SALSASALSA

Overview of Cloud Computing Platforms

July 28, 2010 Big Data for Science Workshop

Judy Qiuxqiu@indiana.edu

http://salsahpc.indiana.edu

Pervasive Technology Institute

School of Informatics and Computing

Indiana University

SALSA

Important Trends

•Implies parallel computing important again

•Performance from extra cores – not extra clock speed

•new commercially supported data center model building on compute grids

•In all fields of science and throughout life (e.g. web!)

•Impacts preservation, access/use, programming model

Data DelugeCloud

Technologies

eScience

Multicore/

Parallel Computing •A spectrum of eScience or

eResearch applications (biology, chemistry, physics social science and

humanities …)

•Data Analysis

•Machine learning

SALSA

Challenges for CS Research

There’re several challenges to realizing the vision on data intensive systems and building generic tools (Workflow, Databases, Algorithms, Visualization ).

• Cluster/Cloud-management software

• Distributed execution engine

• Security and Privacy

• Language constructs e.g. MapReduce Twister …

• Parallel compilers

• Program Development tools

. . .

Science faces a data deluge. How to manage and analyze information?

Recommend CSTB foster tools for data capture, data curation, data analysis

―Jim Gray’sTalk to Computer Science and Telecommunication Board (CSTB), Jan 11, 2007

SALSA

Data We’re Looking at

• Public Health Data (IU Medical School & IUPUI Polis Center)

(65535 Patient/GIS records / 54 dimensions each)

• Biology DNA sequence alignments (IU Medical School & CGB)

(several million Sequences / at least 300 to 400 base pair each)

• NIH PubChem (Cheminformatics)

(60 million chemical compounds/166 fingerprints each)

• Particle physics LHC (Caltech)

(1 Terabyte data placed in IU Data Capacitor)

High volume and high dimension require new efficient computing approaches!

SALSA

Data is too big and gets bigger to fit into memory For “All pairs” problem O(N2),

PubChem data points 100,000 => 480 GB of main memory(Tempest Cluster of 768 cores has 1.536TB)

We need to use distributed memory and new algorithms to solve the problem

Communication overhead is large as main operations include matrix multiplication (O(N2)), moving data between nodes and within one node adds extra overheads

We use hybrid mode of MPI and MapReduce between nodes and concurrent threading internal to node on multicore clusters

Concurrent threading has side effects (for shared memory model like CCR and OpenMP) that impact performance

sub-block size to fit data into cache cache line padding to avoid false sharing

Data Explosion and Challenges

SALSA

Gartner 2009 Hype CurveSource: Gartner (August 2009)

HPC?

SALSA

Clouds hide Complexity

7

SaaS: Software as a Service

(e.g. Clustering is a service)

IaaS (HaaS): Infrastructure as a Service

(get computer time with a credit card and with a Web interface like EC2)

PaaS: Platform as a Service

IaaS plus core software capabilities on which you build SaaS

(e.g. Azure is a PaaS; MapReduce is a Platform)

CyberinfrastructureIs “Research as a Service”

SALSA

Cloud Computing: Infrastructure and Runtimes

• Cloud Infrastructure: outsourcing of servers, computing, data, file space, utility computing, etc.

– Handled through (Web) services that control virtual machine lifecycles.

• Cloud Runtimes or Platform: tools (for using clouds) to do data-parallel (and other) computations.

– Apache Hadoop, Google MapReduce, Microsoft Dryad, Bigtable, Chubby (synchronization) and others

– MapReduce designed for information retrieval but is excellent for a wide range of science data analysis applications

– Can also do much traditional parallel computing for data-mining if extended to support iterative operations

– MapReduce not usually done on Virtual Machines

SALSA

Gri

d/C

lou

dAuthentication and Authorization: Provide single sign in to both FutureGrid and

Commercial Clouds linked by workflow

Workflow: Support workflows that link job components between FutureGrid and

Commercial Clouds. Trident from Microsoft Research is initial candidate

Data Transport: Transport data between job components on FutureGrid and Commercial

Clouds respecting custom storage patterns

Software as a Service: This concept is shared between Clouds and Grids and can be

supported without special attention

SQL: Relational Database

Clo

ud

Program Library: Store Images and other Program material (basic FutureGrid facility)

