Académique Documents
Professionnel Documents
Culture Documents
Parabon Crush
Harnessing The Power of Extreme-Scale Computation on Demand To Perform Statistical Data Mining
Microsoft Excel
Microsoft Excel
Parabon Crush performs statistical data mining and exhaustive regression analysis from within Microsoft Excel by exercising Frontier Grid Services. Workstations
Desktops
Parabon Crush
Statistical Data Mining at Scale
Parabon Crush
Parabon Crush is a statistical data mining application that uses the power of Frontier to identify explanatory regression models among the potentially vast set of all possible models. Unlike traditional statistical modeling tools, which use simple heuristics to rapidly produce answers that are often suboptimal, Crush systematically exhausts the entire space of possible models in its search for the best one and it does so quickly, thanks to the power of the Frontier Grid Platform.
Parabon Crush
Subject
1000 1001 1002 1003 9999
Y-Value
4.03 5.25 4.74 10.43 4.03
X1
20.1 26.2 23.7 52.1 20.1
X2
6 19 3.1415 42 (-1)
X3
1.3 2.1 0.8 4.6 -1.5
Xn
3.14 2.71 9.81 42 (-1)
y = x
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush
Subject
1000 1001 1002 1003 9999
Y-Value
4.03 5.25 4.74 10.43 4.03
X1
20.1 26.2 23.7 52.1 20.1
X2
6 19 3.1415 42 (-1)
X3
1.3 2.1 0.8 4.6 -1.5
Xn
3.14 2.71 9.81 42 (-1)
y = (1/5)x
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush
Subject
1000 1001 1002 1003 9999
Y-Value
4.03 5.25 4.74 10.43 4.03
X1
20.1 26.2 23.7 52.1 20.1
X2
6 19 3.1415 42 (-1)
X3
1.3 2.1 0.8 4.6 -1.5
Xn
3.14 2.71 9.81 42 (-1)
Parabon Crush
Subject
1000 1001 1002 1003 9999
Y-Value
4.03 5.25 4.74 10.43 4.03
X1
20.1 26.2 23.7 52.1 20.1
X2
6 19 3.1415 42 (-1)
X3
1.3 2.1 0.8 4.6 -1.5
Xn
3.14 2.71 9.81 42 (-1)
Parabon Crush
Subject
1000 1001 1002 1003 9999
Y-Value
4.03 5.25 4.74 10.43 4.03
X1
20.1 26.2 23.7 52.1 20.1
X2
6 19 3.1415 42 (-1)
X3
1.3 2.1 0.8 4.6 -1.5
Xn
14.3 10.5 31.4 12.3 70.3
y = (2)x3+(1/10)xn
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush
It can be used for deep correlation analysis, prediction and explanatory modeling. Users have applied Crush to many domains including: cancer research, epidemiology, nancial forecasting and social network analysis.
Parabon Crush
Crush runs as a Microsoft Excel Add-in. Once installed, Crush is available from the Tools menu in Excel. For exceptionally large datasets, Crush can instead be launched directly from the command-line, circumventing Excels data size limitations.
Parabon Crush
Parameters governing a search job are speci ed via standard dialog interfaces.
Parabon Crush
Crush supports linear, logit and ordered regression, and can be extended to handle other types of models as well.
Parabon Crush
By default, Crush employs an exhaustive combinatorial search. This ensures it nds the best possible model. However, the number of possible models is exponential in the number of variables, so even with grid-scale power, the time required to exhaust model spaces with more than 40 variables is prohibitive.
Parabon Crush
n
( )
n k
k=0
Parabon Crush
n
( )
n k n! k!(n-k)!
k=0 n
k=0
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush
Solution space grows EXPONENTIALLY (2n)
n n! k!(n-k)! k=0
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush
Solution space grows EXPONENTIALLY (2n)
7B 60M 500K 4M 30K 200 1 25 50 100 500 1000
Years
Columns
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush
For these cases, Crush employs a sophisticated evolutionary search algorithm to nd the best model in the time allotted. The ability to e ectively search such large model spaces is a breakthrough capability not found in modeling packages that lack Frontier power.
Parabon Crush
Once a job is fully speci ed, it can be launched against a Frontier grid without leaving Excel.
Parabon Crush
After a job is launched, a placeholder for results is created, where they are displayed automatically when the job is complete. The workbook can be closed and reopened at any time without a ecting running jobs.
Parabon Crush
As for all Frontier jobs, progress can be monitored from any browser via the Frontier Dashboard.
Parabon Crush
Jobs that would take years to complete on a single computer can be completed in hours or minutes, depending upon the level of grid capacity used. Upon completion, parameter estimates for the best models are returned and displayed in the workbook for further analysis.
Parabon Crush
Together, the Frontier Grid Platform and Parabon Crush deliver a revolutionary new combination of statistical data mining tools and Computational Work Units grid-scale computational power to answer deep and valuable questions about your data - fast!
2010 Parabon Computation, Inc. All rights reserved.
Parabon Crush