Codebase list dx / scrub-obsolete/main help / dxall272
scrub-obsolete/main

Tree @scrub-obsolete/main (Download .tar.gz)

dxall272 @scrub-obsolete/main — raw · history · blame 

 1
 2
 3
 4
 5
 6
 7
 8
 9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
#!F-adobe-helvetica-medium-r-normal--18*
#!N 
#!CNavyBlue #!N  #!Rall271 Parallelism using Distributed Processing #!N 
#!EC #!N #!N Data Explorer provides the capability of distributing the 
execution of a visual program or a program generated using the 
scripting language over multiple workstations on a network. Distributing the execution 
provides parallelism and enhanced resource utilization. Parallelism is achieved by the 
simultaneous execution of different portions of the visualization on each of 
the workstations. Enhanced resource utilization can be achieved, for example, by 
assigning computationally intensive portions of the visualization to the more powerful 
workstations, or transformation and realization functions that are applied to data 
located remotely can be distributed to the remote workstations, reducing the 
amount of data transfer. #!N #!N Distributed processing is achieved in 
two ways: using "outboard" modules or placing groups of tools into 
"execution groups." These two methods can be used independently or in 
combination. An outboard module is a user-written module controlled by the 
Data Explorer executive but is external to the Data Explorer server 
program. They can be invoked from either a visual program or 
a script program. Execution groups are a set of tools that 
can be assigned to a workstation. Once groups are created, they 
can be assigned to the workstations over which the visualization is 
to be distributed. More than one group can be assigned to 
each workstation. (See also  #!Ldistcmp,dxall484 h Using Distributed Computation  #!EL  .) #!N #!N #!N  #!F-adobe-times-medium-i-normal--18*   Next 
Topic #!EF #!N #!N  #!Lall272,dxall273 h Parallelism for Data Explorer SMP  #!EL  #!N  #!F-adobe-times-medium-i-normal--18*   #!N