Blob: Basic storage concept similar to Azure Blob or Amazon S3

DPFS Data Parallel File System: Support of file systems like Google (MapReduce), HDFS

(Hadoop) or Cosmos (Dryad) with compute-data affinity optimized for data processing

Table: Support of Table Data structures modeled on Apache Hbase (Google Bigtable) or

Amazon SimpleDB/Azure Table (eg. Scalable distributed “Excel”)

Queues: Publish Subscribe based queuing system

Worker Role: This concept is implicitly used in both Amazon and TeraGrid but was first

introduced as a high level construct by Azure

Web Role: This is used in Azure to describe important link to user and can be supported in

FutureGrid with a Portal framework

MapReduce: Support MapReduce Programming model including Hadoop on Linux, Dryad

Key Features of Cloud Platforms

SALSA

MapReduce “File/Data Repository” Parallelism

Instruments

Disks Map1 Map2 Map3

Reduce

Communication

Map = (data parallel) computation reading and writing dataReduce = Collective/Consolidation phase e.g. forming multiple global sums as in histogram

Portals/Users

MPI and Iterative MapReduceMap Map Map Map

Reduce Reduce Reduce

SALSA

MapReduce

• Implementations support:– Splitting of data

– Passing the output of map functions to reduce functions

– Sorting the inputs to the reduce function based on the intermediate keys

– Quality of services

Map(Key, Value)

Reduce(Key, List<Value>)

Data Partitions

Reduce Outputs

A hash function maps the results of the map tasks to r reduce tasks

A parallel Runtime coming from Information Retrieval

SALSA

• Sam thought of “drinking” the apple

Sam’s Problem

He used a to cut the

and a to make juice.

SALSA

(<a’, > , <o’, > , <p’, > )

• Implemented a parallel version of his innovation

Creative Sam

(<a, > , <o, > , <p, > , …)

Each input to a map is a list of <key, value> pairs

Each output of slice is a list of <key, value> pairs

Grouped by key

Each input to a reduce is a <key, value-list> (possibly a list of these, depending on the grouping/hashing mechanism)e.g. <ao, ( …)>

Reduced into a list of values

The idea of Map Reduce in Data Intensive Computing

A list of <key, value> pairs mapped into another list of <key, value> pairs which gets grouped by

the key and reduced into a list of values

SALSA

Hadoop & DryadLINQ

• Apache Implementation of Google’s MapReduce• Hadoop Distributed File System (HDFS) manage data• Map/Reduce tasks are scheduled based on data

locality in HDFS (replicated data blocks)

• Dryad process the DAG executing vertices on compute clusters

• LINQ provides a query interface for structured data

• Provide Hash, Range, and Round-Robin partition patterns

JobTracker

NameNode

1 2

32

3 4

M MM M

R R R R

H

D

F

S

Datablocks

Data/Compute NodesMaster Node

Apache Hadoop Microsoft DryadLINQ

Edge : communication path

Vertex :execution task

Standard LINQ operations

DryadLINQ operations

DryadLINQ Compiler

Dryad Execution Engine

Directed Acyclic Graph (DAG) based execution flows

Job creation; Resource management; Fault tolerance& re-execution of failed tasks/vertices

SALSA

Reduce Phase of Particle Physics “Find the Higgs” using Dryad

• Combine Histograms produced by separate Root “Maps” (of event data to partial histograms) into a single Histogram delivered to Client

• This is an example using MapReduce to do distributed histogramming.

Higgs in Monte Carlo

SALSA

High Energy Physics Data Analysis

Input to a map task: <key, value> key = Some Id value = HEP file Name

Output of a map task: <key, value> key = random # (0<= num<= max reduce tasks)

value = Histogram as binary data

Input to a reduce task: <key, List<value>> key = random # (0<= num<= max reduce tasks)

value = List of histogram as binary data

Output from a reduce task: value value = Histogram file

Combine outputs from reduce tasks to form the final histogram

An application analyzing data from Large Hadron Collider (1TB but 100 Petabytes eventually)

SALSA

AWS/ Azure Hadoop DryadLINQ

Programming patterns

“Master-worker” paradigm

Independent job execution

MapReduce DAG execution, MapReduce + Other

patterns

Fault Tolerance Task re-execution based on a time out

Re-execution of failed and slow tasks.

