D
David E. Lowell
Researcher at University of Michigan
Publications - 5
Citations - 470
David E. Lowell is an academic researcher from University of Michigan. The author has contributed to research in topics: Cache & Failure transparency. The author has an hindex of 4, co-authored 5 publications receiving 469 citations.
Papers
More filters
Proceedings ArticleDOI
The Rio file cache: surviving operating system crashes
Peter M. Chen,Wee Teck Ng,Subhachandra Chandra,Christopher Aycock,Gurushankar Rajamani,David E. Lowell +5 more
TL;DR: The goal of the Rio (RAM I/O) file cache is to make ordinary main memory safe for persistent storage by enabling memory to survive operating system crashes by protecting memory during a crash and restoring it during a reboot.
Proceedings ArticleDOI
Free transactions with Rio Vista
David E. Lowell,Peter M. Chen +1 more
TL;DR: A system that improves transaction overhead by a factor of 2000 for working sets that fit in main memory and lowers transaction overhead to 5 μsec by using no redo log, no system calls, and only one memory-to-memory copy is presented.
Proceedings ArticleDOI
Exploring failure transparency and the limits of generic recovery
TL;DR: It is found that several real applications get failure transparency in the presence of simple stop failures with overhead of 0-12%, and that applications violate one invariant in the course of upholding the other for more than 90% of application faults and 3-15% of operating system faults, rendering transparent recovery impossible for these cases.
Proceedings ArticleDOI
Persistent messages in local transactions
David E. Lowell,Peter M. Chen +1 more
TL;DR: Under this model, messages and data are not lost after crashes, and all sends and receives are performed in local transactions, and Vistagrams are just as fast as traditional messages, despite the recoverability they offer.
Theory and practice of failure transparency
David E. Lowell,Peter M. Chen +1 more
TL;DR: A theory of consistent recovery is constructed that provides the fundamental rules for recovering transparently after a failure, and it is found that failure transparency is feasible, even for the challenging application domain the authors target.