Classification of Big Data Applications and Implications for

Classification of Big Data Applications and Implications for

Classification of Big Data Applications and Implications for the Algorithms and Software Needed for Scalable Data Analytics 70th Annual Meeting of the ORAU Council of Sponsoring Institutions March 4-5, 2015, Oak Ridge, Tennessee Big Data Analytics: Challenges and Opportunities March 4 2015 Geoffrey Fox [email protected] School of Informatics and Computing Digital Science Center Indiana University Bloomington 3/1/2015 1 HPC and Data Analytics/Software

Develop data analytics library SPIDAL (Scalable Parallel Interoperable Data Analytics Library ) of similar quality to PETSc and ScaLAPACK which have been very influential in success of HPC for simulations Approach: 1) Analyze Big Data applications to identify analytics needed and generate benchmark applications and characteristics (Ogres with facets) 2) Analyze existing analytics libraries (in practice limit to some application domains and some general libraries Mahout, R. MLlib) 3) Analyze Big Data Software and identify software model HPC-ABDS (HPC Apache Big Data Stack) to allow interoperability (Cloud/HPC) and high performance merging HPC and commodity cloud software 4) Identify range of big data computer architectures 5) Design or identify new or existing algorithms including parallel implementation Many more data scientists than computational scientists so HPC implications of data analytics could be influential on simulation software and hardware Develop Data Science Curricula 3/1/2015 2 IU Data Science Program Program managed by cross disciplinary Faculty in Data Science. Currently Statistics and Informatics and Computing School but will expand scope to full campus A purely online 4-course Certificate in Data Science has been running since January 2014 (with 100 students so far) Most students are professionals taking courses in free time Masters in Data Science (10 courses) approved October 2014

Online or Residential (Online masters is just $11,500 total) 80 students this semester and 150 applications for Fall 2015 A campus wide Ph.D. Minor in Data Science has been approved. Exploring PhD in Data Science Courses labelled as Decision-maker and Technical paths where McKinsey says an order of magnitude more (1.5 million by 2018) unmet job openings in Decision-maker track I teach big data courses; 70 undergraduates, 10 graduate students and 40 executive education enrolled this semester NIST Big Data Initiative Led by Chaitin Baru, Bob Marcus, Wo Chang 3/1/2015 4 NBD-PWG (NIST Big Data Public Working Group) Subgroups & Co-Chairs There were 5 Subgroups Note mainly industry Requirements and Use Cases Sub Group Geoffrey Fox, Indiana U.; Joe Paiva, VA; Tsegereda Beyene, Cisco Definitions and Taxonomies SG Nancy Grady, SAIC; Natasha Balac, SDSC; Eugene Luster, R2AD

Reference Architecture Sub Group Orit Levin, Microsoft; James Ketner, AT&T; Don Krapohl, Augmented Intelligence Security and Privacy Sub Group Arnab Roy, CSA/Fujitsu Nancy Landreville, U. MD Akhil Manchanda, GE Technology Roadmap Sub Group Carl Buffington, Vistronix; Dan McClary, Oracle; David Boyd, Data Tactics See And 3/1/2015 5 Use Case Template 26 fields completed for 51 areas Government Operation: 4 Commercial: 8 Defense: 3 Healthcare and Life Sciences: 10 Deep Learning and Social Media: 6 The Ecosystem for Research: 4 Astronomy and Physics: 5 Earth, Environmental and Polar Science: 10

Energy: 1 3/1/2015 6 51 Detailed Use Cases: Contributed July-September 2013 Covers goals, data features such as 3 Vs, software, hardware 26 Features for each use case (Section 5) Biased to science Government Operation(4): National Archives and Records Administration, Census Bureau Commercial(8): Finance in Cloud, Cloud Backup, Mendeley (Citations), Netflix, Web Search, Digital Materials, Cargo shipping (as in UPS) Defense(3): Sensors, Image surveillance, Situation Assessment Healthcare and Life Sciences(10): Medical records, Graph and Probabilistic analysis, Pathology, Bioimaging, Genomics, Epidemiology, People Activity models, Biodiversity

