Cyberinfrastructure Technologies and Applications Summit on Cyberinfrastructure Innovation
Cyberinfrastructure Technologies and Applications Summit on Cyberinfrastructure: Innovation At Work Banff Springs Hotel Banff Canada October 11 2007 Geoffrey Fox Computer Science, Informatics, Physics Pervasive Technology Laboratories Indiana University Bloomington IN 47401 http: //grids. ucs. indiana. edu/ptliupages/presentations/ gcf@indiana. edu http: //www. infomall. org 1
e-moreorlessanything ‘e-Science is about global collaboration in key areas of science, and the next generation of infrastructure that will enable it. ’ from its inventor John Taylor Director General of Research Councils UK, Office of Science and Technology e-Science is about developing tools and technologies that allow scientists to do ‘faster, better or different’ research Similarly e-Business captures an emerging view of corporations as dynamic virtual organizations linking employees, customers and stakeholders across the world. This generalizes to e-moreorlessanything including presumably e. Alberta. Enterprise and e-oilandgas, e-geoscience …. A deluge of data of unprecedented and inevitable size must be managed and understood. People (see Web 2. 0), computers, data (including sensors and instruments) must be linked. On demand assignment of experts, computers, networks and storage resources must be supported 2
What is Cyberinfrastructure is (from NSF) infrastructure that supports distributed science (e-Science)– data, people, computers • Clearly core concept more general than Science Exploits Internet technology (Web 2. 0) adding (via Grid technology) management, security, supercomputers etc. It has two aspects: parallel – low latency (microseconds) between nodes and distributed – highish latency (milliseconds) between nodes Parallel needed to get high performance on individual large simulations, data analysis etc. ; must decompose problem Distributed aspect integrates already distinct components – especially natural for data Cyberinfrastructure is in general a distributed collection of parallel systems Cyberinfrastructure is made of services (originally Web services) that are “just” programs or data sources packaged for distributed access 3
Underpinnings of Cyberinfrastructure Distributed software systems are being “revolutionized” by developments from e-commerce, e-Science and the consumer Internet. There is rapid progress in technology families termed “Web services”, “Grids” and “Web 2. 0” The emerging distributed system picture is of distributed services with advertised interfaces but opaque implementations communicating by streams of messages over a variety of protocols • Complete systems are built by combining either services or predefined/pre-existing collections of services together to achieve new capabilities As well as Internet/Communication revolutions (distributed systems), multicore chips will likely be hugely important (parallel systems) Industry not academia is leading innovation in these technologies 4
Service or Web Service Approach One uses GML, CML etc. to define the data structure in a system and one uses services to capture “methods” or “programs” In e. Science, important services fall in three classes • Simulations • Data access, storage, federation, discovery • Filters for data mining and manipulation Services could use something like WSDL (Web Service Definition Language) to define interoperable interfaces but Web 2. 0 follows old library practice: one just specifies interface Service Interface (WSDL) establishes a “contract” independent of implementation between two services or a service and a client Services should be loosely coupled which normally means they are coarse grain Services will be composed (linked together) by mashups (typically scripts) or workflow (often XML – BPEL) Software Engineering and Interoperability/Standards are closely related 5
Tera. Grid resources include more than 250 teraflops of computing capability and more than 30 petabytes of online and archival data storage, with rapid access and retrieval over high-performance networks. Tera. Grid is coordinated at the University of Chicago, working with the Resource Provider sites: Indiana University, Oak Ridge National Laboratory, National Center for Supercomputing Applications, Pittsburgh Supercomputing Center, Purdue University, San Diego Supercomputer Center, Texas Advanced Computing Center, University of Chicago/Argonne National Laboratory, and the National Center for Atmospheric Research. UW Grid Infrastructure Group (UChicago) PSC UC/ANL NCAR PU NCSA Caltech IU UNC/RENCI ORNL USC/ISI SDSC TACC Resource Provider (RP) Software Integration Partner Computing and Cyberinfrastructure: Tera. Grid
