1. CLOUD COMPUTING ARCHITECTURES & APPLICATIONS LECTURERS LAZAR KIRCHEV, PhD ILIYAN NENOV KRUM BAKALSKY 2 May, 2011 LECTURE #11 SCIENTIFIC APPLICATIONS FOR COMPUTING CLOUDS. FREELY ACCESSIBLE LARGE DATA SETS.

2. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications2 OUTLINE Scientific applications characteristics Issues with using current cloud infrastructures for scientific applications Perspectives for scientific applications which clouds offer Scientific projects using cloud computing

3. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications3 Introduction  Cloud computing emerged to meet the needs of large businesses for flexible IT infrastructure  Academia, on the other hand, uses for computational/storage purposes HPC and Grids  Recently scientists turn towards cloud infrastructures to evaluate if scientific applications can benefit from them

4. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications4 Introduction  Efforts in scientific community to use clouds - US Government funded research projects  SDSC funded by NSF  Department of Energy project  Science clouds  Small clouds, made available by several scientific institutions  Amazon hosts Public Data Sets on their Elastic Block Store

5. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications5 Characteristics of scientific applications  Data sharing, integration and analysis supported by metadata  Very large data sets  Very large execution systems (e.g. supercomputers)  High performance – HPC and HTC, high system utilization  Resource usage – exclusive, space-shared

6. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications6 Characteristics of scientific applications  Consist of parallel jobs with extensive inter-process communication, workflows or bag-of-tasks with no inter-process communication  Heterogeneous workloads, bottleneck could be CPU, I/O, memory or network  Examples for scientific applications: CARMEN, BLAST, SNFactory

7. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications7 Issues with cloud infrastructures  Clouds are meant to replace the small to medium-size enterprise data centers; these are much less utilized than the systems used for scientific computing  Time sharing of resources and virtualization – increase concurrency of users, decrease performance  The performance is not enough – performance analysis of four clouds (EC2, GoGrid, Elastic Hosts, Mosso) with scientific applications show this

8. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications8 Issues with cloud infrastructures  Lack of dedicated access to the hardware and fine-grained sharing of resources associated with virtualization decreases performance  Slow network connections between the virtual machines  Optimized for business applications instead of HPC applications

9. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications9 Issues with cloud infrastructures  Cloud environment not suitable for scientific applications (HPC cluster environment expected, or the application should be modified)  Abstraction (provided by virtualization) vs. control (over hardware resources)  Data access and interoperability – dynamically provisioned resources increases the need to easily integrate distributed data and repositories  Difficult management and utilization of the virutalized resources

10. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications10 Perspectives provided by the cloud infrastructures  Cost-effective for scientific computing  in comparison with supporting the respective infrastructure in-house  users can choose the most effective computing resources suited for their application and budget  Appropriate alternative when resources are needed instantly and temporarily  Can accommodate large data sets  Elastic resource provisioning

11. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications11 Perspectives provided by the cloud infrastructures  Easily provide access to a shared environment, which may evolve  Virtual ownership of resources  always can access your resources when needed  Ease of deployment  package the OS, libraries, patches and application codes on an image

12. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications12 Scientific projects using cloud computing - CARMEN  Developed to solve neuroscience problems  Deploy a set of generic e-science services (data management, service management, security, workflow enactment) on the cloud  build domain-specific services on top

13. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications13 Scientific projects using cloud computing - Open Science Data Cloud  A persistent, distributed storage and computing resource designed to manage, analyze, share, and archive scientific data  Managed by Open Cloud Consortium  Hadoop DFS and MapReduce, Sector, UDT (UDP-based protocol)  Eucaliptus elasticity, Images compatible with Amazon’s

14. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications14 Scientific projects using cloud computing - AzureBlast  Implementation of the BLAST – an algorithm used in bioinformatics  Implemented for Windows Azure, Microsoft’s PaaS solution  AzureBlast is specifically tailored to utilize the Azure infrastructure  Therefore it is well supported by Microsoft’s cloud

15. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications15 Conclusion  Current cloud infrastructures are not optimized for scientific computing  Performance  Network latency  Management of virtualized resources  However, they are promising  Economic efficiency  Elasticity  When HPC becomes in-scope for clouds performance will improve

