Distributing Application and OS Functionality to Improve Application Performance

In this paper we demonstrate that the placement of functionality can have a significant impact on the performance of applications. OS bypass distributes OS policies to the network interface and protocol processing to the application to enhance application performance. We take this notion one step further and consider the distribution of application functionality to the network interface and/or the operating system. We illustrate the advantages of this approach by considering double buffering at the application level, a standard technique used to hide communication latency.

[1]  Peter Steenkiste.  A systematic approach to host interface design for high-speed networks , 1994, Computer.

[2]  D.E. Culler,et al.  Effects Of Communication Latency, Overhead, And Bandwidth In A Cluster Architecture , 1997, Conference Proceedings. The 24th Annual International Symposium on Computer Architecture.

[3]  Scott Pakin,et al.  Efficient layering for high speed communication: Fast Messages 2.x , 1998, Proceedings. The Seventh International Symposium on High Performance Distributed Computing (Cat. No.98TB100244).

[4]  Rolf Riesen,et al.  SUNMOS for the Intel Paragon - a brief user`s guide , 1994 .

[5]  Ron Brightwell,et al.  The Portals 3.0 Message Passing Interface Revision 1.0 , 1999 .

[6]  David S. Greenberg,et al.  Massively parallel computing using commodity components , 2000, Parallel Comput..