• No results found

High Performance Robust Datamining for Cheminformatics

N/A
N/A
Protected

Academic year: 2020

Share "High Performance Robust Datamining for Cheminformatics"

Copied!
23
0
0

Loading.... (view fulltext now)

Full text

(1)

1

High Performance

Robust

Datamining for

Cheminformatics

Division of Chemical Information

Session: Cheminformatics: From Teaching to Research ACS Spring Meeting New Orleans

April 8 2008

Geoffrey Fox

Community Grids Laboratory, School of informatics Indiana University

http://www.chembiogrid.org http://www.infomall.org/multicore

(2)

Too much Computing?

n Historically both grids and parallel computing have tried to

increase computing capabilities by

Optimizing performance of codes at cost of re-usabilityExploiting all possible CPU’s such as Graphics

co-processors and “idle cycles” (across administrative domains)

Linking central computers together such as NSF/DoE/DoD

supercomputer networks without clear user requirements

n Next Crisis in technology area will be the opposite problem

commodity chips will be 32-128way parallel in 5 years time and we currently have no idea how to use them on commodity systems – especially on clients

Only 2 releases of standard software (e.g. Office) in this

time span so need solutions that can be implemented in next 3-5 years

n Intel RMS analysis: Gaming and Generalized decision

(3)
(4)

Too much Data to the Rescue?

n Multicore servers have clear “universal parallelism” as many

users can access and use machines simultaneously

n Maybe also need application parallelism (e.g. datamining) as

needed on client machines

n Over next years, we will be submerged of course in data

deluge

Scientific observations for e-Science including

cheminformatics (high throughput screening)

Local (video, environmental) sensors

Data fetched from Internet defining users interests

n Maybe data-mining of this “too much data” will use up the

“too much computing” both for science and commodity PC’s

PC will use this data(-mining) to be intelligent user

assistant?

Must have highly parallel algorithms and new algorithms

(5)

CICC Chemical Informatics and Cyberinfrastructure Collaboratory Web Service Infrastructure

Portal Services

RSS Feeds User Profiles

Collaboration as in Sakai

Core Grid Services

Service Registry

Job Submission and Management Local Clusters

IU Big Red, TeraGrid, Open Science Grid

Varuna.net

Quantum Chemistry

OSCAR Document Analysis InChI Generation/Search

Computational Chemistry (Gamess, Jaguar etc.)

Need to make all this parallel Hide parallelism in service

(6)

Service Aggregated Linked Sequential Activities

GOALS: Increasing number of cores accompanied by continued

data deluge

Develop scalable parallel data mining algorithms with good

multicore and cluster performance; understand software runtime and parallelization method. Use

managed code (C#) and package

algorithms as services to encourage broad use assuming experts

parallelize core algorithms.

CURRENT RESUTS: Microsoft CCR supports MPI, dynamic threading and via DSS a Service model of computing; detailed performance

measurements

Speedups of 7.5 or above on 8-core systems for “large problems” with deterministic annealed

(avoid local minima) algorithms for clustering, Gaussian Mixtures, GTM and MDS (dimensional reduction) etc.

SALSA Team (funded by

Geoffrey Fox Microsoft) Xiaohong Qiu Seung-Hee Bae Huapeng Yuan Indiana University Technology Collaboration George Chrysanthakopoulos Henrik Frystyk Nielsen

Microsoft

Application Collaboration

Cheminformatics(funded by NIH

Rajarshi Guha ECCR)

David Wild Bioinformatics

Haiku Tang

Demographics (GIS) Neil Devadasan

IU Bloomington and IUPUI

(7)

Unsupervised Modeling

Find clusters without prejudice

Model distribution as clusters formed from Gaussian distributions with general shape

Both can use multi-resolution annealing

SALSA

N data points X(x) in D dimensional space OR points with dissimilarityij defined between them

General Problem

Classes

Dimensional Reduction/Embedding

Given vectors, map into lower dimension space

“preserving topology” for visualization: SOM and GTM

Givenijassociate data points with vectors in a

Euclidean space with Euclidean distance approximately

ij : MDS (can anneal) and Random Projection

(8)

Minimize Free Energy F = E-TS where E objective function

(energy) and S entropy.

Reduce temperature T logarithmically; T=is dominated by

Entropy, T small by objective function

S regularizes E in a natural fashion

In simulated annealing, use Monte Carlo but in deterministic

annealing, use mean field averages

<F> =exp(-E0/T) F over the Gibbs distribution

P0 = exp(-E0/T) using an energy function E0similar to E but for

which integrals can be calculated

E0 = E for clustering and related problems

General simple choice is E0 =(xi -i)2where xi parameters to

be annealed

(9)

Deterministic Annealing Clustering (DAC)

a(x) = 1/N or generally p(x) withp(x) =1

g(k)=1 and s(k)=0.5

T is annealing temperature varied down from

with final value of 1

Vary cluster center Y(k)

K starts at 1 and is incremented by algorithm; pick resolution NOT number of clusters