Data and Cyberinfrastructure DIKW: Data Information Knowledge Wisdom transformation Applies to e-Science, Distributed Business Enterprise (including outsourcing), Military Command Control and general decision support (SOAP or just RSS) messages transport information expressed in a semantically rich fashion between sources and services that enhance and transform information so that complete system provides • Semantic Web technologies like RDF and OWL might help us to have rich expressivity but they might be too complicated We are meant to build application specific information management/transformation systems for each domain • Each domain has Specific Services/Standards (for API’s and Information such as KML and GML for Geographical Information Systems) • and will use Generic Services (like R for datamining) and • Generic Standards (such as RDF, WSDL) Standards made before consensus or not observant of technology progress are dubious 7
Raw Dataand Information Knowledge Wisdom Information Cyberinfrastructure Another Decisions Grid SS SS FS OS OS FS FS SS FS FS MD SS SS FS F S ge s OS FS MD OS FS SS Other Service OS OS FS SS M es sa FS MD SS al MD Filter Service SS ce rt SS SS Meta. Data Sensor Service SS SS rv i Po MD MD SS Se F S OS FS r- OS FS MD FS SS te FS OS Another Grid In MD FS SS SS FS MD SS Another Service FS OS SS SS Another Grid Database Another Service 8
Information Cyberinfrastructure Architecture The Party Line approach to Information Infrastructure is clear – one creates a Cyberinfrastructure consisting of distributed services accessed by portals/gadgets/gateways/RSS feeds Services include: • Computing • “original data” • Transformations or filters implementing DIKW (Data Information Knowledge Wisdom) pipeline • Final “Decision Support” step converting wisdom into action • Generic services such as security, profiles etc. Some filters could correspond to large simulations Infrastructure will be set up as a System of Systems (Grids of Grids) • Services and/or Grids just accept some form of DIKW and produce another form of DIKW • “Original data” has no explicit input; just output 9
Virtual Observatory Astronomy Grid Integrate Experiments Radio Far-Infrared Visible Dust Map Visible + X-ray Galaxy Density Map 10
11 CYBERINFRASTRUCTURE CENTER FOR PO L A R SCIENCE (CICPS)
CRe. SIS Polar. Grid • Important CRe. SIS-specific Cyberinfrastructure components include – Managed data from sensors and satellites – Data analysis such as SAR processing – possibly with parallel algorithms – Electromagnetic simulations (currently commercial codes) to design instrument antennas – 3 D simulations of ice-sheets (glaciers) with non-uniform meshes – GIS Geographical Information Systems • Also need capabilities present in many Grids – Portal i. e. Science Gateway – Submitting multiple sequential or parallel jobs • The need for three distinct types of components: Continental USA with multiple base and field camps – Base and field camps must be power efficient – Terrible connectivity from base and field camps to Continental sub. Grid 12
CICC Chemical Informatics and Cyberinfrastructure Collaboratory Web Service Infrastructure OSCAR Document Analysis In. Ch. I Generation/Search Computational Chemistry (Gamess, Jaguar etc. ) Core Grid Services Service Registry Job Submission and Management Local Clusters IU Big Red, Tera. Grid, Open Science Grid Varuna. net Quantum Chemistry Portal Services RSS Feeds User Profiles Collaboration as in Sakai
Process Chemistry-Biology Interaction Data from HTS (High Throughput Screening) Percent Inhibition or IC 50 data is retrieved from HTS Question: Was this screen successful? Scientists at IU prefer Web 2. 0 to Grid/Web Service for workflow Workflows encoding plate & control well statistics, distribution analysis, etc Question: What should the active/inactive cutoffs be? Workflows encoding distribution analysis of screening results Question: What can we learn about the target protein or cell line from this screen? Workflows encoding statistical comparison of results to similar screens, docking of compounds into proteins to correlate binding, with activity, literature search of active compounds, etc Compound data submitted to Pub. Chem PROCESS CHEMINFORMATICS Grids can link data analysis ( e. g image processing developed in existing Grids), traditional Cheminformatics tools, as well as annotation tools (Semantic Web, del. icio. us) and enhance lead ID and SAR analysis A Grid of Grids linking collections of services at Pub. Chem ECCR centers 14 MLSCN centers GRIDS
