Conference on File and Storage Technologies

Conference on File and Storage Technologies
Name	Conference on File and Storage Technologies
Abbreviation	FAST
Discipline	Computer science
Venue	Various
Country	International
Established	2002
Frequency	Biennial (originally annual)

Contents

Conference on File and Storage Technologies The Conference on File and Storage Technologies was a leading international forum for research on file systems, storage architectures, and data management, attracting contributors from Carnegie Mellon University, University of California, Berkeley, Massachusetts Institute of Technology, Stanford University, and University of Washington. The conference featured peer-reviewed program committees and produced proceedings cited by researchers at Google, Microsoft Research, Facebook, Amazon Web Services, and NetApp. FAST fostered collaboration among academics from institutions such as Princeton University, University of California, San Diego, University of Illinois Urbana–Champaign, Cornell University, and University of Texas at Austin.

Overview

FAST focused on design, implementation, and evaluation of storage systems including file systems, object stores, distributed storage, and archival systems, drawing attendees from Intel, IBM Research, Hewlett Packard Enterprise, Dell Technologies, and Hitachi. The conference emphasized empirical evaluation methods used by teams at AT&T Labs, Bell Labs, Xerox PARC, Oracle Corporation, and Seagate Technology. FAST proceedings were indexed by databases associated with Association for Computing Machinery, IEEE Computer Society, USENIX Association, ACM SIGOPS, and ACM SIGMOD.

FAST was established in 2002 following precursors like workshops at USENIX Annual Technical Conference and initiatives by researchers from University of California, Santa Cruz, University of Maryland, College Park, Harvard University, Yale University, and Columbia University. Early organizers included faculty affiliated with International Conference on Data Engineering, Symposium on Operating Systems Principles, Conference on Empirical Methods in Natural Language Processing (as methodological exemplars), and committees involving members from European Conference on Computer Systems, ASPLOS, and SOSP. Over time FAST evolved alongside technologies from RAID, solid-state drive, SSD, NVMe, SATA, and standards from JEDEC, SNIA, and NFS.

Topics covered durable storage, consistency, crash recovery, metadata management, caching, tiering, deduplication, erasure coding, and compression, with contributions intersecting work at INRIA, Max Planck Institute for Informatics, TELENOR, Netflix, and Dropbox. FAST sessions included papers on distributed file systems in the tradition of Google File System, Hadoop Distributed File System, and Ceph, as well as object store research related to Amazon S3 and OpenStack Swift. Related research drew on theory and practice from MIT Lincoln Laboratory, Los Alamos National Laboratory, Lawrence Berkeley National Laboratory, Argonne National Laboratory, and Sandia National Laboratories.

FAST proceedings were published and archived alongside proceedings of USENIX FAST Workshop editions and cited at venues such as VLDB, SIGMOD, ICDE, EuroSys, and SoCC. Notable program chairs and organizers hailed from University of Michigan, University of California, Irvine, Northwestern University, Brown University, and Duke University. The conference often included keynote addresses from leaders at Google Research, Microsoft Azure, Facebook AI Research, Amazon Research, and IBM Watson. Proceedings were presented at campuses like Stanford University, University of California, Los Angeles, Georgia Institute of Technology, Purdue University, and University of Toronto.

FAST published influential papers that informed systems like ZFS from Sun Microsystems, Btrfs from Oracle Corporation, and research prototypes that influenced Facebook storage efforts, Google infrastructure, and Netflix streaming backend. Landmark contributions addressed concepts from log-structured file system, shadow paging, copy-on-write, write-ahead logging, and distributed consensus such as implementations of Paxos and Raft used in storage metadata. Authors included researchers from University of Cambridge, University of Oxford, ETH Zurich, EPFL, and Tsinghua University.

FAST was typically organized by committees composed of academics and industry practitioners representing ACM, USENIX, IEEE, SNIA, and Storage Networking Industry Association. Sponsors included corporations like Intel Corporation, Western Digital, Samsung Electronics, Micron Technology, and Seagate Technology. Funding and support came from research agencies such as National Science Foundation, European Research Council, DARPA, and agencies collaborating with NIH and NSF-funded centers.

FAST influenced product development and academic curricula through cross-pollination with projects at Google Cloud Platform, Microsoft Azure Storage, Amazon Web Services, Alibaba Cloud, and Tencent Cloud. Its findings informed standards work at IETF, IEEE Standards Association, and drove innovations adopted by vendors like EMC Corporation, Pure Storage, NetApp, IBM, and HPE. FAST-trained researchers advanced careers at institutions including LUMS, KAUST, Indian Institute of Science, National University of Singapore, and University of Melbourne, while contributions seeded startups and influenced open-source communities such as Linux Kernel, Kubernetes, Ceph, GlusterFS, and OpenStack.