My 4th most cited article but little used; probably

as no good software compared to simple K-means

Avoid local minima

SALSA

(10)

Deterministic Annealing Clustering of Indiana Census Data

Decrease temperature (distance scale) to discover more clusters

(11)

Minimum evolving as temperature decreases

Movement at fixed temperature going to local

minima if not initialized “correctly”

Solve Linear Equations for each

temperature

Nonlinearity removed by approximating with solution at previous higher

temperature

Deterministi Annealing

F({Y}, T)

(12)

Deterministic Annealing Clustering (DAC)

a(x) = 1/N or generally p(x) withp(x) =1

g(k)=1 and s(k)=0.5

T is annealing temperature varied down from

with final value of 1

Vary cluster center Y(k) but can calculate weight Pk and correlation matrix s(k) =(k)2 (even for

matrix(k)2) using IDENTICAL formulae for

Gaussian mixtures

•K starts at 1 and is incremented by algorithm

Deterministic Annealing Gaussian

Mixture models (DAGM

)

a(x) = 1

g(k)={Pk/(2(k)2)D/2}1/T

s(k)=(k)2 (taking case of spherical Gaussian)

T is annealing temperature varied down from

with final value of 1

Vary Y(k) Pk and(k)

K starts at 1 and is incremented by algorithm

SALSA

N data points E(x) in D dim. space and Minimize F by EM

a(x) = 1 and g(k) = (1/K)(/2)D/2

s(k) = 1/and T = 1

Y(k) =m=1M Wmm(X(k))

Choose fixedm(X) = exp( - 0.5 (X-m)2/2 )

Vary Wm andbut fix values of M and K a priori

Y(k) E(x) Wmare vectors in original high D dimension space

X(k) andmare vectors in 2 dimensional mapped space

Generative Topographic Mapping (GTM)

As DAGM but set T=1 and fix K

Traditional Gaussian

mixture models GM

GTM has several natural annealing

versions based on either DAC or DAGM: under investigation

DAMDS different form as different Gibbs distribution (different E0)

(13)

Speedup = Number of cores/(1+f)

f= (Sum of Overheads)/(Computation per core)

ComputationGrain Size n . # Clusters K

Overheads are

Synchronization: small with CCR

Load Balance: good

Memory Bandwidth Limit:0 as K  

Cache Use/Interference: Important

Runtime Fluctuations: Dominant large n, K All our “real” problems have f ≤ 0.05 and speedups on 8 core systems greater than 7.6

(14)

We implement micro-parallelism using Microsoft CCR

(Concurrency and Coordination Runtime) as it supports both MPI rendezvous and dynamic (spawned) threading style of parallelism

http://msdn.microsoft.com/robotics/

CCR Supports exchange of messages between threads using named ports

and has primitives like:

FromHandler: Spawn threads without reading ports

Receive: Each handler reads one item from a single port

MultipleItemReceive: Each handler reads a prescribed number of items

of a given type from a given port. Note items in a port can be general structures but all must have same type.

MultiplePortReceive: Each handler reads a one item of a given type from

multiple ports.

CCR has fewer primitives than MPI but can implement MPI collectives

efficiently

Use DSS (Decentralized System Services) built in terms of CCR for service

model

DSS has ~35 µs and CCR a few µs overhead

(15)

MPI Exchange Latency in µs (20-30 µs computation between messaging)

Machine OS Runtime Grains Parallelism MPI Latency

Intel8c:gf12

(8 core 2.33 Ghz) (in 2 chips)

Redhat MPJE(Java) Process 8 181

MPICH2 (C) Process 8 40.0

MPICH2:Fast Process 8 39.3

Nemesis Process 8 4.21

Intel8c:gf20

(8 core 2.33 Ghz)

Fedora MPJE Process 8 157

mpiJava Process 8 111

MPICH2 Process 8 64.2

Intel8b

(8 core 2.66 Ghz)

Vista MPJE Process 8 170

Fedora MPJE Process 8 142

Fedora mpiJava Process 8 100

Vista CCR (C#) Thread 8 20.2

AMD4

(4 core 2.19 Ghz)

XP MPJE Process 4 185

Redhat MPJE Process 4 152

mpiJava Process 4 99.4

MPICH2 Process 4 39.3

XP CCR Thread 4 16.3

Intel(4 core) XP CCR Thread 4 25.8

SALSA

(16)

GTM Projection of 2 clusters of 335

compounds in 155 dimensions

GTM Projection of PubChem:

10,926,94 compounds in 166 dimension binary property space takes 4 days on 8 cores. 64X64 mesh of GTM clusters interpolates

PubChem. Could usefully use 1024 cores! David Wild will use for GIS style 2D browsing interface to chemistry

PCA GTM

Linear PCA v. nonlinear GTM on 6 Gaussians in 3D PCA is Principal Component Analysis

Parallel Generative Topographic Mapping

GTM Reduce dimensionality preserving

topology and perhaps distance Here project to 2D

(17)

