The Alloc Stream Facility: a redesign of application-level stream I/O

Krieger, Orran; Stumm, Michael; Unrau, Ronald C.

doi:10.1109/2.268889

Cited by 15 publications

(6 citation statements)

References 7 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…For every network and disk intensive application an experimentation regarding the buffer size is needed. There have also been a number of researches regarding the application level I/O facility such as Alloc Stream Facility [7]. During all tests, 64MB file size was used.…”

Section: Related Workmentioning

confidence: 99%

Distributed File System: Efficiency Experiments for Data Access and Communication

Upadhyaya

Azimov

Doan

et al. 2008

2008 Fourth International Conference on Networked Computing and Advanced Information Management

View full text Add to dashboard Cite

In a Distributed File System, operations such as read, write, send and receive should be efficient and must increase the overall performance of the system. Especially every remote read and write operation we encounter network and secondary storage latencies. In order to reduce this latency time and improve the performance of Distributed File System, we analyze the problem of selecting application-level buffer size for the Network communication and Disk Input/Output operations. A number of experiments were carried out to identify the optimal size for our application-level Buffer.

show abstract

Section: Related Workmentioning

confidence: 99%

Distributed File System: Efficiency Experiments for Data Access and Communication

Upadhyaya

Azimov

Doan

et al. 2008

2008 Fourth International Conference on Networked Computing and Advanced Information Management

View full text Add to dashboard Cite

show abstract

“…Another user-level approach for fast I/O is the Alloc Stream Facility [38]. Fundamentally, all these schemes place the application in the center of data transfers, making main memory and the application itself (i.e., the main CPU) potential performance bottlenecks.…”

Section: Related Workmentioning

confidence: 99%

High-speed I/O

Burnside

Keromytis

2003

Proceedings of the ACM SIGCOMM Workshop on Network-I/O Convergence: Experience, Lessons, Implications

View full text Add to dashboard Cite

The design of modern operating systems is based around the concept of memory as a cache for data that flows between applications, storage, and I/O devices. With the increasing disparity between I/O bandwidth and CPU performance, this architecture exposes the processor and memory subsystems as the bottlenecks to system performance. Furthermore, this design does not easily lend itself to exploitation of new capabilities in peripheral devices, such as programmable network cards or special-purpose hardware accelerators, capable of card-to-card data transfers.We propose a new operating system architecture that removes the memory and CPU from the data path. The role of the operating system becomes that of data-flow management, while applications operate purely at the signaling level. This design parallels the evolution of modern network routers, and has the potential to enable high-performance I/O for end-systems, as well as fully exploit recent trends in programmability of peripheral (I/O) devices.

show abstract

“…The Alloc Stream Facility (ASF) is a file system library in the application address space that translates application file read and write requests into accesses to a mapped file region [Krieger 1994;Krieger et al 1994]. On the first access to a page in the region, the process will page-fault.…”

Section: The File Systemmentioning

confidence: 99%

Hierarchical clustering: A structure for scalable multiprocessor operating system design

et al. 1995

View full text Add to dashboard Cite

Abstract. We introduce the concept of hierarchical clustering as a way to structure shared-memory multiprocessor operating systems for scalability. The concept is based on clustering and hierarchical system design. Hierarchical clustering leads to a modular system, composed of easy-to-design and efficient building blocks. The resulting structure is scalable because it 1) maximizes locality, which is key to good performance in NUMA (non-uniform memory access) systems and 2) provides for concurrency that increases linearly with the number of processors. At the same time, there is tight coupling within a cluster, so the system performs well for local interactions that are expected to constitute the common case. A clustered system can easily be adapted to different hardware configurations and architectures by changing the size of the clusters. We show how this structuring technique is applied to the design of a microkernel-based operating system called HURRICANE. This prototype system is the first complete and running implementation of its kind and demonstrates the feasibility of a hierarchically clustered system. We present performance results based on the prototype, demonstrating the characteristics and behavior of a clustered system. In particular, we show how clustering trades off the efficiencies of tight coupling for the advantages of replication, increased locality, and decreased lock contention.

show abstract

The Alloc Stream Facility: a redesign of application-level stream I/O

Cited by 15 publications

References 7 publications

Distributed File System: Efficiency Experiments for Data Access and Communication

Distributed File System: Efficiency Experiments for Data Access and Communication

High-speed I/O

Hierarchical clustering: A structure for scalable multiprocessor operating system design

Contact Info

Product

Resources

About