Deep Learning and Social Media(6): Driving Car, Geolocate images/cameras, Twitter, Crowd Sourcing, Network Science, NIST benchmark datasets The Ecosystem for Research(4): Metadata, Collaboration, Language Translation, Light source experiments Astronomy and Physics(5): Sky Surveys including comparison to simulation, Large Hadron Collider at CERN, Belle Accelerator II in Japan Earth, Environmental and Polar Science(10): Radar Scattering in Atmosphere, Earthquake, Ocean, Earth Observation, Ice sheet Radar scattering, Earth radar mapping, Climate simulation datasets, Atmospheric turbulence identification, Subsurface Biogeochemistry (microbes to watersheds), AmeriFlux and FLUXNET gas sensors 3/1/2015 7 Energy(1): Smart grid Application Example Montage Table 4: Characteristics of 6 Distributed Applications Execution Unit Communication Coordination Execution Environment Multiple sequential and parallel executable Multiple concurrent parallel executables Multiple seq. and parallel executables

Files Pub/sub Dataflow and events Climate Prediction (generation) Climate Prediction (analysis) SCOOP Multiple seq. & parallel executables Files and messages Multiple seq. & parallel executables Files and messages MasterWorker, events

Dataflow Coupled Fusion Multiple executable NEKTAR ReplicaExchange Multiple Executable Stream based Files and messages Stream-based Dataflow (DAG) Dataflow Dataflow Dataflow Dynamic process creation, execution Co-scheduling, data streaming, async. I/O

Decoupled coordination and messaging @Home (BOINC) Dynamics process creation, workflow execution Preemptive scheduling, reservations Co-scheduling, data streaming, async I/O Part of Property Summary Table 3/1/2015 8 Features and 2 Examples 3/1/2015 9 51 Use Cases: What is Parallelism Over? People: either the users (but see below) or subjects of application and often both

Decision makers like researchers or doctors (users of application) Items such as Images, EMR, Sequences below; observations or contents of online store Images or Electronic Information nuggets EMR: Electronic Medical Records (often similar to people parallelism) Protein or Gene Sequences; Material properties, Manufactured Object specifications, etc., in custom dataset Modelled entities like vehicles and people Sensors Internet of Things Events such as detected anomalies in telescope or credit card data or atmosphere (Complex) Nodes in RDF Graph Simple nodes as in a learning network Tweets, Blogs, Documents, Web Pages, etc. And characters/words in them

Files or data to be backed up, moved or assigned metadata Particles/cells/mesh points as in parallel simulations 3/1/2015 10 Features of 51 Use Cases I PP (26) All Pleasingly Parallel or Map Only MR (18) Classic MapReduce MR (add MRStat below for full count) MRStat (7) Simple version of MR where key computations are simple reduction as found in statistical averages such as histograms and averages MRIter (23) Iterative MapReduce or MPI (Spark, Twister) Graph (9) Complex graph data structure needed in analysis Fusion (11) Integrate diverse data to aid discovery/decision making; could involve sophisticated algorithms or could just be a portal Streaming (41) Some data comes in incrementally and is processed this way Classify (30) Classification: divide data into categories S/Q (12) Index, Search and Query 3/1/2015 11 Features of 51 Use Cases II CF (4) Collaborative Filtering for recommender engines LML (36) Local Machine Learning (Independent for each parallel entity) application could have GML as well

GML (23) Global Machine Learning: Deep Learning, Clustering, LDA, PLSI, MDS, Large Scale Optimizations as in Variational Bayes, MCMC, Lifted Belief Propagation, Stochastic Gradient Descent, L-BFGS, Levenberg-Marquardt . Can call EGO or Exascale Global Optimization with scalable parallel algorithm Workflow (51) Universal GIS (16) Geotagged data and often displayed in ESRI, Microsoft Virtual Earth, Google Earth, GeoServer etc. HPC (5) Classic large-scale simulation of cosmos, materials, etc. generating (visualization) data Agent (2) Simulations of models of data-defined macroscopic entities represented as agents 3/1/2015 12 13 Image-based Use Cases 13-15 Military Sensor Data Analysis/ Intelligence PP, LML, GIS, MR 7:Pathology Imaging/ Digital Pathology: PP, LML, MR for search becoming terabyte 3D images, Global Classification 18&35: Computational Bioimaging (Light Sources): PP, LML Also materials 26: Large-scale Deep Learning: GML Stanford ran 10 million images and 11 billion parameters on a 64 GPU HPC; vision (drive car), speech, and Natural Language Processing 27: Organizing large-scale, unstructured collections of photos: GML Fit position and camera direction to assemble 3D photo ensemble 36: Catalina Real-Time Transient Synoptic Sky Survey (CRTS): PP, LML followed by classification of events (GML)