People and Cyberinfrastructure: Web 2. 0 has tools (sites) and technologies • Technologies (later) are “competition” for Grids and Web Services • Sites (below) are the best way to integrate people into Cyberinfrastructure Kazaa, Instant Messengers, Skype, Napster, Bit. Torrent for P 2 P Collaboration – text, audio-video conferencing, files del. icio. us, Connotea, Citeulike, Bibsonomy, Biolicious manage shared bookmarks My. Space, You. Tube, Bebo, Hotornot, Facebook, or similar sites allow you to create (upload) community resources and share them; Friendster, Linked. In create networks • http: //en. wikipedia. org/wiki/List_of_social_networking_websites Writely, Wikis and Blogs are powerful specialized shared document systems Google Scholar and Windows Live Academic Search tells you who has cited your papers while publisher sites tell you about coauthors 15
“Best Web 2. 0 Sites” -- 2006 Extracted from http: //web 2. wsj 2. com/ Social Networking Start Pages Social Bookmarking Peer Production News Social Media Sharing Online Storage (Computing) 16
Web 2. 0 Systems are Portals, Services, Resources Captures the incredible development of interactive Web sites enabling people to create and collaborate 17
Web 2. 0 and Web Services I Web Services have clearly defined protocols (SOAP) and a well defined mechanism (WSDL) to define service interfaces • There is good. NET and Java support • The so-called WS-* specifications provide a rich sophisticated but complicated standard set of capabilities for security, fault tolerance, metadata, discovery, notification etc. “Narrow Grids” build on Web Services and provide a robust managed environment with growing adoption in Enterprise systems and distributed science (so called e-Science) Web 2. 0 supports a similar architecture to Web services but has developed in a more chaotic but remarkably successful fashion with a service architecture with a variety of protocols including those of Web and Grid services • Over 500 Interfaces defined at http: //www. programmableweb. com/apis Web 2. 0 also has many well known capabilities with Google Maps and Amazon Compute/Storage services of clear general relevance There also Web 2. 0 services supporting novel collaboration modes and user interaction with the web as seen in social networking sites, portals, My. Space, You. Tube, 18
Web 2. 0 and Web Services II I once thought Web Services were inevitable but this is no longer clear to me Web services are complicated, slow and non functional • WS-Security is unnecessarily slow and pedantic (canonicalization of XML) • WS-RM (Reliable Messaging) seems to have poor adoption and doesn’t work well in collaboration • WSDM (distributed management) specifies a lot There are de facto standards like Google Maps and powerful suppliers like Google which “define the rules” One can easily combine SOAP (Web Service) based services/systems with HTTP messages but the “lowest common denominator” suggests additional structure/complexity of SOAP will not easily survive 19
Applications, Infrastructure, Technologies The discussion is confused by inconsistent use of terminology – this is what I mean Multicore, Narrow and Broad Grids and Web 2. 0 (Enterprise 2. 0) are technologies These technologies combine and compete to build infrastructures termed e-infrastructure or Cyberinfrastructure • Although multicore can and will support “standalone” clients probably most important client and server applications of the future will be internet enhanced/enabled so key aspect of multicore is its role and integration in e -infrastructure e-moreorlessanything is an emerging application area of broad importance that is hosted on the infrastructures e-infrastructure or Cyberinfrastructure 20
Some Web 2. 0 Activities at IU Use of Blogs, RSS feeds, Wikis etc. Use of Mashups for Cheminformatics Grid workflows Moving from Portlets to Gadgets in portals (or at least supporting both) Use of Connotea to produce tagged document collections such as http: //www. connotea. org/user/crmc for parallel computing Semantic Research Grid integrates multiple tagging and search systems and copes with overlapping inconsistent annotations MSI-CIEC portal augments Connotea to tag a mix of URL and URI’s e. g. NSF Tera. Grid use, PI’s and Proposals • Hopes to support collaboration (for Minority Serving Institution faculty) 21
Use blog to create posts. Display blog RSS feed in Media. Wiki. 22