Minimize Stress

(X) =i<j=1n weight(i,j) (ij - d(Xi, Xj))2

 ij are input dissimilarities and d(Xi, Xj) the Euclidean distance squared in embedding space (2D here)

SMACOF or Scaling by minimizing a complicated function is clever

steepest descent algorithm

Use GTM to initialize SMACOF

(18)

Use deterministically annealed version of GTM

Do not use GTM at all but rather find clusters by DAC

algorithm and then use MDS iteratively with one point

(cluster center) added each iteration

and/or use Newton’s method for MDS as only thousands

of parameters (# clusters times dimension l)

and/or use deterministically annealed MDS (DAMDS

(X,T) =

i<j

=1n

weight(

i,j

) (d(X

i

,

X

j

) + 2T(l+2)-

ij

)

2

Where

T

annealing temperature and

l

dimension of

embedding space (2 in example)

d(X

i

,

X

j

) = (X

i

X

i

)

2

in l dimensional latent space

(19)

(X,T) =

i<j

=1n

weight(

i,j

) (d(X

i

,

X

j

) + 2T(l+2)-

ij

)

2

Note that that at T=

,

2T(l+2)-

ij

is positive and all

points

X

i

are at origin. As T decreases, the terms with

large

ij

become negative and associated points

gradually expand from origin

Physical Optimization

”: Think of points

X

i

as “particles”

moving under influence of forces with other points.

Forces are in direction of vector between particles

Attractive:

d(X

i

,

X

j

) >

ij

-

2T(l+2)

Repulsive:

d(X

i

,

X

j

) <

ij

-

2T(l+2)

Can use iterative method based on this particle

(20)

Use Data Decomposition as in classic distributed memory

but use shared memory for read variables. Each thread uses a “local” array for written variables to get good cache performance

Multicore and Cluster use same parallel algorithms but

different runtime implementations; algorithms are

Accumulate matrix and vector elements in each process/thread

At iteration barrier, combine contributions (MPI_Reduce)

Linear Algebra (multiplication, equation solving, SVD) “Main Thread” and Memory

M 1 m 1 0 m 0 2 m 2 3 m 3 4 m 4 5 m 5 6 m 6 7 m 7 Subsidiary threads t with memory

mt MPI/CCR/DSS From other nodes MPI/CCR/DSS From other nodes

(21)
(22)

All parallel algorithms packaged as services and not traditional

libraries

MPI-Style Micro-parallelism uses low latency CCR threads or MPI

processes

CCR microseconds; local services 10’s microseconds; distributed services milliseconds

Services can be used where loose coupling natural

Input dataAlgorithms

PCA

DAC GTM GM DAGM DAGTM – both for complete algorithm and for each iteration

Linear Algebra used inside or outside above

Metric embedding MDS, Bourgain, Quadratic Programming ….

HMM, SVM ….

User interface: GIS (Web map Service) or equivalent

(23)

This class of data mining does/will parallelize well on current/future multicore

nodes

Several engineering issues for use in large applications

How to take CCR in multicore node to cluster (MPI or cross-cluster CCR?)Use Google MapReduce on Cloud/Grid

Need high performance linear algebra for C# (PLASMA from UTenn)

Access linear algebra services in a different language?

Need equivalent of Intel C Math Libraries for C# (vector arithmetic – level 1

BLAS)

Service model to integrate module

Although work used C#, similar results in C, C++, Java, Fortran

Future work is more applications; any suggestions?

Refine current algorithms such as DAGTM, SMACOF, DAMDSNew parallel algorithms

Clustering with pairwise distances but no vector spaces. Deterministic

annealing here well understood but even less used

Bourgain Random Projection for metric embedding

Support use of Newton’s Method (Marquardt’s method) as EM alternativeLater HMM and SVM

References

Related documents

the cost of a hand-held pulse oximeter, per DALY averted, as a function of baseline anaesthetic-related mortality, assuming that pulse oximetry prevents 10% of

Por otro lado, si f '(x) ≥ 0, entonces DF &gt; 0 y trDF ≤ -γb &lt; 0, por lo tanto los equilibrios que se encuentran en la parte decreciente de la ceroclina-x son

The dyes used in this treatment are Naphthol Green B which comes under Industrial dyes, Acid Orange 74 which comes under Protein textile dyes, Disperse Blue 14 comes under Synthetic

En primer lugar, para la descripción del perfil de Twitter se emplean las siguientes variables: número de seguidores (followers) y de personas a las que sigue la cuenta del

Based on the results of this study, emotional factor identified as the highest score related to mathematics anxiety followed by environmental and assessment factors with no

We experiment the proposed model with monthly demand data that are relevant to Hong Kong’s tourism industry, and compare the performance of the sparse GPR model with those of

Humidity Operation Endurance test applying the electric stress (voltage &amp; current) and the high thermal with high humidity stress for a long time. 60 sec in each of 3 directions

However, since Fossbakk made the error of adding an additional extra digit to a sequence, many of the account numbers used have sequences of two, three, four and even five