developed techniques for encryption, integrity protection, and access control. Topics include: Protection of data in flight vs. data at rest Block-layer encryption Tweakable block ciphers Capabilities in object storage Designs for key management in file systems Encryption key management with lazy revocation Hash trees for integrity protection in file systems Principles of current tape and disk encryption products