Semantic Research Grid (SRG) Architecture 10/19/2021 23 23
MSI-CIEC Portal MSI-CIEC Minority Serving Institution Cyber. Infrastructure Empowerment Coalition 24
Mashups v Workflow? Mashup Tools are reviewed at http: //blogs. zdnet. com/Hinchcliffe/? p=63 Workflow Tools are reviewed by Gannon and Fox http: //grids. ucs. indiana. edu/ptliupages/publications/Workflow-overview. pdf Both include scripting in PHP, Python, sh etc. as both implement distributed programming at level of services Mashups use all types of service interfaces and perhaps do not have the potential robustness (security) of Grid service approach Mashups typically “pure” HTTP (REST) 25
Grid Workflow Datamining in Earth Science NASA GPS Work with Scripps Institute Grid services controlled by workflow process real time data from ~70 GPS Sensors in Southern California Earthquake Streaming Data Support Archival Transformations Data Checking Hidden Markov Datamining (JPL) Display (GIS) Real Time 26
Grid Workflow Data Assimilation in Earth Science Grid services triggered by abnormal events and controlled by workflow process real time data from radar and high resolution simulations for tornado forecasts Typical graphical interface to service composition 27
Web 2. 0 uses all types of Services Here a Gadget Mashup uses a 3 service workflow with a Java. Script Gadget Client 28
Web 2. 0 Mashups and APIs http: //www. programmable web. com/apis has (Sept 12 2007) 2312 Mashups and 511 Web 2. 0 APIs and with Google. Maps the most often used in Mashups The Web 2. 0 UDDI (service registry) 29
The List of Web 2. 0 API’s Each site has API and its features Divided into broad categories Only a few used a lot (49 API’s used in 10 or more mashups) RSS feed of new APIs Amazon S 3 growing in popularity 30
Grid-style portal as used in Earthquake Grid The Portal is built from portlets – providing user interface fragments for each service that are composed into the full interface – uses OGCE technology as does planetary science VLAB portal with University of Minnesota Now to Portals 31
Note the many competitions powering Web 2. 0 Mashup Development Portlets v. Google Gadgets Portals for Grid Systems are built using portlets with software like Grid. Sphere integrating these on the server-side into a single web-page Google (at least) offers the Google sidebar and Google home page which support Web 2. 0 services and do not use a server side aggregator Google is more user friendly! The many Web 2. 0 competitions is an interesting model for promoting development in the world-wide distributed collection of Web 2. 0 developers I guess Web 2. 0 model will win! 32
Typical Google Gadget Structure Google Gadgets are an example of Start Page technology See http: //blogs. zdnet. com/Hinchcliffe/? p=8 … Lots of HTML and Java. Script </Content> </Module> Portlets build User Interfaces by combining fragments in a standalone Java Server Google Gadgets build User Interfaces by combining fragments with Java. Script on the client
Web 2. 0 v Narrow Grid I Web 2. 0 and Grids are addressing a similar application class although Web 2. 0 has focused on user interactions • So technology has similar requirements Web 2. 0 chooses simplicity (REST rather than SOAP) to lower barrier to everyone participating Web 2. 0 and Parallel Computing tend to use traditional (possibly visual) (scripting) languages for equivalent of workflow whereas Grids use visual interface backend recorded in BPEL Web 2. 0 and Grids both use SOA Service Oriented Architectures “System of Systems”: Grids and Web 2. 0 are likely to build systems hierarchically out of smaller systems • We need to support Grids of Grids, Webs of Grids, Grids of Services etc. i. e. systems of all sorts 34
Web 2. 0 v Narrow Grid II Web 2. 0 has a set of major services like Google. Maps or Flickr but the world is composing Mashups that make new composite services • End-point standards are set by end-point owners • Many different protocols covering a variety of de-facto standards Narrow Grids have a set of major software systems like Condor and Globus and a different world is extending with custom services and linking with workflow Popular Web 2. 0 technologies are PHP, Java. Script, JSON, AJAX and REST with “Start Page” e. g. (Google Gadgets) interfaces Popular Narrow Grid technologies are Apache Axis, BPEL WSDL and SOAP with portlet interfaces Robustness of Grids demanded by the Enterprise? Not so clear that Web 2. 0 won’t eventually dominate other application areas and with Enterprise 2. 0 it’s invading Grids The world does itself in large numbers!