43: Radar Data Analysis for CReSIS Remote Sensing of Ice Sheets: PP, LML to identify glacier beds; GML for full ice-sheet 44: UAVSAR Data Processing, Data Product Delivery, and Data Services: PP to find slippage from radar images 45, 46: Analysis of Simulation visualizations: PP LML ?GML find paths, classify orbits, classify patterns that signal earthquakes, instabilities, climate, turbulence 3/1/2015 13 Internet of Things and Streaming Apps It is projected that there will be 24 (Mobile Industry Group) to 50 (Cisco) billion devices on the Internet by 2020. The cloud natural controller of and resource provider for the Internet of Things. Smart phones/watches, Wearable devices (Smart People), Intelligent River Smart Homes and Grid and Ubiquitous Cities, Robotics. Majority of use cases are streaming experimental science gathers data in a stream sometimes batched as in a field trip. Below is sample 10: Cargo Shipping Tracking as in UPS, Fedex PP GIS LML 13: Large Scale Geospatial Analysis and Visualization PP GIS LML 28: Truthy: Information diffusion research from Twitter Data PP MR for Search, GML for community determination 39: Particle Physics: Analysis of LHC Large Hadron Collider Data: Discovery of Higgs particle PP Local Processing Global statistics 50: DOE-BER AmeriFlux and FLUXNET Networks PP GIS LML 3/1/2015 14

51: Consumption forecasting in Smart Grids PP GIS LML Big Data Patterns the Ogres 3/1/2015 15 7 Computational Giants of NRC Massive Data Analysis Report 1) 2) 3) 4) 5) 6) 7) G1: G2: G3: G4: G5: G6: G7:

3/1/2015 Basic Statistics e.g. MRStat Generalized N-Body Problems Graph-Theoretic Computations Linear Algebraic Computations Optimizations e.g. Linear Programming Integration e.g. LDA and other GML Alignment Problems e.g. BLAST 16 HPC Benchmark Classics Linpack or HPL: Parallel LU factorization for solution of linear equations NPB version 1: Mainly classic HPC solver kernels MG: Multigrid CG: Conjugate Gradient FT: Fast Fourier Transform IS: Integer sort EP: Embarrassingly Parallel BT: Block Tridiagonal SP: Scalar Pentadiagonal LU: Lower-Upper symmetric Gauss Seidel 3/1/2015 17 13 Berkeley Dwarfs

1) Dense Linear Algebra 2) Sparse Linear Algebra 3) Spectral Methods 4) N-Body Methods 5) Structured Grids 6) Unstructured Grids 7) MapReduce 8) Combinational Logic 9) Graph Traversal 10)Dynamic Programming 11)Backtrack and Branch-and-Bound 12)Graphical Models 3/1/2015 13)Finite State Machines First 6 of these correspond to Colellas original. Monte Carlo dropped. N-body methods are a subset of Particle in Colella. Note a little inconsistent in that MapReduce is a programming model and spectral method is a numerical method. Need multiple facets! 18

Facets of the Ogres 3/1/2015 19 Big Data Ogres and their Facets Big Data Ogres are an attempt to characterize applications and algorithms with a set of general common features that are called Facets Originally derived from NIST collection of 51 use cases but refined with experience The 50 facets capture common characteristics (shared by several problems)which are inevitably multi-dimensional and often overlapping. Divided into 4 views One view of an Ogre is the overall problem architecture which is naturally related to the machine architecture needed to support data intensive application. The execution (computational) features view, describes issues such as I/O versus compute rates, iterative nature and regularity of computation and the classic Vs of Big Data: defining problem size, rate of change, etc. The data source & style view includes facets specifying how the data is collected, stored and accessed. Has classic database characteristics Processing view has facets which describe types of processing steps including nature of algorithms and kernels e.g. Linear Programming, Learning, Maximum Likelihood Instances of Ogres are particular big data problems and a set of Ogre instances that cover enough of the facets could form a comprehensive benchmark/mini-app