Re-execution of failed and slow tasks.

Data Storage S3/Azure Storage. HDFS parallel file system. Local files

Environments EC2/Azure clouds, local compute resources

Linux cluster, Amazon Elastic MapReduce

Windows HPCS cluster

Ease of Programming

EC2 : **

Azure: ******* ****

Ease of use EC2 : ***

Azure: ***** ****

Scheduling & Load Balancing

Dynamic scheduling through a global queue,

Good natural load balancing

Data locality, rack aware dynamic task scheduling through a global queue,

Good natural load balancing

Data locality, network topology aware

scheduling. Static task partitions at the node level, suboptimal load

balancing

SALSA

Some Life Sciences Applications

• EST (Expressed Sequence Tag) sequence assembly program using DNA sequence assembly program software CAP3.

• Metagenomics and Alu repetition alignment using Smith Waterman dissimilarity computations followed by MPI applications for Clustering and MDS (Multi Dimensional Scaling) for dimension reduction before visualization

• Mapping the 60 million entries in PubChem into two or three dimensions to aid selection of related chemicals with convenient Google Earth like Browser. This uses either hierarchical MDS (which cannot be applied directly as O(N2)) or GTM (Generative Topographic Mapping).

• Correlating Childhood obesity with environmental factors by combining medical records with Geographical Information data with over 100 attributes using correlation computation, MDS and genetic algorithms for choosing optimal environmental factors.

SALSA

DNA Sequencing Pipeline

Illumina/Solexa Roche/454 Life Sciences Applied Biosystems/SOLiD

Modern Commerical Gene Sequences

Internet

Read Alignment

Visualization

PlotvizBlocking

Sequence

alignment

MDS

Dissimilarity

Matrix

N(N-1)/2 values

FASTA File

N Sequences

block

Pairings

Pairwise

clustering

MapReduce

MPI

• This chart illustrate our research of a pipeline mode to provide services on demand (Software as a Service SaaS) • User submit their jobs to the pipeline. The components are services and so is the whole pipeline.

SALSA

Alu and Metagenomics Workflow

“All pairs” problem Data is a collection of N sequences. Need to calcuate N2 dissimilarities (distances) between

sequnces (all pairs).

• These cannot be thought of as vectors because there are missing characters• “Multiple Sequence Alignment” (creating vectors of characters) doesn’t seem to work if N larger than

O(100), where 100’s of characters long.

Step 1: Can calculate N2 dissimilarities (distances) between sequencesStep 2: Find families by clustering (using much better methods than Kmeans). As no vectors, use vector

free O(N2) methodsStep 3: Map to 3D for visualization using Multidimensional Scaling (MDS) – also O(N2)

Results: N = 50,000 runs in 10 hours (the complete pipeline above) on 768 cores

Discussions:• Need to address millions of sequences …..• Currently using a mix of MapReduce and MPI• Twister will do all steps as MDS, Clustering just need MPI Broadcast/Reduce

SALSA

Biology MDS and Clustering Results

Alu Families

This visualizes results of Alu repeats from Chimpanzee and Human Genomes. Young families (green, yellow) are seen as tight clusters. This is projection of MDS dimension reduction to 3D of 35399 repeats – each with about 400 base pairs

Metagenomics

This visualizes results of dimension reduction to 3D of 30000 gene sequences from an environmental sample. The many different genes are classified by clustering algorithm and visualized by MDS dimension reduction

SALSA

All-Pairs Using DryadLINQ

0

5000

10000

15000

20000

35339 50000

DryadLINQ

MPI

Calculate Pairwise Distances (Smith Waterman Gotoh)

125 million distances4 hours & 46 minutes

• Calculate pairwise distances for a collection of genes (used for clustering, MDS)

• Fine grained tasks in MPI

• Coarse grained tasks in DryadLINQ

• Performed on 768 cores (Tempest Cluster)

Moretti, C., Bui, H., Hollingsworth, K., Rich, B., Flynn, P., & Thain, D. (2009). All-Pairs: An Abstraction for Data Intensive Computing on Campus Grids. IEEE Transactions on Parallel and Distributed Systems , 21, 21-36.

