Cues : File Subsystem Data Streams
نویسندگان
چکیده
We present a system call which enables directing high performance data transfers with in-kernel streams. The streams are de ned and run using abstract data types called Cues, formed over the Unix le subsystem. The system call, named cue() , returns a descriptor, which can be used to write requests to the newly created stream. These requests de ne the ow of the stream. The abstraction simpli es the design of applications that transfer large amounts of data from les or devices. It also enables high throughput when multiple transfers are in progress. Cue code is compact, modular, and portable. This model also results in a simple mechanism for remotely cueing data ow using standard connections with peer processes. The implementation and tests are also described in this paper. Also with Department of Computer Science. Also with Institute of Advanced Computer Studies. Also with Department of Electrical Engineering. This material is based upon work supported in part by the National Science Foundation under Grant No. NSF EEC 94-02384, and by the Center for Satellite and Hybrid Communication Networks under NASA contract NAGW-2777, by the University of Maryland Institute of Advanced Computing Studies, by ARPA under Grant No. F30602{93{C{0177, Maryland Industrial Partnerships and Loral Corporation. Part of the design for this work was also done by the rst author as a summer intern at AT&T Bell Laboratories, Murray Hill in Summer '94.
منابع مشابه
Coordination between parallel TCP streams on Long Fat Pipe Network
“Data Reservoir” is a file sharing facility system which uses parallel streams for huge data transfer on Long Fat Pipe Network. While experimenting Data Reservoir system, we observed that performances of parallel streams disperse a lot, which results in damage on total performance. To equalize throughput of parallel streams, we propose “Dulling Edges of Cooperative Parallel streams(DECP)”. This...
متن کاملManaging File Subsystem Data Streams for Databases on Networked Systems
One important activity for networked database systems that distribute data across several workstations is moving data between the le and network subsystems. It is possible to create data streams in the operating system kernel. If provided on a system, they allow user level processes to request transfer of data without having to copy it into the user space. This is particularly useful for data w...
متن کاملAn Efficient and Low-Cost Input/Output Subsystem for Network Processors
† Also with the Institute of Computer Science (ICS), Foundation for Research and Technology-Hellas (FORTH). ABSTRACT We present the architecture and implementation of an input/output subsystem for a cost-effective network processor. We believe that adding processing power to a networking chip is relatively straightforward. However, transferring data to and from the processor(s) is insufficient ...
متن کاملOn The Relationship Of Server Disk Workloads And Client File Requests
In this study, we consider the relationship of client IO requests submitted to a file server and the server's disk subsystem IO traffic. We collected traces of client-server IOs and, during each trace period, we also traced server storage subsystem workloads. We analyze and compare trace pairs collected in each trace period. We evaluate file server performance and investigate the relationship b...
متن کاملStreamline: Efficient OS Communication Through Versatile Streams
Streamline is a stream-based communication subsystem that spans from embedded hardware to userspace processes. It improves performance of memory and I/O bound applications by constructing tailor-made datapaths for each application at runtime. Datapath optimisation removes unnecessary copying, context switching and cache replacement. It simplifies integration of embedded and distributed hardware...
متن کامل