set Problem Architecture View of Ogres (Meta or MacroPatterns) i. Pleasingly Parallel as in BLAST, Protein docking, some (bio-)imagery including Local Analytics or Machine Learning ML or filtering pleasingly parallel, as in bio-imagery, radar images (pleasingly parallel but sophisticated local analytics) ii. Classic MapReduce: Search, Index and Query and Classification algorithms like collaborative filtering (G1 for MRStat in Features, G7) iii. Map-Collective: Iterative maps + communication dominated by collective operations as in reduction, broadcast, gather, scatter. Common datamining pattern iv. Map-Point to Point: Iterative maps + communication dominated by many small point to point messages as in graph algorithms v. Map-Streaming: Describes streaming, steering and assimilation problems vi. Shared Memory: Some problems are asynchronous and are easier to parallelize on shared rather than distributed memory see some graph algorithms vii. SPMD: Single Program Multiple Data, common parallel programming feature viii. BSP or Bulk Synchronous Processing: well-defined compute-communication phases

ix. Fusion: Knowledge discovery often involves fusion of multiple methods. x. Dataflow: Important application features often occurring in composite Ogres xi. Use Agents: as in epidemiology (swarm approaches) xii. Workflow: All applications often involve orchestration (workflow) of multiple components Note problem and machine architectures are related 3/1/2015 21 Hardware, Software, Applications In my old papers (especially book Parallel Computing Works!), I discussed computing as multiple complex systems mapped into each other Problem Numerical formulation Software Hardware Each of these 4 complex systems has an architecture that can be described in similar language One gets an easy programming model if architecture of problem matches that of Software One gets good performance if architecture of hardware matches that of software and problem So MapReduce can be used as architecture of software (programming model) or Numerical formulation of problem 1/26/2015 22 (1) Map Only

6 Forms of MapReduce Input PP Local Analytics (3) Iterative Map Reduce (2) Classic or Map-Collective M MapReduce Input Iterations Input MR Basic Statistics map map reduce Output ap Reduce (4) Point to Point or llective Map-Communication ations

map (5) Map Streaming maps brokers reduce Iterative (6) Shared memory Map Communicates Shared Memory Map & Communicate Local Graph 1/26/2015 Graph Streaming Events Shared Memory

23 8 Data Analysis Problem Architectures 1) Pleasingly Parallel PP or map-only in MapReduce BLAST Analysis; Local Machine Learning 2A) Classic MapReduce MR, Map followed by reduction High Energy Physics (HEP) Histograms; Web search; Recommender Engines 2B) Simple version of classic MapReduce MRStat Final reduction is just simple statistics 3) Iterative MapReduce MRIter Expectation maximization Clustering Linear Algebra, PageRank 4A) Map Point to Point Communication Classic MPI; PDE Solvers and Particle Dynamics; Graph processing Graph 4B) GPU (Accelerator) enhanced 4A) especially for deep learning 5) Map + Streaming + Communication Images from Synchrotron sources; Telescopes; Internet of Things IoT 6) Shared memory allowing parallel threads which are tricky to program but lower latency Difficult to parallelize asynchronous parallel Graph Algorithms 1/26/2015 24

There are a lot of Big Data and HPC Software systems in 17 (21) layers Build on do not compete with the 293 HPC-ABDS systems Kaleidoscope of (Apache) Big Data Stack (ABDS) and HPC Technologies CrossCutting Functions 1) Message and Data Protocols: Avro, Thrift, Protobuf 2) Distributed Coordination: Zookeeper, Giraffe, JGroups 17) Workflow-Orchestration: ODE, ActiveBPEL, Airavata, Pegasus, Kepler, Swift, Taverna, Triana, Trident, BioKepler, Galaxy, IPython, Dryad, Naiad, Oozie, Tez, Google FlumeJava, Crunch, Cascading, Scalding, e-Science Central, Azure Data Factory, Google Cloud Dataflow, NiFi (NSA) 16) Application and Analytics: Mahout , MLlib , MLbase, DataFu, R, pbdR, Bioconductor, ImageJ, Scalapack, PetSc, Azure Machine Learning, Google Prediction API, Google Translation API, mlpy, scikit-learn, PyBrain, CompLearn, Caffe, Torch, Theano, H2O, IBM Watson, Oracle PGX, GraphLab, GraphX, IBM System G, GraphBuilder(Intel), TinkerPop, Google Fusion Tables, CINET, NWB, Elasticsearch 15B) Frameworks: Google App Engine, AppScale, Red Hat OpenShift, Heroku, Aerobatic, AWS Elastic Beanstalk, Azure, Cloud Foundry, Pivotal, IBM BlueMix, Ninefold, Jelastic, Stackato, appfog, CloudBees, Engine Yard, CloudControl, dotCloud, Dokku, OSGi, HUBzero, OODT 15A) High level Programming: Kite, Hive, HCatalog, Tajo, Shark, Phoenix, Impala, MRQL, SAP HANA, HadoopDB, PolyBase, Presto, Google Dremel, Google BigQuery, Amazon Redshift, Drill, Pig, Sawzall, Google Cloud DataFlow, Summingbird 14B) Streams: Storm, S4, Samza, Google MillWheel, Amazon Kinesis, LinkedIn Databus, Facebook Scribe/ODS, Azure Stream Analytics 14A) Basic Programming model and runtime, SPMD, MapReduce: Hadoop, Spark, Twister, Stratosphere (Apache Flink), Reef, Hama, Giraph,