SALSA

Hadoop/Dryad ComparisonInhomogeneous Data I

1500

1550

1600

1650

1700

1750

1800

1850

1900

0 50 100 150 200 250 300

Tim

e (

s)

Standard Deviation

Randomly Distributed Inhomogeneous Data Mean: 400, Dataset Size: 10000

DryadLinq SWG Hadoop SWG Hadoop SWG on VM

Inhomogeneity of data does not have a significant effect when the sequence lengths are randomly distributedDryad with Windows HPCS compared to Hadoop with Linux RHEL on Idataplex (32 nodes)

SALSA

Hadoop/Dryad ComparisonInhomogeneous Data II

0

1,000

2,000

3,000

4,000

5,000

6,000

0 50 100 150 200 250 300

Tota

l Tim

e (

s)

Standard Deviation

Skewed Distributed Inhomogeneous dataMean: 400, Dataset Size: 10000

DryadLinq SWG Hadoop SWG Hadoop SWG on VM

This shows the natural load balancing of Hadoop MR dynamic task assignment using a global pipe line in contrast to the DryadLINQ static assignmentDryad with Windows HPCS compared to Hadoop with Linux RHEL on Idataplex (32 nodes)

DryadLINQ out performs Hadoop in other cases with data locality awareness

SALSA

Hadoop VM Performance Degradation

• 15.3% Degradation at largest data set size

10000 20000 30000 40000 50000

0%

5%

10%

15%

20%

25%

30%

No. of Sequences

Perf. Degradation On VM (Hadoop)

SALSA

Application Classes

1 Synchronous Lockstep Operation as in SIMD architectures SIMD

2 Loosely Synchronous

Iterative Compute-Communication stages with independent compute (map) operations for each CPU. Heart of most MPI jobs

MPP

3 Asynchronous Compute Chess; Combinatorial Search often supported by dynamic threads

MPP

4 Pleasingly Parallel Each component independent MPP, Grids, Clouds

5 Metaproblems Coarse grain (asynchronous) combinations of classes 1)-4). The preserve of workflow.

Grids, Clouds

6 MapReduce++ It describes file(database) to file(database) operations which has subcategories including.

1) Pleasingly Parallel Map Only (e.g. Cap3)2) Map followed by reductions (e.g. HEP)3) Iterative “Map followed by reductions” –

Extension of Current Technologies that supports much linear algebra and datamining

Clouds

Hadoop/Dryad

Twister

Classification of Parallel software/hardware use in terms of “Application architecture” Structures

SALSA

Applications & Different Interconnection PatternsMap Only Classic

MapReduceIterative Reductions

MapReduce++Loosely

Synchronous

CAP3 AnalysisDocument conversion(PDF -> HTML)Brute force searches in cryptographyParametric sweeps

High Energy Physics (HEP) HistogramsSWG gene alignmentDistributed searchDistributed sortingInformation retrieval

Expectation maximization algorithmsClusteringLinear Algebra

Many MPI scientific applications utilizingwide variety of communication constructs including local interactions

- CAP3 Gene Assembly- PolarGrid Matlab data analysis

- Information Retrieval -HEP Data Analysis- Calculation of Pairwise Distances for ALU Sequences

- Kmeans - Deterministic Annealing Clustering- Multidimensional Scaling MDS

- Solving Differential Equations and - particle dynamics with short range forces

Input

Output

map

Input

map

reduce

Input

map

reduce

iterations

Pij

Domain of MapReduce and Iterative Extensions MPI

SALSA

Twister(MapReduce++)• Streaming based communication• Intermediate results are directly

transferred from the map tasks to the reduce tasks – eliminates local files

• Cacheable map/reduce tasks• Static data remains in memory

• Combine phase to combine reductions• User Program is the composer of

MapReduce computations• Extends the MapReduce model to

iterative computationsData Split

D MRDriver

UserProgram

Pub/Sub Broker Network

D

File System

M

R

M

R

M

R

M

R

Worker Nodes

M

R

D

Map Worker

Reduce Worker

MRDeamon

Data Read/Write

Communication

Reduce (Key, List<Value>)

