Hadoop extensions for distributed computing on reconfigurable active SSD clusters

Simple item page
dc.contributor.authorKaitoua, Abdulrahman
dc.contributor.authorHajj, Hazem M.
dc.contributor.authorSaghir, Mazen A.R.
dc.contributor.authorArtail, Hassan Ali
dc.contributor.authorAkkary, Haitham H.
dc.contributor.authorAwad, Mariette
dc.contributor.authorSharafeddin, Mageda
dc.contributor.authorMershad, Khaleel W.
dc.contributor.departmentDepartment of Electrical and Computer Engineering
dc.contributor.facultyMaroun Semaan Faculty of Engineering and Architecture (MSFEA)
dc.contributor.institutionAmerican University of Beirut
dc.date.accessioned2025-01-24T11:29:08Z
dc.date.available2025-01-24T11:29:08Z
dc.date.issued2014
dc.description.abstractIn this article, we propose new extensions to Hadoop to enable clusters of reconfigurable active solid-state drives (RASSDs) to process streaming data from SSDs using FPGAs. We also develop an analytical model to estimate the performance of RASSD clusters running under Hadoop. Using the Hadoop RASSD platform and network simulators, we validate our design and demonstrate its impact on performance for different workloads taken from Stanford's Phoenix MapReduce project. Our results show that for a hardware acceleration factor of 20×, compute-intensive workloads processing 153MB of data can run up to 11× faster than a standard Hadoop cluster. © 2014 ACM.
dc.identifier.doihttps://doi.org/10.1145/2608199
dc.identifier.eid2-s2.0-84907372451
dc.identifier.urihttp://hdl.handle.net/10938/27094
dc.language.isoen
dc.publisherAssociation for Computing Machinery
dc.relation.ispartofTransactions on Architecture and Code Optimization
dc.sourceScopus
dc.subjectActive storage
dc.subjectData-intensive computing
dc.subjectMiddleware
dc.subjectDigital storage
dc.subjectField programmable gate arrays (fpga)
dc.subjectHardware acceleration
dc.subjectMap-reduce
dc.subjectNetwork simulators
dc.subjectReconfigurable
dc.subjectSolid-state drives
dc.subjectStreaming data
dc.subjectReconfigurable hardware
dc.titleHadoop extensions for distributed computing on reconfigurable active SSD clusters
dc.typeArticle

Files

Original bundle

Now showing 1 - 1 of 1
Thumbnail Image
Name:
2014-10772.pdf
Size:
1.36 MB
Format:
Adobe Portable Document Format
Download

Collections

Department of Electrical and Computer Engineering