Pregel, Pegasus 13) Inter process communication Collectives, point-to-point, publish-subscribe: Harp, MPI, Netty, ZeroMQ, ActiveMQ, RabbitMQ, QPid, Kafka, Kestrel, JMS, AMQP, Stomp, MQTT, Azure Event Hubs, Amazon Lambda Public Cloud: Amazon SNS, Google Pub Sub, Azure Queues 12) In-memory databases/caches: Gora (general object from NoSQL), Memcached, Redis (key value), Hazelcast, Ehcache, Infinispan 12) Object-relational mapping: Hibernate, OpenJPA, EclipseLink, DataNucleus, ODBC/JDBC 12) Extraction Tools: UIMA, Tika 11C) SQL(NewSQL): Oracle, DB2, SQL Server, SQLite, MySQL, PostgreSQL, SciDB, Apache Derby, Google Cloud SQL, Azure SQL, Amazon RDS, rasdaman, BlinkDB, N1QL, Galera Cluster, Google F1, IBM dashDB 11B) NoSQL: HBase, Accumulo, Cassandra, Solandra, MongoDB, CouchDB, Lucene, Solr, Berkeley DB, Riak, Voldemort, Neo4J, Yarcdata, Jena, Sesame, AllegroGraph, RYA, Espresso, Sqrrl, Facebook Tao, Google Megastore, Google Spanner, Titan:db, IBM Cloudant Public Cloud: Azure Table, Amazon Dynamo, Google DataStore 4) 11A) File management: iRODS, NetCDF, CDF, HDF, OPeNDAP, FITS, RCFile, ORC, Parquet Monitoring: 10) Data Transport: BitTorrent, HTTP, FTP, SSH, Globus Online (GridFTP), Flume, Sqoop Ambari, 9) Cluster Resource Management: Mesos, Yarn, Helix, Llama, Celery, HTCondor, SGE, OpenPBS, Moab, Slurm, Torque, Google Omega, Ganglia, Facebook Corona Nagios, Inca 8) File systems: HDFS, Swift, Cinder, Ceph, FUSE, Gluster, Lustre, GPFS, GFFS, Haystack, f4 Public Cloud: Amazon S3, Azure Blob, Google Cloud Storage 21 layers 7) Interoperability: Whirr, JClouds, OCCI, CDMI, Libcloud, TOSCA, Libvirt 6) DevOps: Docker, Puppet, Chef, Ansible, Boto, Cobbler, Xcat, Razor, CloudMesh, Juju, Foreman, OpenStack Heat, Rocks, Cisco Intelligent 293 Automation for Cloud, Ubuntu MaaS, Facebook Tupperware, AWS OpsWorks, OpenStack Ironic, Google Kubernetes, Buildstep, Gitreceive 5) IaaS Management from HPC to hypervisors: Xen, KVM, Hyper-V, VirtualBox, OpenVZ, LXC, Linux-Vserver, VMware ESXi, vSphere,

Software 3/1/2015OpenStack, OpenNebula, Eucalyptus, Nimbus, CloudStack, VMware vCloud, Amazon, Azure, Google and other public Clouds, 25 Packages Networking: Google Cloud DNS, Amazon Route 53 3) Security & Privacy: InCommon, OpenStack Keystone, LDAP, Sentry, Sqrrl 1/26/2015 26 i. ii. iii. One View of Ogres has Facets that are micropatterns or Execution Features Performance Metrics; property found by benchmarking Ogre Flops per byte; memory or I/O Execution Environment; Core libraries needed: matrix-matrix/vector algebra, conjugate gradient, reduction, broadcast; Cloud, HPC etc. iv. Volume: property of an Ogre instance v.

