Proceedings ArticleDOI
A distributed shared buffer space for data-intensive applications
R. Lachaize,J.S. Hansen +1 more
- Vol. 2, pp 913-920
TLDR
This work proposes separating control and data transfer traffic by accessing data through a DSM-like cluster-wide shared buffer space and only including buffer references in the control messages, using a generic API for accessing buffers.Citations
More filters
Patent
Virtual disk storage techniques
John A. Starks,Dustin L. Green,Todd William Harris,Mathew John,Senthil Rajaram,Karan Mehra,Neal R. Christiansen,Chung Lang Dai +7 more
TL;DR: In this paper, the authors describe techniques for storing virtual disk payload data in an exemplary configuration, each virtual disk extent can be associated with state information that indicates whether the virtual disk file is described by a virtual disk.
Patent
Virtualization and offload reads and writes
TL;DR: In this paper, an offload read allows a requestor to obtain a token that represents data while an offloading write allows the requestors to request that the data (or a part thereof) represented by a token be logically written.
Patent
Offload Read and Write Offload Provider
TL;DR: In this article, an offload provider may provide a token that represents data, and the offload providers may be expected to ensure that the data the token represents will not change while the token is valid.
Patent
Compatibly extending offload token size
TL;DR: In this paper, the authors describe a mechanism that allows an offload provider to use larger tokens, such as physical or virtual tokens, in response to offload read and write commands.
References
More filters
Proceedings ArticleDOI
Cooperative caching: using remote client memory to improve file system performance
TL;DR: In this article, the authors examine four cooperative caching algorithms using a trace-driven simulation study and conclude that cooperative caching can significantly improve file system read response time and that relatively simple cooperative caching methods are sufficient to realize most of the potential performance gain.
Proceedings ArticleDOI
Implementing global memory management in a workstation cluster
Michael J. Feeley,W. E. Morgan,E. P. Pighin,Anna R. Karlin,Henry M. Levy,Chandramohan A. Thekkath +5 more
TL;DR: The objective is to use a single, unified, but distributed memory management algorithm at the lowest level of the operating system to manage memory globally at this level so that all system- and higher-level software, including VM, file systems, transaction systems, and user applications, can benefit from available cluster memory.
Journal ArticleDOI
Disk-directed I/O for MIMD multiprocessors
TL;DR: In this article, the authors proposed a disk-directed I/O technique, which allows the disk servers to determine the flow of data for maximum performance, and demonstrated that this technique provided consistent high performance that was largely independent of data distribution.
Book
The virtual interface architecture
D. Dunning,G. Regnier,Gary L. McAlpine,D. Cameron,B. Shubert,F. Berry,A.M. Merritt,E. Gronke,C. Dodd +8 more
TL;DR: In this paper, a protected, zero-copy, user-level network interface architecture is proposed to reduce the system overhead for sending and receiving messages between high-performance CPU/memory subsystems and networks to less than 10 microseconds.
Proceedings ArticleDOI
IO-lite: a unified I/O buffering and caching system
TL;DR: IO-Lite eliminates all copying and multiple buffering of I/O data, and enables various cross-subsystem optimizations, and shows performance improvements between 40 and 80% on real workloads as a result of IO -Lite.