Filters








11 Hits in 3.4 sec

iHarmonizer: Improving the Disk Efficiency of I/O-intensive Multithreaded Codes

Yizhe Wang, Kei Davis, Yuehai Xu, Song Jiang
2012 2012 IEEE 26th International Parallel and Distributed Processing Symposium  
Specifically, we use the compiler to insert code into OpenMP programs so that data usage can be transmitted at run time to a supporting run-time library that prefetches data in a diskfriendly way and coordinates  ...  Challenged by serious power and thermal constraints and limited by available instruction-level parallelism, processor designs have evolved to multi-core architectures.  ...  This work was also funded in part by the Accelerated Strategic Computing program of the Department of Energy.  ... 
doi:10.1109/ipdps.2012.87 dblp:conf/ipps/WangDXJ12 fatcat:4sduoffqgrgtpngj42bjnblmj4

Throttling I/O Streams to Accelerate File-IO Performance [chapter]

Seetharami Seelam, Andre Kerstens, Patricia J. Teller
2007 Lecture Notes in Computer Science  
To increase the scale and performance of high-performance computing (HPC) applications, it is common to distribute computation across multiple processors.  ...  Only 16 processors, i.e., one processor on each of 16 compute nodes (16 MPI tasks) were employed in the experiments.  ...  Due to the advent of multicore processors, as the number of processors in nextgeneration systems grows to solve larger problem sizes and as the ratio of compute processors to I/O nodes increases, the number  ... 
doi:10.1007/978-3-540-75444-2_67 fatcat:ivn3fummmveb3oy5j2p6tuyoue

Malthusian Locks [article]

Dave Dice
2017 arXiv   pre-print
Applications running in modern multithreaded environments are sometimes over-threaded.  ...  In the worst case CR does no harm, but it often yields performance benefits.  ...  Normally the "pop" operator would employ CAS in a loop.  ... 
arXiv:1511.06035v7 fatcat:led5wcfcfnaoznmdxjrbhrxrwa

Software engineering for mobility: reflecting on the past, peering into the future

Gian Pietro Picco, Christine Julien, Amy L. Murphy, Mirco Musolesi, Gruia-Catalin Roman
2014 Proceedings of the on Future of Software Engineering - FOSE 2014  
Further, we ask ourselves to what extent the software engineering community is still interested in taking up the challenges mobility bears.  ...  Indeed, some trends and visions we considered in our original paper materialized, while others faded, disappeared, or morphed into something else.  ...  Anticipatory mobile systems must integrate both on-device hardware and software capabilities with infrastructure hardware and software capabilities in a manner that is fluid, flexible, and aware of resource  ... 
doi:10.1145/2593882.2593884 dblp:conf/icse/PiccoJMMR14 fatcat:tnxj3xfjpzeutpjfiba6iiblge

GDM

Kaibo Wang, Xiaoning Ding, Rubao Lee, Shinpei Kato, Xiaodong Zhang
2014 Performance Evaluation Review  
In this paper, we present the design and implementation of GDM, a fully functional GPGPU device memory manager to address the above problems and unleash the computing power of GPGPUs in general-purpose  ...  In large applications, managing the usage of device memory space is a heavy burden for programmers.  ...  It invokes an in-place add operation for 100 times in a loop.  ... 
doi:10.1145/2637364.2592002 fatcat:zcpzph5qfzh3jczn7nsnj2iruy

Intentional networking

Brett D. Higgins, Azarias Reda, Timur Alperovich, Jason Flinn, T. J. Giuli, Brian Noble, David Watson
2010 Proceedings of the sixteenth annual international conference on Mobile computing and networking - MobiCom '10  
This work is is supported in part by NSF CAREER award CNS-0346686, NSF award CNS-0509089, and the Ford Motor Company.  ...  The views and conclusions contained in this document are those of the authors and should not be interpreted as representing the official policies, either expressed or implied, of NSF, the University of  ...  Multi-Sockets Labels are used in conjunction with label-aware sockets. We call such sockets multi-sockets.  ... 
doi:10.1145/1859995.1860005 dblp:conf/mobicom/HigginsRAFGNW10 fatcat:bsw4rdd6xffubikn3mcgoifhvm

Life-cycle management and placement of service function chains in MEC-enabled 5G networks

Rasoul Behravesh
2021
Firstly, we define the motivations behind video content prefetching and caching in mobile networks and the need for anticipatory prefetching of video segments at the right bitrate during streaming at the  ...  In this context, prediction, anticipatory prefetching, and caching of video segments at right bitrate during streaming at the MEC servers play a pivotal role in MEC-enabled DASH video streaming.  ...  10 in our scenario).  ... 
doi:10.48676/unibo/amsdottorato/9732 fatcat:vdn7s6zwezf7xpfusl6mnfx3oa

Advancing Operating Systems via Aspect-Oriented Programming

Michael Engel, Mathematik Und Informatik, Freisleben, Bernd (Prof. Dr.)
2011
Operating system kernels are among the most complex pieces of software in existence to- day.  ...  Many interactions in kernel code are dynamic, so in order to implement non-static behavior and improve performance, a dynamic weaver that deploys and undeploys aspects at system runtime is required.  ...  a loop.  ... 
doi:10.17192/z2006.0138 fatcat:ojsjec4yizfyxli3h6pqgr2v5e

Proceedings of the Seminar Innovative Internet Technologies and Mobile Communications (IITM), Winter Semester 2018/2019

Georg Carle, Stephan Günther, Benedikt Jaeger, Chair Of Network Architectures
2019
Each semester, the seminar takes place in two different ways: once as a block seminar during the semester break and once in the course of the semester.  ...  In the context of the seminar, each student individually works on a relevant topic in the domain of computer networks supervised by one or more advisors.  ...  We refer to this as anticipatory fetching or prefetching.  ... 
doi:10.2313/net-2019-06-1 fatcat:6bojwbiwjfaljip2a6hplhfqqm

Modeling and Prediction of I/O Performance in Virtualized Environments

Omar-Qais Noorshams
2015
In this thesis, we present a novel performance modeling approach tailored to I/O performance prediction in virtualized environments. First we identify important performance-influencing factors.  ...  Concluding Remarks • Self-aware Computing Systems The long-term vision by Kounev et al. (2010a) is to develop self-aware computing systems that are able to optimally allocate resources in an autonomic  ...  Evaluation Criteria Note: The Anticipatory scheduler was removed from recent Linux distributions and is highly discouraged in virtualized environments.  ... 
doi:10.5445/ir/1000046750 fatcat:pcpbsg7rsbfhdnl3tksgv5w47m

Modeling and Prediction of I/O Performance in Virtualized Environments

Omar-Qais Noorshams
2017
In this thesis, we present a novel performance modeling approach tailored to I/O performance prediction in virtualized environments.  ...  Our approach is validated in a variety of case studies in state-of-the-art, real-world environments based on IBM System z and Sun Fire hardware.  ...  vision by Kounev et al. (2010a) is to develop self-aware computing systems that are able to optimally allocate resources in an autonomic manner.  ... 
doi:10.5445/ksp/1000046300 fatcat:qknjskpcq5f6vnqc5o2hhdsy7m