Velocity: qualitative property of Ogre with value associated with instance vi. Variety: important property especially of composite Ogres vii. Veracity: important property of mini-applications but not kernels viii. Communication Structure; Interconnect requirements; Is communication BSP, Asynchronous, Pub-Sub, Collective, Point to Point? ix. Is application (graph) static or dynamic? x. Most applications consist of a set of interconnected entities; is this regular as a set of pixels or is it a complicated irregular graph? xi. Are algorithms Iterative or not? xii. Data Abstraction: key-value, pixel, graph(G3), vector, bags of words or items xiii. Are data points in metric or non-metric spaces? xiv. Is algorithm O(N2) or O(N) (up to logs) for N points per iteration (G2) 3/1/2015 27 Data Source and Style View of Ogres I i. ii. iii. iv. v. SQL NewSQL or NoSQL: NoSQL includes Document, Column, Key-value, Graph, Triple store; NewSQL is SQL redone to exploit NoSQL performance

Other Enterprise data systems: 10 examples from NIST integrate SQL/NoSQL Set of Files or Objects: as managed in iRODS and extremely common in scientific research File systems, Object, Blob and Data-parallel (HDFS) raw storage: Separated from computing or colocated? HDFS v Lustre v. Openstack Swift v. GPFS Archive/Batched/Streaming: Streaming is incremental update of datasets with new algorithms to achieve real-time response (G7); Before data gets to compute system, there is often an initial data gathering phase which is characterized by a block size and timing. Block size varies from month (Remote Sensing, Seismic) to day 3/1/2015 (genomic) to seconds or lower (Real time control, streaming) 28 Data Source and Style View of Ogres II vi. Shared/Dedicated/Transient/Permanent: qualitative property of data; Other characteristics are needed for permanent auxiliary/comparison datasets and these could be interdisciplinary, implying nontrivial data movement/replication vii. Metadata/Provenance: Clear qualitative property but not for kernels as important aspect of data collection process viii. Internet of Things: 24 to 50 Billion devices on Internet by 2020 ix. HPC simulations: generate major (visualization) output that often needs to be mined x. Using GIS: Geographical Information Systems provide attractive access to geospatial data Note 10 Bob Marcus (lead NIST effort) access examples illustrate this

3/1/2015 29 2. Perform real time analytics on data source streams and notify users when specified events occur Specify filter Filter Identifying Events Streaming Data Streaming Data Streaming Data Post Selected Events Fetch streamed Data Posted Data Identified Events Archive Repository 3/1/2015 Storm, Kafka, Hbase, Zookeeper 30

5A. Perform interactive analytics on observational scientific data Science Analysis Code, Mahout, R Grid or Many Task Software, Hadoop, Spark, Giraph, Pig Data Storage: HDFS, Hbase, File Collection Direct Transfer Streaming Twitter data for Social Networking Record Scientific Data in field 3/1/2015 Transport batch of data to primary analysis data system Local Accumulate and initial computing NIST examples include LHC, Remote Sensing, Astronomy and Bioinformatics 31

Facets in Processing (run time) View of Ogres I i. Micro-benchmarks ogres that exercise simple features of hardware such as communication, disk I/O, CPU, memory performance ii. Local Analytics executed on a single core or perhaps node iii. Global Analytics requiring iterative programming models (G5,G6) across multiple nodes of a parallel system iv. Optimization Methodology: overlapping categories i. ii. iii. iv. v. vi. vii. v. Nonlinear Optimization (G6) Machine Learning Maximum Likelihood or 2 minimizations Expectation Maximization (often Steepest descent) Combinatorial Optimization Linear/Quadratic Programming (G5) Dynamic Programming Visualization is key application capability with algorithms like MDS useful but it itself part of mini-app or composite Ogre

vi. 3/1/2015 Alignment (G7) as in BLAST compares samples with repository 32 Facets in Processing (run time) View of Ogres II vii. Streaming divided into 5 categories depending on event size and synchronization and integration Set of independent events where precise time sequencing unimportant. Time series of connected small events where time ordering important. Set of independent large events where each event needs parallel processing with time sequencing not critical Set of connected large events where each event needs parallel processing with time sequencing critical. Stream of connected small or large events to be integrated in a complex way. viii. Basic Statistics (G1): MRStat in NIST problem features ix. Search/Query/Index: Classic database which is well studied (Baru, Rabl tutorial) x. Recommender Engine: core to many e-commerce, media businesses; collaborative filtering key technology xi. Classification: assigning items to categories based on many methods MapReduce good in Alignment, Basic statistics, S/Q/I, Recommender, Calssification

