dc.contributor.author | Arandi, Samer | en |
dc.contributor.author | Matheou, George | en |
dc.contributor.author | Kyriacou, Costas | en |
dc.contributor.author | Evripidou, Paraskevas | en |
dc.creator | Arandi, Samer | en |
dc.creator | Matheou, George | en |
dc.creator | Kyriacou, Costas | en |
dc.creator | Evripidou, Paraskevas | en |
dc.date.accessioned | 2019-11-13T10:38:20Z | |
dc.date.available | 2019-11-13T10:38:20Z | |
dc.date.issued | 2017 | |
dc.identifier.uri | http://gnosis.library.ucy.ac.cy/handle/7/53574 | |
dc.description.abstract | In this paper we report our experience in implementing and evaluating the Data-Driven Multithreading (DDM) model on a heterogeneous multi-core processor. DDM is a non-blocking multithreading model that decouples the synchronization from the computation portions of a program, allowing them to execute asynchronously in a dataflow manner. Thread dependencies are determined by the compiler/programmer while thread scheduling is done dynamically at runtime based on data availability. The target processor for this implementation is the Cell processor. We call this implementation the Data-Driven Multithreading Virtual Machine for the Cell processor (DDM-(Formula presented.)). Thread scheduling is handled in software by the Power Processing Element core of the Cell while the Synergistic Processing Element cores execute the program threads. DDM-(Formula presented.) virtualizes the parallel resources of the Cell, handles the heterogeneity of the cores, manages the Cell memory hierarchy efficiently and supports distributed execution across a cluster of Cell nodes. DDM-(Formula presented.) has been implemented on a single Cell processor with six computation cores, as well as, on a four Cell processor cluster with 24 computation cores. We present an in-depth performance analysis of DDM-(Formula presented.), using a suite of standard computational benchmarks. The evaluation shows that DDM-(Formula presented.) scales well and tolerates scheduling overheads, memory and communication latencies effectively. Furthermore, DDM-(Formula presented.) compares favorably with other platforms targeting the Cell processor, such as, the CellSs and Sequoia. © 2017 Springer Science+Business Media New York | en |
dc.source | International Journal of Parallel Programming | en |
dc.source.uri | https://www.scopus.com/inward/record.uri?eid=2-s2.0-85011880092&doi=10.1007%2fs10766-016-0486-6&partnerID=40&md5=7f06b43eeb05381c70496cea9d0fa07e | |
dc.subject | Cells | en |
dc.subject | Cytology | en |
dc.subject | Performance analysis | en |
dc.subject | Scheduling | en |
dc.subject | Multicore programming | en |
dc.subject | Program processors | en |
dc.subject | Data-flow scheduling | en |
dc.subject | Multi-core systems | en |
dc.subject | Multitasking | en |
dc.subject | CacheFlow | en |
dc.subject | Communication latency | en |
dc.subject | Data flow analysis | en |
dc.subject | Data-Driven Multithreading | en |
dc.subject | Dataflow scheduling | en |
dc.subject | Heterogeneous multi core processors | en |
dc.subject | Heterogeneous processors | en |
dc.subject | Virtual machine | en |
dc.title | Data-Driven Thread Execution on Heterogeneous Processors | en |
dc.type | info:eu-repo/semantics/article | |
dc.identifier.doi | 10.1007/s10766-016-0486-6 | |
dc.description.startingpage | 1 | |
dc.description.endingpage | 27 | |
dc.author.faculty | 002 Σχολή Θετικών και Εφαρμοσμένων Επιστημών / Faculty of Pure and Applied Sciences | |
dc.author.department | Τμήμα Πληροφορικής / Department of Computer Science | |
dc.type.uhtype | Article | en |
dc.description.notes | <p>Article in Press</p> | en |
dc.source.abbreviation | Int J Parallel Program | en |
dc.contributor.orcid | Evripidou, Paraskevas [0000-0002-2335-9505] | |
dc.contributor.orcid | Matheou, George [0000-0002-3019-7102] | |
dc.gnosis.orcid | 0000-0002-2335-9505 | |
dc.gnosis.orcid | 0000-0002-3019-7102 | |