Actas JP2011 - Universidad de La Laguna

Actas XXII Jornadas de Paralelismo (JP2011) , La Laguna, Tenerife, 7-9 septiembre 2011to limit the number of threads that can be created inpolicy S3. Thus, policy S3 must be redesigned.Fig. 10. Execution times incrementing the number ofsearching engines.VI. CONCLUSIONSThe essential objective of this study was to decreasethe run time of the application Biblio-MetReS byparallelizing the parsing and analysis of the documentsby the application. In doing so, we wanted to establish astrategy that would be close to optimal in creating theadequate number of threads for decreasing the run timeof the application. Our preliminary results suggest that astrategy that fairly divides the number of documents tobe analyzed by the number of physical hardware threadsthat are available in the machine is, in most cases, thebest policy.VII. FUTURE WORKWe are now planning the design of new efficientscheduling algorithms to distribute the parser phasebetween cores of one node. Once this is accomplished,we will consider clusters of workstations. In doing so,we want to classify documents according to their types(pdf, text or HTML) and sizes. Then schedulingdecisions will try to balance the load between thethreads assigned to the cores according to such aclassification.Future challenges will go in the direction of alsoparallelize the search phase, which is another step that ishighly amenable to parallelization. We will analyze thebest policy for distributing the bandwidth worldwidebetween nodes located in Internet.ACKNOWLEDGEMENTSThis work was supported by the MEyC-Spain undercontracts BFU2007-62772/BMC, BFU2010-17704,TIN2008-05913 and CSD-2007-00050, by Generalitatde Catalunya, through research groups 2009SGR809 and2009SGR145 and the CUR of DIUE of GENCAT, andby the European Social Fund.REFERENCES1. Alves R, Sorribas A: In silico pathwayreconstruction: Iron-sulfur cluster biogenesis inSaccharomyces cerevisiae. BMC Syst Biol 2007,1:10.2. Markowetz F, Spang R: Inferring cellular networks--a review. BMC Bioinformatics 2007, 8 Suppl 6:S5.3. Hoffmann R, Valencia A: Implementing the iHOPconcept for navigation of biomedical literature.Bioinformatics 2005, 21 Suppl 2:ii252-258.4. Hoffmann R, Valencia A: A gene network fornavigating the literature. Nat Genet 2004,36(7):664.5. von Mering C, Jensen LJ, Kuhn M, Chaffron S,Doerks T, Kruger B, Snel B, Bork P: STRING 7--recent developments in the integration andprediction of protein interactions. Nucleic Acids Res2007, 35(Database issue):D358-362.6. Barbosa-Silva A, Soldatos TG, Magalhaes IL,Pavlopoulos GA, Fontaine JF, Andrade-NavarroMA, Schneider R, Ortega JM: LAITOR--LiteratureAssistant for Identification of Terms co-Occurrences and Relationships. BMCBioinformatics 2010, 11:70.7. Kemper B, Matsuzaki T, Matsuoka Y, Tsuruoka Y,Kitano H, Ananiadou S, Tsujii J: PathText: a textmining integrator for biological pathwayvisualizations. Bioinformatics 2010, 26(12):i374-381.8. Krallinger M, Leitner F, Valencia A: Analysis ofbiological processes and diseases using text miningapproaches. Methods Mol Biol 2010, 593:341-382.9. Krallinger M, Valencia A, Hirschman L: Linkinggenes to literature: text mining, informationextraction, and retrieval applications for biology.Genome Biol 2008, 9 Suppl 2:S8.10. Hahn U, Valencia A: Semantic Mining inBiomedicine (Introduction to the papers selectedfrom the SMBM 2005 Symposium, Hinxton, U.K.,April 2005). Bioinformatics 2006, 22(6):643-644.11. McIntosh T, Curran JR: Challenges forautomatically extracting molecular interactionsfrom full-text articles. BMC Bioinformatics 2009,10:311.12. Aoki KF, Kanehisa M: Using the KEGG databaseresource. Curr Protoc Bioinformatics 2005, Chapter1:Unit 1 12.13. Geer LY, Marchler-Bauer A, Geer RC, Han L, He J,He S, Liu C, Shi W, Bryant SH: The NCBIBioSystems database. Nucleic Acids Res 2010,38(Database issue):D492-496.14. Usié A, Karathia H, Solsona F, Alves R. Biblio-MetReS: A Bibliometric Reconstruction Server.ICMSB 2011.JP2011-62