Iterate

Map(Key, Value)

Combine (Key, List<Value>)

User Program

Close()

Configure()Staticdata

δ flow

Different synchronization and intercommunication mechanisms used by the parallel runtimes

SALSA

Twister New Release

SALSA

Iterative Computations

K-meansMatrix

Multiplication

Performance of K-Means Parallel Overhead Matrix Multiplication

Smith Waterman

SALSA

Performance of Pagerank using

ClueWeb Data (Time for 20 iterations)

using 32 nodes (256 CPU cores) of Crevasse

SALSA

TwisterMPIReduce

• Runtime package supporting subset of MPI mapped to Twister

• Set-up, Barrier, Broadcast, Reduce

TwisterMPIReduce

PairwiseClusteringMPI

Multi Dimensional Scaling MPI

GenerativeTopographic Mapping

MPIOther …

Azure Twister (C# C++) Java Twister

Microsoft AzureFutureGrid Local

ClusterAmazon EC2

SALSA

Google MapReduce Apache Hadoop Microsoft Dryad Twister Azure Twister

Programming

Model

MapReduce MapReduce DAG execution,

Extensible to

MapReduce and

other patterns

Iterative

MapReduce

MapReduce-- will

extend to Iterative

MapReduce

Data Handling GFS (Google File

System)

HDFS (Hadoop

Distributed File

System)

Shared Directories &

local disks

Local disks

and data

management

tools

Azure Blob Storage

Scheduling Data Locality Data Locality; Rack

aware, Dynamic

task scheduling

through global

queue

Data locality;

Network

topology based

run time graph

optimizations; Static

task partitions

Data Locality;

Static task

partitions

Dynamic task

scheduling through

global queue

Failure Handling Re-execution of failed

tasks; Duplicate

execution of slow tasks

Re-execution of

failed tasks;

Duplicate execution

of slow tasks

Re-execution of failed

tasks; Duplicate

execution of slow

tasks

Re-execution

of Iterations

Re-execution of

failed tasks;

Duplicate execution

of slow tasks

High Level

Language

Support

Sawzall Pig Latin DryadLINQ Pregel has

related

features

N/A

Environment Linux Cluster. Linux Clusters,

Amazon Elastic

Map Reduce on

EC2

Windows HPCS

cluster

Linux Cluster

EC2

Window Azure

Compute, Windows

Azure Local

Development Fabric

Intermediate

data transfer

File File, Http File, TCP pipes,

shared-memory

FIFOs

Publish/Subscr

ibe messaging

Files, TCP

SALSA

High Performance Dimension Reduction and Visualization

• Need is pervasive

– Large and high dimensional data are everywhere: biology, physics, Internet, …

– Visualization can help data analysis

• Visualization of large datasets with high performance

– Map high-dimensional data into low dimensions (2D or 3D).

– Need Parallel programming for processing large data sets

– Developing high performance dimension reduction algorithms: • MDS(Multi-dimensional Scaling), used earlier in DNA sequencing application

• GTM(Generative Topographic Mapping)

• DA-MDS(Deterministic Annealing MDS)

• DA-GTM(Deterministic Annealing GTM)

– Interactive visualization tool PlotViz

• We are supporting drug discovery by browsing 60 million compounds in

PubChem database with 166 features each

SALSA

Dimension Reduction Algorithms

• Multidimensional Scaling (MDS) [1]o Given the proximity information among

points.

o Optimization problem to find mapping in target dimension of the given data based on pairwise proximity information while minimize the objective function.

o Objective functions: STRESS (1) or SSTRESS (2)

o Only needs pairwise distances ij between original points (typically not Euclidean)

o dij(X) is Euclidean distance between mapped (3D) points

• Generative Topographic Mapping (GTM) [2]o Find optimal K-representations for the given

data (in 3D), known as K-cluster problem (NP-hard)

o Original algorithm use EM method for optimization

o Deterministic Annealing algorithm can be used for finding a global solution

o Objective functions is to maximize log-likelihood:

[1] I. Borg and P. J. Groenen. Modern Multidimensional Scaling: Theory and Applications. Springer, New York, NY, U.S.A., 2005.[2] C. Bishop, M. Svens´en, and C. Williams. GTM: The generative topographic mapping. Neural computation, 10(1):215–234, 1998.

