Filters








2,283 Hits in 4.8 sec

Solving Large-Scale Granular Resource Allocation Problems Efficiently with POP [article]

Deepak Narayanan, Fiodar Kazhamiaka, Firas Abuzaid, Peter Kraft, Akshay Agrawal, Srikanth Kandula, Stephen Boyd, Matei Zaharia
2021 arXiv   pre-print
We observe, however, that many allocation problems are granular: they consist of a large number of clients and resources, each client requests a small fraction of the total number of resources, and clients  ...  Our technique, Partitioned Optimization Problems (POP), randomly splits the problem into smaller problems (with a subset of the clients and resources in the system) and coalesces the resulting sub-allocations  ...  More Efficient Solving. The optimization community has developed various methods for scaling optimization solvers to handle large problems.  ... 
arXiv:2110.11927v1 fatcat:3softued4vczhbg5vfe4gg5e3m

Don't Give Up on Large Optimization Problems; POP Them! [article]

Deepak Narayanan, Fiodar Kazhamiaka, Firas Abuzaid, Peter Kraft, Matei Zaharia
2021 arXiv   pre-print
Resource allocation problems in many computer systems can be formulated as mathematical optimization problems.  ...  Deciding how to split a large problem into smaller sub-problems, and how to coalesce split allocations into a unified allocation, needs to be performed carefully in a domain-aware way.  ...  problems at the required time granularity for systems with thousands of entities and resource units.  ... 
arXiv:2104.06513v1 fatcat:jzdzdhsdabhnnibut4ivxrf7vu

An energy and deadline aware resource provisioning, scheduling and optimization framework for cloud systems

Yue Gao, Yanzhi Wang, Sandeep K. Gupta, Massoud Pedram
2013 2013 International Conference on Hardware/Software Codesign and System Synthesis (CODES+ISSS)  
In this large-scale, heterogeneous, multi-user environment of a cloud system, profit maximization for the cloud service provider (CSP) is a key objective.  ...  In this paper, the problem of global optimization of the cloud system operation (in the sense of lowering operation costs by maximizing energy efficiency, while satisfying user deadlines defined in the  ...  This problem is similar to the classical load balancing problem in internet services [7] , which can be concurrently solved during resource allocation for independent workloads.  ... 
doi:10.1109/codes-isss.2013.6659018 dblp:conf/codes/GaoWGP13 fatcat:biufaoltdzg6vpjjvusxl4um2y

NFV Platform Design: A Survey [article]

Tianzhu Zhang
2020 arXiv   pre-print
provisioning paradigm with much lower capital expenditure (CapEx) and operational expenditure (OpEx).  ...  We begin with a top-down architectural view of the standard reference NFV platform and present our taxonomy of existing NFV platforms based on principal purpose of design.  ...  It dynamically traces the workload demands for each VNF and allocates compute resources at fine granularity.  ... 
arXiv:2002.11059v2 fatcat:zgafnd6xmvdzngkukq6qicf3gu

Resource allocation and utilization in the Blue Gene/L supercomputer

Y. Aridor, T. Domany, O. Goldshmidt, J. E. Moreira, E. Shmueli
2005 IBM Journal of Research and Development  
The scalability challenge is solved in BG/L by sacrificing granularity of system management.  ...  We show that in these difficult cases, the advantage of BG/L compared with a 3D toroidal machine of the same size is very significant, with resource utilization better by a factor of 2.  ...  The design of BG/L solves the scalability problem by trading off the granularity of management.  ... 
doi:10.1147/rd.492.0425 fatcat:cg4iwmscknddxeglupih32uf6u

Optimization of Decentralized Scheduling for Physic Applications in Grid Environments [article]

Florin Pop
2008 arXiv   pre-print
Our work addresses the problem of scheduling various computationally intensive and data intensive applications that are required for extracting information from satellite images.  ...  It demonstrates a very good behavior for scheduling and executing groups of applications, while also achieving a near-optimal utilization of the resources.  ...  At the opposite end of the granularity scale are coarse-grained or embarrassingly parallel calculations, where each sub-problem is independent of all others.  ... 
arXiv:0812.2164v1 fatcat:kwcjbo4sjjh67fx3otn7n77r5e

An Efficient OpenMP Loop Scheduler for Irregular Applications on Large-Scale NUMA Machines [chapter]

Marie Durand, François Broquedis, Thierry Gautier, Bruno Raffin
2013 Lecture Notes in Computer Science  
Parallel environments like OpenMP provide ways of parallelizing them efficiently, but the achieved performance is closely related to the choice of parameters like the granularity of work or the loop scheduler  ...  Nowadays shared memory HPC platforms expose a large number of cores organized in a hierarchical way.  ...  However, tasks will not solve the problem of granularity, as defining the right number of tasks can be challenging, as studied in our previous work [3] .  ... 
doi:10.1007/978-3-642-40698-0_11 fatcat:qru56qdbjzdjvcu4z7giq73bbi

A parallel, real-time garbage collector

Perry Cheng, Guy E. Blelloch
2001 SIGPLAN notices  
large and small objects.  ...  This paper presents the extensions necessary for a practical implementation: reducing excessive interleaving, handling stacks and global variables, reducing double allocation, and special treatment of  ...  This problem can be solved by replicating globals like other heap-allocated object.  ... 
doi:10.1145/381694.378823 fatcat:kasudneqlfdatgiaq4meoobndq