Web 2. 0 v Narrow Grid III Narrow Grids have a strong emphasis on standards and structure; Web 2. 0 lets a 1000 flowers (protocols) and a million developers bloom and focuses on functionality, broad usability and simplicity • Semantic Web/Grid has structure to allow reasoning • Annotation in sites like del. icio. us and uploading to My. Space/You. Tube is unstructured and free text search replaces structured ontologies Portals are likely to feature both Web and “desktop client” technology although it is possible that Web approach will be adopted more or less uniformly Web 2. 0 has a very active portal activity which has similar architecture to Grids • A page has multiple user interface fragments Web 2. 0 user interface integration is typically Client side using Gadgets AJAX and Java. Script while • Grids are in a special JSR 168 portal server side using Portlets WSRP and Java 36
The Ten areas covered by the 60 core WS-* Specifications WS-* Specification Area Typical Grid/Web Service Examples 1: Core Service Model XML, WSDL, SOAP 2: Service Internet WS-Addressing, WS-Message. Delivery; Reliable Messaging WSRM; Efficient Messaging MOTM 3: Notification WS-Notification, WS-Eventing (Publish. Subscribe) 4: Workflow and Transactions BPEL, WS-Choreography, WS-Coordination 5: Security WS-Security, WS-Trust, WS-Federation, SAML, WS-Secure. Conversation 6: Service Discovery UDDI, WS-Discovery 7: System Metadata and State WSRF, WS-Metadata. Exchange, WS-Context 8: Management WSDM, WS-Management, WS-Transfer 9: Policy and Agreements WS-Policy, WS-Agreement 10: Portals and User Interfaces WSRP (Remote Portlets) 37
WS-* Areas and Web 2. 0 WS-* Specification Area Web 2. 0 Approach 1: Core Service Model XML becomes optional but still useful SOAP becomes JSON RSS ATOM WSDL becomes REST with API as GET PUT etc. Axis becomes Xml. Http. Request 2: Service Internet No special Qo. S. Use JMS or equivalent? 3: Notification Hard with HTTP without polling– JMS perhaps? 4: Workflow and Transactions (no Transactions in Web 2. 0) Mashups, Google Map. Reduce Scripting with PHP Java. Script …. 5: Security SSL, HTTP Authentication/Authorization, Open. ID is Web 2. 0 Single Sign on 6: Service Discovery http: //www. programmableweb. com 7: System Metadata and State Processed by application – no system state – Microformats are a universal metadata approach 8: Management==Interaction WS-Transfer style Protocols GET PUT etc. 9: Policy and Agreements Service dependent. Processed by application 10: Portals and User Interfaces Start Pages, AJAX and Widgets(Netvibes) Gadgets 38
Too much Computing? Historically one has tried to increase computing capabilities by • Optimizing performance of codes • Exploiting all possible CPU’s such as Graphics co-processors and “idle cycles” • Making central computers available such as NSF/Do. E/Do. D supercomputer networks Next Crisis in technology area will be the opposite problem – commodity chips will be 32 -128 way parallel in 5 years time and we currently have no idea how to use them – especially on clients • Only 2 releases of standard software (e. g. Office) in this time span Gaming and Generalized decision support (data mining) are two obvious ways of using these cycles • Intel RMS analysis • Note even cell phones will be multicore There is “Too much data” as well as “Too much computing” but unclear implications 39
Intel’s Projection 40
RMS: Recognition Mining Synthesis What is …? Is it …? What if …? Model Find a model instance Create a model instance Today Model-less Real-time streaming and transactions on static – structured datasets Very limited realism Tomorrow Model-based multimodal recognition Real-time analytics on dynamic, unstructured, multimodal datasets Pradeep K. Dubey, pradeep. dubey@intel. com Photo-realism and physics-based animation 41
Recognition What is a tumor? Mining Synthesis Is there a tumor here? What if the tumor progresses? It is all about dealing efficiently with complex multimodal datasets Images courtesy: http: //splweb. bwh. harvard. edu: 8000/pages/images_movies. html Pradeep K. Dubey, pradeep. dubey@intel. com 42
Intel’s Application Stack 43