xii. Deep Learning of growing importance due to success in speech recognition etc. xiii. Problem set up as a graph (G3) as opposed to vector, grid, bag of words etc. 33 xiv. 3/1/2015 Using Linear Algebra Kernels: much machine learning uses linear algebra kernels Data Source and Style View 6 5 4 3 2 1 3 2 1 HDFS/Lustre/GPFS Files/Objects Enterprise Data Model SQL/NoSQL/NewSQL 4 Ogre Views and

50 Facets Pleasingly Parallel Classic MapReduce Map-Collective Map Point-to-Point Map Streaming Shared Memory Single Program Multiple Data Bulk Synchronous Parallel Fusion Problem Dataflow Agents Architecture Workflow View Geospatial Information System HPC Simulations Internet of Things Metadata/Provenance Shared / Dedicated / Transient / Permanent Archived/Batched/Streaming 1 2 3

4 5 6 7 8 9 10 11 12 Execution View 1 2 3 4 5 6 7 8 9 10 11 12 13 14 = NN / = N Metric = M / Non-Metric = N Data Abstraction Iterative / Simple Regular = R / Irregular = I Dynamic = D / Static = S Communication Structure Veracity Variety Velocity Volume Execution Environment; Core libraries

Flops per Byte; Memory I/O Performance Metrics 7 Micro-benchmarks Local Analytics Global Analytics Base Statistics Processing View 8 Recommendations Search / Query / Index Classification Learning Optimization Methodology Streaming Alignment Linear Algebra Kernels Graph Algorithms Visualization 14 13 12 11 10 9

10 9 8 7 6 5 4 Benchmarks based on Ogres 3/1/2015 35 Benchmarks/Mini-apps spanning Facets Look at NSF SPIDAL Project, NIST 51 use cases, Baru-Rabl review Catalog facets of benchmarks and choose entries to cover all facets Micro Benchmarks: SPEC, EnhancedDFSIO (HDFS), Terasort, Wordcount, Grep, MPI, Basic Pub-Sub . SQL and NoSQL Data systems, Search, Recommenders: TPC (-C to xHS for Hadoop), BigBench, Yahoo Cloud Serving, Berkeley Big Data, HiBench, BigDataBench, Cloudsuite, Linkbench includes MapReduce cases Search, Bayes, Random Forests, Collaborative Filtering Spatial Query: select from image or earth data Alignment: Biology as in BLAST Streaming: Online classifiers, Cluster tweets, Robotics, Industrial Internet of Things, Astronomy; BGBenchmark; choose to cover all 5 subclasses Pleasingly parallel (Local Analytics): as in initial steps of LHC, Pathology,

Bioimaging (differ in type of data analysis) Global Analytics: Outlier, Clustering, LDA, SVM, Deep Learning, MDS, PageRank, Levenberg-Marquardt, Graph 500 entries Workflow and Composite (analytics on xSQL) linking above Parallel Data Analytics Issues 3/1/2015 37 Remarks on Parallelism I Most use parallelism over items in data set Entities to cluster or map to Euclidean space Except deep learning (for image data sets)which has parallelism over pixel plane in neurons not over items in training set as need to look at small numbers of data items at a time in Stochastic Gradient Descent SGD Need experiments to really test SGD as no easy to use parallel implementations tests at scale NOT done Maybe got where they are as most work sequential Maximum Likelihood or 2 both lead to structure like Minimize sum items=1N (Positive nonlinear function of unknown parameters for item i) All solved iteratively with (clever) first or second order approximation to shift in objective function

Sometimes steepest descent direction; sometimes Newton 11 billion deep learning parameters; Newton impossible Have classic Expectation Maximization structure Steepest descent shift is sum over shift calculated from each point SGD take randomly a few hundred of items in data set and calculate shifts over these and move a tiny distance 3/1/2015 38 Remarks on Parallelism II Need to cover non vector semimetric and vector spaces for clustering and dimension reduction (N points in space) MDS Minimizes Stress (X) = i