SALSA

GTM vs. MDS

MDS also soluble by viewing as nonlinear χ2 with iterative linear equation solver

GTM MDS (SMACOF)

Maximize Log-Likelihood Minimize STRESS or SSTRESSObjectiveFunction

O(KN) (K << N) O(N2)Complexity

• Non-linear dimension reduction• Find an optimal configuration in a lower-dimension• Iterative optimization method

Purpose

EM Iterative Majorization (EM-like)Optimization

Method

SALSA

MDS and GTM Example

37

Chemical compounds shown in literatures, visualized by MDS (left) and GTM (right)Visualized 234,000 chemical compounds which may be related with a set of 5 genes of interest (ABCB1, CHRNB2, DRD2, ESR1, and F2) based on the dataset collected from major journal literatures which is also stored in Chem2Bio2RDF system.

SALSA

Interpolation Method• MDS and GTM are highly memory and time consuming

process for large dataset such as millions of data points• MDS requires O(N2) and GTM does O(KN) (N is the number of

data points and K is the number of latent variables)• Training only for sampled data and interpolating for out-of-

sample set can improve performance• Interpolation is a pleasingly parallel application suitable for

MapReduce and Clouds

n in-sample

N-nout-of-sample

Total N data

Training

Interpolation

Trained data

Interpolated MDS/GTM

map

SALSA

Quality Comparison (O(N2) Full vs. Interpolation)

MDS

• Quality comparison between Interpolated result upto 100k based on the sample data (12.5k, 25k, and 50k) and original MDS result w/ 100k.

• STRESS:

wij = 1 / ∑δij2

GTM

Interpolation result (blue) is getting close to the original (red) result as sample size is increasing.

16 nodes

12.5K 25K 50K 100K Run on 16 nodes of Tempest

Note that we gain performance of over a factor of 100 for this data size. It would be more for larger data set.

SALSA

Summary of Initial Results

• Cloud technologies (Dryad/Hadoop/Azure/EC2) promising for Life Science computations

• Dynamic Virtual Clusters allow one to switch between different modes

• Overhead of VM’s on Hadoop (15%) acceptable

• Twister allows iterative problems (classic linear algebra/datamining) to use MapReduce model efficiently

– Prototype Twister released

• Dimension Reduction is important for visualization

SALSA

Convergence is Happening

Multicore

Clouds

Data IntensiveParadigms

Data intensive application with basic activities:capture, curation, preservation, and analysis (visualization)

Cloud infrastructure and runtime

Parallel threading and processes

SALSA

• Dynamic Virtual Cluster provisioning via XCAT• Supports both stateful and stateless OS images

iDataplex Bare-metal Nodes

Linux Bare-system

Linux Virtual Machines

Windows Server 2008 HPC

Bare-system Virtualization

Microsoft DryadLINQ / Twister / MPIApache Hadoop / Twister/ MPI

Smith Waterman Dissimilarities, CAP-3 Gene Assembly, PhyloD Using DryadLINQ, High Energy Physics, Clustering, Multidimensional Scaling,

Generative Topological Mapping

XCAT Infrastructure

Xen Virtualization

Applications

Runtimes

Infrastructure software

Hardware

Windows Server 2008 HPC

Science Cloud (Dynamic Virtual Cluster) Architecture

Services and Workflow

SALSA

Acknowledgements

SALSA Group

http://salsahpc.indiana.edu

Judy Qiu, Adam Hughes

Jaliya Ekanayake, Thilina Gunarathne, Jong Youl Choi, Seung-Hee Bae,

Yang Ruan, Hui Li, Bingjing Zhang, Saliya Ekanayake, Stephen Wu

Collaborators

Yves Brun, Peter Cherbas, Dennis Fortenberry, Roger Innes, David Nelson, Homer Twigg,

Craig Stewart, Haixu Tang, Mina Rho, David Wild, Bin Cao, Qian Zhu, Maureen Biggers, Gilbert Liu,

Neil Devadasan

Support by

Research Technologies of UITS and School of Informatics and Computing