A parallel, real-time garbage collector

Perry Cheng, Guy E. Blelloch
2001 Proceedings of the ACM SIGPLAN 2001 conference on Programming language design and implementation - PLDI '01  
large and small objects.  ...  This paper presents the extensions necessary for a practical implementation: reducing excessive interleaving, handling stacks and global variables, reducing double allocation, and special treatment of  ...  This problem can be solved by replicating globals like other heap-allocated object.  ... 
doi:10.1145/378795.378823 dblp:conf/pldi/ChengB01 fatcat:gy2vfwuiqvfw7m57jgg3ls7pke

Parallel Genetic Algorithms with GPU Computing [chapter]

John Runwei Cheng, Mitsuo Gen
2020 Industry 4.0 - Impact on Intelligent Logistics and Manufacturing [Working Title]  
A hybrid parallel model, based on the feature of GPU architecture, is suggested to build up efficient parallel GAs for hyper-scale problems.  ...  It helps to find better solutions for complex and difficult cases, which are hard to be solved by using strict optimization methods.  ...  Most published works tackle relatively small-scale problems. In handling large-scale problems, an efficient implementation of parallel GAs on multiple GPUs becomes imperative.  ... 
doi:10.5772/intechopen.89152 fatcat:ktigfcnb35hpve7jtvwobuv3d4

QoE-driven resource optimization for user generated video content in next generation mobile networks

Ali El Essaili, Eckehard Steinbach, Daniele Munaretto, Srisakul Thakolsri, Wolfgang Kellerer
2011 2011 18th IEEE International Conference on Image Processing  
Need arises for optimizing the network resource allocation while preserving the user satisfaction.  ...  We compare our service-centric approach with a QoE-driven one that does not consider video popularity and evaluate both approaches for the uplink of an LTE system.  ...  The analysis of large-scale User-Generated Content (UGC) shows that the users' requests are highly skewed towards popular videos [1] .  ... 
doi:10.1109/icip.2011.6116708 dblp:conf/icip/EssailiSMTK11 fatcat:2qpwkqptczgizpnqsgiia2wz6e

Toward Hierarchical Mobile Edge Computing: An Auction-Based Profit Maximization Approach

Abbas Kiani, Nirwan Ansari
2017 IEEE Internet of Things Journal  
Accordingly, we explore a two time scale approach in which the computing resources are offered in an auction-based profit maximization manner and then the communications resources are allocated to satisfy  ...  However, in such three tier network, it is still desirable to investigate efficient strategies to offer the computing, storage and communications resources to the users.  ...  Based on the proposed model, a two time scale optimization approach for resource allocation is introduced.  ... 
doi:10.1109/jiot.2017.2750030 fatcat:yftlvifz4vgb3frtcuk4qp4jfu

Pushing CDN-ISP collaboration to the limit

Benjamin Frank, Ingmar Poese, Yin Lin, Georgios Smaragdakis, Anja Feldmann, Bruce Maggs, Jannis Rake, Steve Uhlig, Rick Weber
2013 Computer communication review  
Recently, major CDNs have formed strategic alliances with large ISPs to provide content delivery network solutions.  ...  We identify two key enablers for supporting collaboration and improving content delivery performance: informed end-user to server assignment and in-network server allocation.  ...  ., the server locations and the number of servers, by solving an optimization problem (namely the SiSL or the CFL problem, see Section 3.2.3).  ... 
doi:10.1145/2500098.2500103 fatcat:qfi3jycdvrdqro2c2lyzaoduc4

Opening the Information Bottleneck in Complex Scheduling Problems with a Novel Representation: STARK Diagrams [chapter]

Peter C-H. Cheng, Rossano Barone, Peter I. Cowling, Samad Ahmadi
2002 Lecture Notes in Computer Science  
Multiple tables are ubiquitous in representations of schedule information, but they impose large cognitive demands and inhibit the comprehension of highlevel patterns.  ...  An experiment compared performance on STARK diagrams and a conventional representation with features typical of current commercial scheduling software interfaces.  ...  A problem can be more than an order of magnitude more difficult to solve with a poor representation than a good representation [11] .  ... 
doi:10.1007/3-540-46037-3_26 fatcat:zjd5rooolzhj7da2guizbmduvu

An adaptive online system for efficient processing of hierarchical data

Athanasia Asiki, Dimitrios Tsoumakos, Nectarios Koziris
2009 Proceedings of the 18th ACM international symposium on High performance distributed computing - HPDC '09  
We motivate our approach by applying it on a large-scale Grid system: Specifically, we plan to apply our fully decentralized scheme that creates, queries and updates large volumes of hierarchical data  ...  In our method, peers individually decide on the level of indexing according to the granularity of the incoming queries.  ...  Grid computing allows for coordinated resource sharing and problem solving in dynamic virtual organizations (VOs).  ... 
doi:10.1145/1551609.1551627 dblp:conf/hpdc/AsikiTK09 fatcat:kao5t4qc45gwllfmhnrauxlnti
« Previous Showing results 1 — 15 out of 2,283 results