Results 1 - 10 of 42,157

Table 2. Overhead for a virtual memory manager.

in A Case for Multi-Level Main Memory
by Magnus Ekman And, Magnus Ekman, Per Stenstrom 2004
"... In PAGE 8: ... Finally, the pageout thread writes the pages marked for evic- tion to disk. Table2 shows where the time is spent in these three parts in the existing virtual memory manager. The second column describes how many instructions it takes to run one loop in each part.... ..."
Cited by 1

Table 2. Abstract functions: communication.

in Lazy Parallelization: A Finite State Machine Based Optimization Approach for Data Parallel Image Processing Applications
by F. J. Seinstra, D. Koelma 2003
"... In PAGE 2: ...lude (a.o.) unary and binary pixel operations, neighbor- hood operations, and geometric transformations. Notation for unbreakable instruction streams relating to interprocess communication is presented in Table2 , and contains ab- stractions similar to operations in MPI [11]. The additional CreatLclPart/Full and DelLcl functions constitute creators and destructors for partial data structures (i.... ..."
Cited by 3

Table 2. Abstract functions: communication.

in Lazy Parallelization: A Finite State Machine Based Optimization Approach for Data Parallel Image Processing Applications
by F. J. Seinstra, D. Koelma 2003
"... In PAGE 2: ...lude (a.o.) unary and binary pixel operations, neighbor- hood operations, and geometric transformations. Notation for unbreakable instruction streams relating to interprocess communication is presented in Table2 , and contains ab- stractions similar to operations in MPI [11]. The additional CreatLclPart/Full and DelLcl functions constitute creators and destructors for partial data structures (i.... ..."
Cited by 3

Table 2. Abstract functions: communication.

in Lazy Parallelization: A Finite State Machine Based Optimization Approach for Data Parallel Image Processing Applications
by F. J. Seinstra, D. Koelma 2003
"... In PAGE 2: ...lude (a.o.) unary and binary pixel operations, neighbor- hood operations, and geometric transformations. Notation for unbreakable instruction streams relating to interprocess communication is presented in Table2 , and contains ab- stractions similar to operations in MPI [11]. The additional CreatLclPart/Full and DelLcl functions constitute creators and destructors for partial data structures (i.... ..."
Cited by 3

Table 1. Characteristics of 11 communication systems built for Myrinet.

in unknown title
by unknown authors
"... In PAGE 1: ... In this article, we refer to 11 such systems. Table1 lists the characteristics of each. As the table shows, the systems differ significantly in how they resolve the design issues we describe, but all aim for high performance and pro- vide a lean, low-level, and more or less generic com- munication facility.... In PAGE 4: ... Even if the operating system swaps out the page dur- ing transfer, the next memory reference by the host will generate a page fault, causing the operating sys- tem to swap the page back in. As Table1 shows, many protocols use DMA. FM, FM/MC, and LFC use pro- grammed I/O for host-to-interface transfers; AM-II, Hamlyn, and BIP do so only for small messages.... In PAGE 8: ...7 M uch research has been done on all these design issues for both Myrinet and other network technologies. In addition to the 11 systems in Table1 , which originate mostly from academic envi- ronments, industry has recently developed the Virtual Interface Architecture (http://www.viarch.... ..."

Table 5h: File and Virtual Memory system latencies (microseconds) 0K file 10 K file Host OS

in and Micro-Benchmarks ABSTRACT
by Mariam Salloum
"... In PAGE 6: ... As we can see, even when the amount of data transferred was less the Challenge performed worse than the Atlix. Table5 g: Local Communication bandwidth in MB/s Host OS Pipe AF UNIX TCP File reread Mmap reread Bcopy (libc) Bcopy (hand) Mem read Meme write Sgi-2 Linux 2.6.... In PAGE 7: ... Table5 g shows the local communication bandwidth, which includes inter-process communication and cache bandwidth. It evaluates the inter-process communication bandwidth, via evaluating several processes communicating through pipes or TCP sockets.... ..."

TABLE 2.5 Communication Management and Content in the Development of the Virtual Community

in DOING Complementary I NTERNET E xplorative RESEARCH Data Analysis CHAPTER 2 Complementary Explorative Data Analysis The Reconciliation of Quantitative and Qualitative Principles
by Fay Sudweeks, Simeon J. Simoff

Table 1: Comparison of Communication Abstractions.

in Implications of Hierarchical N-body Methodsfor Multiprocessor Architecture
by Jaswinder Pal Singh, John L. Hennessy, Anoop Gupta
"... In PAGE 26: ... Table1 summarizes the differences between machines that support the two paradigms. The primary advantage of explicit message-passing is the ease and efficiency of building scalable machines, since processing nodes require minimal hardware/software support for communication management.... ..."

Table 1 Parallel environments involved in network performance comparisons. Variations in architecture, interprocess communication vehicles, numbers of processes and com- puters are examined.

in A Hierarchical Partition Model for Adaptive Finite Element Computation
by J. D. Teresco, M. W. Beall, J. E. Flaherty, M. S. Shephard 1998
"... In PAGE 19: ... The speci c test sequence was (i) load the mesh onto one processor, (ii) partition the mesh onto n processors, (iii) re ne a portion of the mesh, and (iv) repartition the mesh to rebalance. Computers involved in the test con gurations ( Table1 ) were a 36-node IBM SP2, with communications over the high-speed switch and over an Ethernet network that connects all nodes; an eight-processor SGI Onyx2 with communication via shared memory; and three dual-processor Sun Ultra 2/2200 workstations, using shared mem- ory and both slow (10Mb/s) and fast (100Mb/s) Ethernet. The intial meshes contained 44,177 elements (coarse case), 79,944 elements (medium case), and 169,733 elements ( ne case).... ..."
Cited by 16

Table 1: Average call time for MinArg (ms) The average time for calling MinArg is 0.328 ms using nx, while 1.909 ms are required when using Norma. As Norma is about 6 times slower than nx on inter-node communications, our shared virtual memory facility uses nx when inter-node communications are required (e.g., when two external pagers communicate when a page fault occurs). Figures obtained for intra-node 2At the time this paper is written, the second i860 processor of each node is not exploited yet as a dedicated com- munication processor by the Paragon osf/1 operating system. Consequently, inter-process communication overhead is likely to decrease in future releases of the operating system for both nx and Norma. 4

in MYOAN: a Shared Virtual Memory Facility for the Intel Paragon Supercomputer
by Cabillic Priol Puaut
"... In PAGE 4: ... Measurements were made in multi-user mode. Table1 shows the average time required for calling MinArg when: (i) the communicating threads execute on distinct nodes; (ii) the communicating threads run on the same node but belong to di erent tasks; (iii) the communicating threads belong to the same task. The given values were obtained by dividing the total elapsed time of 10000 calls to MinArg... ..."
Results 1 - 10 of 42,157