Multicore SALSA at IU Service Aggregated Linked Sequential Activities • http: //www. infomall. org/multicore Aims to link parallel and distributed (Grid) computing by developing parallel applications as services and not as programs or libraries • Improve traditionally poor parallel programming development environments Can use messaging to link parallel and Grid services but performance – functionality tradeoffs different • Parallelism needs few µs latency for message latency and thread spawning • Network overheads in Grid 10 -100’s µs Developing Service (library) of multicore parallel data mining algorithms 44
Microsoft CCR for Parallelism • Use Microsoft CCR/DSS where DSS is mash-up/workflow service model built from CCR and CCR supports MPI or Dynamic threads • CCR Supports exchange of messages between threads using named ports • From. Handler: Spawn threads without reading ports • Receive: Each handler reads one item from a single port • Multiple. Item. Receive: 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. • Multiple. Port. Receive: Each handler reads a one item of a given type from multiple ports. • Joined. Receive: Each handler reads one item from each of two ports. The items can be of different type. • Choice: Execute a choice of two or more port-handler pairings • Interleave: Consists of a set of arbiters (port -- handler pairs) of 3 types that are Concurrent, Exclusive or Teardown (called at end for clean up). Concurrent arbiters are run concurrently but exclusive handlers are • http: //msdn. microsoft. com/robotics/ 45 45
DSS Service Measurements Timing of HP Opteron Multicore as a function of number of simultaneous twoway service messages processed (November 2006 DSS Release) Measurements of Axis 2 shows about 500 microseconds – DSS is 10 times better 46 46
MPI Exchange Latency in µs (20 -30 µs computation between messaging) Machine OS Runtime Grains Parallelism MPI Exchange Latency Intel 8 c: gf 12 (8 core 2. 33 Ghz) (in 2 chips) Redhat MPJE (Java) Process 8 181 MPICH 2 (C) Process 8 40. 0 MPICH 2: Fast Process 8 39. 3 Nemesis Process 8 4. 21 MPJE Process 8 157 mpi. Java Process 8 111 MPICH 2 Process 8 64. 2 Vista MPJE Process 8 170 Fedora MPJE Process 8 142 Fedora mpi. Java Process 8 100 Vista CCR (C#) Thread 8 20. 2 XP MPJE Process 4 185 Redhat MPJE Process 4 152 mpi. Java Process 4 99. 4 MPICH 2 Process 4 39. 3 XP CCR Thread 4 16. 3 XP CCR Thread 4 25. 8 Intel 8 c: gf 20 (8 core 2. 33 Ghz) Intel 8 b (8 core 2. 66 Ghz) AMD 4 (4 core 2. 19 Ghz) Intel 4 (4 core 2. 8 Ghz) Fedora 47
Clustering algorithm annealing by decreasing distance scale and gradually finds more clusters as resolution improved Here we see 10 increasing to 30 as algorithm progresses 48
Parallel Multicore Clustering (C# on Windows) Parallel Overhead on 8 Threads running on Intel 8 core Speedup = 8/(1+Overhead) 10 Clusters Overhead = Constant 1 + Constant 2/n Constant 1 = 0. 05 to 0. 1 (Client Windows) due to thread runtime fluctuations 20 Clusters 10000/(Grain Size n = points per core) PC 07 Intro gcf@indiana. edu 49
We use DSS as Service Framework as Integrated with CCR Supporting MPI/Threading 50
Intel 8 -core C# with 80 Clusters: Vista Run Time Fluctuations for Clustering Kernel • 2 Quadcore Processors • This is average of standard deviation of run time of the 8 threads between messaging synchronization points Standard Deviation/Run Time Number of Threads PC 07 Intro gcf@indiana. edu 51
Intel 8 core with 80 Clusters: Redhat Run Time Fluctuations for Clustering Kernel • This is average of standard deviation of run time of the 8 threads between messaging synchronization points Standard Deviation/Run Time PC 07 Intro gcf@indiana. edu Number of Threads 52
What should one do? i. e. How does one Cyberinfrastructure enable a given area/application XYZ As computing free, focus on identifying data/information/knowledge/wisdom needed (there is probably too much data but not so much wisdom in DIKW pipeline) • Should we care just about “original data” or also about the whole pipeline DIKW? Scope out supercomputer/computer services needed and exploit OGF standards Identify services (filters, often data mining) needed by XYZ? • Will we need parallel implementations of filters – if so use multicore compatible frameworks Identify standards for application XYZ Set up distributed XYZ Services Use Web 2. 0 (as it makes things easier) not current Grids (which makes things harder) • Build a “Programmable XYZ Web”’ • Emphasize Simplicity • Is “Secrecy” important and in fact viable? Often important but hard What are synergies of XYZ to pervasive capabilities such as Web 2. 0 sites, National resources like Tera. Grid, and “Personal aides in an information rich world” (future of PC) ? 53
- Slides: 53