16. END OF LECTURE #11

17. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications17 The information in this document is compiled using varous public sources, freely available in internet. These sources include:  http://www.scribd.com/doc/17929394/Cloud-Computing-Use-Cases-Whitepaperhttp://www.scribd.com/doc/17929394/Cloud-Computing-Use-Cases-Whitepaper  http://www.enisa.europa.eu/act/rm/files/deliverables/cloud-computing-risk-assessmenthttp://www.enisa.europa.eu/act/rm/files/deliverables/cloud-computing-risk-assessment  http://code.google.com/edu/parallel/index.html http://code.google.com/edu/parallel/index.html  Google: Cluster Computing and MapReduce: http://code.google.com/edu/submissions/mapreduce-minilecture/listing.htmlhttp://code.google.com/edu/submissions/mapreduce-minilecture/listing.html  Google Course: MapReduce in a Week http://code.google.com/edu/submissions/mapreduce/listing.htmlhttp://code.google.com/edu/submissions/mapreduce/listing.html  Intensive MapReduce course at MIT http://mr.iap.2008.googlepages.comhttp://mr.iap.2008.googlepages.com  Hadoop Virtual Image Documentation http://code.google.com/edu/parallel/tools/hadoopvm/index.htmlhttp://code.google.com/edu/parallel/tools/hadoopvm/index.html  http://www.umiacs.umd.edu/~jimmylin/cloud-computinghttp://www.umiacs.umd.edu/~jimmylin/cloud-computing  Colby Ranger, Ramanan Raghuraman, Arun Penmetsa, Gary Bradski, Christos Kozyrakis,  Evaluating MapReduce for Multi-core and Multiprocessor Systems, http://csl.stanford.edu/~christos/publications/2007.cmp_mapreduce.hpca.pdfhttp://csl.stanford.edu/~christos/publications/2007.cmp_mapreduce.hpca.pdf  http://www.dbms2.com/2008/08/26/why-mapreduce-matters-to-sql-data-warehousinghttp://www.dbms2.com/2008/08/26/why-mapreduce-matters-to-sql-data-warehousing  Bingsheng He, Wenbin Fang, Qiong Luo, Mars: A MapReduce Framework on Graphics Processors http://www.cse.ust.hk/catalac/users/saven/GPGPU/MapReduce/PACT08/171.pdfhttp://www.cse.ust.hk/catalac/users/saven/GPGPU/MapReduce/PACT08/171.pdf  Hung-chih Yang, Ali Dasdan, Map-reduce-merge: simplified relational data processing on large clusters http://portal.acm.org/citation.cfm?doid=1247480.1247602http://portal.acm.org/citation.cfm?doid=1247480.1247602  Foto N. Afrati, Jeffrey D. Ullman, A New Computation Model for Rack-Based Computing http://infolab.stanford.edu/~ullman/pub/mapred.pdfhttp://infolab.stanford.edu/~ullman/pub/mapred.pdf  Ralf Lammel, Google’s MapReduce Programming Model Revisite http://www.cs.vu.nl/~ralf/MapReduce/paper.pdfhttp://www.cs.vu.nl/~ralf/MapReduce/paper.pdf  http://www.baselinemag.com/c/a/Infrastructure/How-Google-Works-1http://www.baselinemag.com/c/a/Infrastructure/How-Google-Works-1  Joe Hellerstein, Parallel Programming in the Age of Big Data http://gigaom.com/2008/11/09/mapreduce-leads-the-way-for-parallel-programminghttp://gigaom.com/2008/11/09/mapreduce-leads-the-way-for-parallel-programming  Jeffrey Dean and Sanjay Ghemawat, MapReduce: Simplified Data Processing on Large Clusters https://sites.google.com/a/colgate.edu/cloudintro/Homehttps://sites.google.com/a/colgate.edu/cloudintro/Home © 2011 COPYRIGHTS DISCLAIMER The information in this document is proprietary to Sofia University “Sv. Kliment Ohridski” (called THE UNIVERSITY bellow) http://uni-sofia.bg THE UNIVERSITY assumes no responsibility for errors or omissions in this document. THE UNIVERSITY does not warrant the accuracy or completeness of the information, text, graphics, links, or other items contained within this material. This document is provided without a warranty of any kind, either express or implied, including but not limited to the implied warranties of merchantability, fitness for a particular purpose, or non-infringement. This document is used only for educational purposes related to the masters programs of THE UNIVERSITY, Faculty of Mathematics and Informatics. This document is compiled using various public sources freely available in internet or offered by SAP AG. This document is not used directly or indirectly for any type of commercial use. http://fmi.uni-sofia.bg THE UNIVERSITY shall have no liability for damages of any kind including without limitation direct, special, indirect, or consequential damages that may result from the use of these materials. This limitation shall not apply in cases of intent or gross negligence. The statutory liability for personal injury and defective products is not affected. THE UNIVERSITY has no control over the information that you may access through the use of hot links contained in these materials and does not endorse your use of third-party Web pages nor provide any warranty whatsoever relating to third-party Web pages.

18. 2011 Sofia University “Sv. Kliment Ohridski” > Faculty of Mathematics and Informatics > Cloud Computing Architecture and Applications18 Headline area Drawing area White space The Grid