Note matrix solvers all use conjugate gradient converges in 5-100 iterations a big gain for matrix with a million rows. This removes factor of N in time complexity Ratio of #clusters to #points important; new ideas if ratio >~ 0.1 3/1/2015 39 Algorithm Challenges See NRC Massive Data Analysis report O(N) algorithms for O(N2) problems Parallelizing Stochastic Gradient Descent Streaming data algorithms balance and interplay between batch methods (most time consuming) and interpolative streaming methods Graph algorithms Machine Learning Community uses parameter servers; Parallel Computing (MPI) would not recommend this? Is classic distributed model for parameter service better? Apply best of parallel computing communication and load balancing to Giraph/Hadoop/Spark Are data analytics sparse?; many cases are full matrices BTW Need Java Grande Some C++ but Java most popular in ABDS,

with Python, Erlang, Go, Scala (compiles to JVM) .. 3/1/2015 40 Lessons / Insights Proposed classification of Big Data applications and Benchmarks with features generalized as facets Data intensive algorithms do not have the well developed high performance libraries familiar from HPC Global Machine Learning or (Exascale Global Optimization) particularly challenging Develop SPIDAL (Scalable Parallel Interoperable Data Analytics Library) New algorithms and new high performance parallel implementations Challenges with O(N2) problems Integrate (dont compete) HPC with Commodity Big data (Google to Amazon to Enterprise/Startup Data Analytics) i.e. improve Mahout; dont compete with it Use Hadoop plug-ins rather than replacing Hadoop Enhanced Apache Big Data Stack HPC-ABDS has ~290 members with HPC opportunities at Resource management, Storage/Data, Streaming, 3/1/2015 41 Programming, monitoring, workflow layers.

Recently Viewed Presentations

  • Basic Nursing: Foundations of Skills and Concepts Chapter 24
  • Strategic Planning in Public Health - University of Pittsburgh

    Strategic Planning in Public Health - University of Pittsburgh

    The Big Picture: Strategic Planning in Public Health TH Tulchinsky MD MPH Braun School of Public Health Hebrew University-Hadassah, Jerusalem Skopje, Macedonia 25 Oct 2010 Strategy and Tactics Strategy - A plan or method or series of maneuvers or stratagems...
  • Small Feature Reproducibility Measuring, Understanding and ...

    Small Feature Reproducibility Measuring, Understanding and ...

    Small Feature Reproducibility Small Feature Reproducibility Measuring, Understanding and Controlling Variability in Deep Sub-micron Patterning UC-SMART Major Program Award E. Aydil, J. Bokor, N. Cheung, D. Dornfeld, F. Doyle, B. Dunn, D. Graves, E. Haller, M. Lieberman, A. Neureuther, K....
  • ILC Test Facility at New Muon Lab - Fermilab

    ILC Test Facility at New Muon Lab - Fermilab

    ILC/SRF R&D Robert Kephart Outline Fermilab ILC Goals Fermilab's role in the GDE & the ILC machine design Main Linac design Accelerator physics Main Linac components (Cryomodule, RF components) ILC/SRF R&D and Infrastructure Cavity fabrication and processing Cavity test 3.9...
  • Glenn York Elementary PTO

    Glenn York Elementary PTO

    Henna by Kylie. Glossy Glam Girl - Lipsense - Colleen Stewart. Color Street TX Girls. ... The Dickson Family. The Stewart Family. The Sansalone Family. The Mata Family. The Montgomery Family. The Flores Family. The Griffin Family. The Robinson Family....
  • Self Organizing Maps

    Self Organizing Maps

    A histogram is formed based on the hits on WCM. Use this histogram as fingerprint for DCM. References [1] T.Honkela, S.Kaski, K.Lagus,T.Kohonen. WEBSOM- Self Organizing Maps of Document Collection. (1997) [2] T.Honkela, S.Kaski, K.Lagus,T.Kohonen. Exploration of full text Databases with...
  • Schenker's Conception of Melody

    Schenker's Conception of Melody

    Compound melody or polyphonic melody often suggests long-range voice-leading connections between nonadjacent tones. 14. To Schenker, a melodic leap is often not a mere skip, but a temporary switch to an inner voice. ^3 ^2 ^1 15. Upward leaps can...
  • Getting the reading habit -

    Getting the reading habit -

    Whatisextensivereading? readinglots of differenttexts. reading a variety of texts. readingforpleasure. readingwithout a task, a test or a set of exercises