FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2213038203> ?p ?o ?g. }

• W2213038203 abstract "Big data processing frameworks (MapReduce, Hadoop, Dryad) are hugely popular today because they greatly simplify the deployment and execution of big data analysis jobs requiring the use of many machines in parallel. A strong selling point of these frameworks is their built-in failure resilience support. Big data frameworks can run computations to completion despite occasional failures in the system. However, an important but overlooked point has been the efficiency of their failure resilience. The vision of this thesis is that big data frameworks should not only be failure resilient but that they should provide the resilience in an efficient manner. This means both minimizing the impact of failures on computations as well as minimizing the cost of running proactive failure resilience algorithms during failure-free periods. Towards the end goal of enabling efficient failure resilience for big data frameworks, this thesis makes two contributions. The first part of the thesis presents the first in-depth analysis of the efficiency of the failure resilience provided by Hadoop, the most popular big data processing framework today. The results show that even single machine failures can lead to large, variable and unpredictable job running times. This thesis discovers the causes behind this inefficient behavior, determines the responsible Hadoop mechanisms and points out their limitations. The second part of the thesis focuses on providing efficient failure resilience for the common case of computations comprised of multiple jobs. We present the design, implementation and evaluation of RCMP, a MapReduce system originating from the fundamental insight that using data replication to enable failure resilience oftentimes leads to significant and unnecessary increases in computation running time. In contrast, RCMP is designed to use job re-computation as a first-order failure resilience strategy. Job re-computations under RCMP are efficient. Specifically, RCMP re-computes the minimum amount of work necessary and uniquely it ensures that this minimum re-computation work is performed efficiently. In particular, RCMP mitigates hot-spots that affect data transfers during job re-computations and also ensures that the available compute node parallelism is well exploited." @default.
• W2213038203 created "2016-06-24" @default.
• W2213038203 creator A5079167224 @default.
• W2213038203 date "2013-10-30" @default.
• W2213038203 modified "2023-09-27" @default.
• W2213038203 title "Understanding and Improving the Efficiency of Failure Resilience for Big Data Frameworks" @default.
• W2213038203 cites W1236737278 @default.
• W2213038203 cites W1480850417 @default.
• W2213038203 cites W1510496002 @default.
• W2213038203 cites W1519276763 @default.
• W2213038203 cites W1536639265 @default.
• W2213038203 cites W1598064945 @default.
• W2213038203 cites W1623461676 @default.
• W2213038203 cites W1845494277 @default.
• W2213038203 cites W1846416616 @default.
• W2213038203 cites W1857009450 @default.
• W2213038203 cites W1861377444 @default.
• W2213038203 cites W1900515362 @default.
• W2213038203 cites W1903497807 @default.
• W2213038203 cites W192446467 @default.
• W2213038203 cites W1967890297 @default.
• W2213038203 cites W1973768838 @default.
• W2213038203 cites W1985419898 @default.
• W2213038203 cites W1987932453 @default.
• W2213038203 cites W1993892970 @default.
• W2213038203 cites W2001276096 @default.
• W2213038203 cites W2003597767 @default.
• W2213038203 cites W2010805714 @default.
• W2213038203 cites W2010929544 @default.
• W2213038203 cites W2013344760 @default.
• W2213038203 cites W2025549137 @default.
• W2213038203 cites W2027720485 @default.
• W2213038203 cites W2035543557 @default.
• W2213038203 cites W2035829578 @default.
• W2213038203 cites W2044490410 @default.
• W2213038203 cites W2048554864 @default.
• W2213038203 cites W2070275167 @default.
• W2213038203 cites W2081804145 @default.
• W2213038203 cites W2090673017 @default.
• W2213038203 cites W2091765165 @default.
• W2213038203 cites W2092086632 @default.
• W2213038203 cites W2092643753 @default.
• W2213038203 cites W2096125134 @default.
• W2213038203 cites W2097926925 @default.
• W2213038203 cites W2098935637 @default.
• W2213038203 cites W2104171511 @default.
• W2213038203 cites W2106019582 @default.
• W2213038203 cites W2110086534 @default.
• W2213038203 cites W2110104287 @default.
• W2213038203 cites W2118676041 @default.
• W2213038203 cites W2119528150 @default.
• W2213038203 cites W2119565742 @default.
• W2213038203 cites W2119638333 @default.
• W2213038203 cites W2119738171 @default.
• W2213038203 cites W2123016589 @default.
• W2213038203 cites W2125520775 @default.
• W2213038203 cites W2125775320 @default.
• W2213038203 cites W2126969025 @default.
• W2213038203 cites W2129424879 @default.
• W2213038203 cites W2129542763 @default.
• W2213038203 cites W2130010548 @default.
• W2213038203 cites W2130531694 @default.
• W2213038203 cites W2131975293 @default.
• W2213038203 cites W2136717145 @default.
• W2213038203 cites W2141249441 @default.
• W2213038203 cites W2147470852 @default.
• W2213038203 cites W2153889808 @default.
• W2213038203 cites W2157614013 @default.
• W2213038203 cites W2158733823 @default.
• W2213038203 cites W2158865579 @default.
• W2213038203 cites W2163291889 @default.
• W2213038203 cites W2168595508 @default.
• W2213038203 cites W2173213060 @default.
• W2213038203 cites W2406836379 @default.
• W2213038203 cites W2588083655 @default.
• W2213038203 cites W2912802084 @default.
• W2213038203 cites W65853127 @default.
• W2213038203 cites W69713013 @default.
• W2213038203 cites W91677537 @default.
• W2213038203 hasPublicationYear "2013" @default.
• W2213038203 type Work @default.
• W2213038203 sameAs 2213038203 @default.
• W2213038203 citedByCount "0" @default.
• W2213038203 crossrefType "dissertation" @default.
• W2213038203 hasAuthorship W2213038203A5079167224 @default.
• W2213038203 hasConcept C105339364 @default.
• W2213038203 hasConcept C105795698 @default.
• W2213038203 hasConcept C11413529 @default.
• W2213038203 hasConcept C115903868 @default.
• W2213038203 hasConcept C120314980 @default.
• W2213038203 hasConcept C121332964 @default.
• W2213038203 hasConcept C124101348 @default.
• W2213038203 hasConcept C12590798 @default.
• W2213038203 hasConcept C165136773 @default.
• W2213038203 hasConcept C2522767166 @default.
• W2213038203 hasConcept C2524010 @default.
• W2213038203 hasConcept C2779585090 @default.
• W2213038203 hasConcept C28719098 @default.
• W2213038203 hasConcept C33923547 @default.
• W2213038203 hasConcept C41008148 @default.

• next