1 General and Effective Monetary Optimizations for Workflows in IaaS Clouds Amelie Chi Zhou Xtra Computing Group

Slides:



Advertisements
Similar presentations
L3S Research Center University of Hanover Germany
Advertisements

Revisiting Co-Processing for Hash Joins on the Coupled CPU- GPU Architecture School of Computer Engineering Nanyang Technological University 27 th Aug.
Hadi Goudarzi and Massoud Pedram
SkewReduce YongChul Kwon Magdalena Balazinska, Bill Howe, Jerome Rolia* University of Washington, *HP Labs Skew-Resistant Parallel Processing of Feature-Extracting.
SLA-Oriented Resource Provisioning for Cloud Computing
1 A GPU Accelerated Storage System NetSysLab The University of British Columbia Abdullah Gharaibeh with: Samer Al-Kiswany Sathish Gopalakrishnan Matei.
Locality-Aware Dynamic VM Reconfiguration on MapReduce Clouds Jongse Park, Daewoo Lee, Bokyeong Kim, Jaehyuk Huh, Seungryoul Maeng.
ElasticTree: Saving Energy in Data Center Networks Brandon Heller, Srini Seetharaman, Priya Mahadevan, Yiannis Yiakoumis, Puneed Sharma, Sujata Banerjee,
Cloud Computing Resource provisioning Keke Chen. Outline  For Web applications statistical Learning and automatic control for datacenters  For data.
A system Performance Model Instructor: Dr. Yanqing Zhang Presented by: Rajapaksage Jayampthi S.
Meeting Service Level Objectives of Pig Programs Zhuoyao Zhang, Ludmila Cherkasova, Abhishek Verma, Boon Thau Loo University of Pennsylvania Hewlett-Packard.
Tai, Yu-Chang 4/29/2013 Future Generation Computer Systems(FGCS.J) journal homepage: Saeid Abrishami a, ∗, Mahmoud Naghibzadeha,
Software Architecture of High Efficiency Video Coding for Many-Core Systems with Power- Efficient Workload Balancing Muhammad Usman Karim Khan, Muhammad.
System design-related Optimization problems Michela Milano Joint work DEIS Università di Bologna Dip. Ingegneria Università di Ferrara STI Università di.
Cloud based Dynamic workflow with QOS for Mass Spectrometry Data Analysis Thesis Defense: Ashish Nagavaram Graduate student Computer Science and Engineering.
WORKFLOWS IN CLOUD COMPUTING. CLOUD COMPUTING  Delivering applications or services in on-demand environment  Hundreds of thousands of users / applications.
MATE-EC2: A Middleware for Processing Data with Amazon Web Services Tekin Bicer David Chiu* and Gagan Agrawal Department of Compute Science and Engineering.
CISC673 – Optimizing Compilers1/34 Presented by: Sameer Kulkarni Dept of Computer & Information Sciences University of Delaware Phase Ordering.
CREATING A MULTI-WAVELENGTH GALACTIC PLANE ATLAS WITH AMAZON WEB SERVICES G. Bruce Berriman, John Good IPAC, California Institute of Technolog y Ewa Deelman,
HeteroPar 2013 Optimization of a Cloud Resource Management Problem from a Consumer Perspective Rafaelli de C. Coutinho, Lucia M. A. Drummond and Yuri Frota.
A User Experience-based Cloud Service Redeployment Mechanism KANG Yu.
Transfer Learning From Multiple Source Domains via Consensus Regularization Ping Luo, Fuzhen Zhuang, Hui Xiong, Yuhong Xiong, Qing He.
Authors: Weiwei Chen, Ewa Deelman 9th International Conference on Parallel Processing and Applied Mathmatics 1.
MobSched: An Optimizable Scheduler for Mobile Cloud Computing S. SindiaS. GaoB. Black A.LimV. D. AgrawalP. Agrawal Auburn University, Auburn, AL 45 th.
EECE **** Embedded System Design
A Budget Constrained Scheduling of Workflow Applications on Utility Grids using Genetic Algorithms Jia Yu and Rajkumar Buyya Grid Computing and Distributed.
Elastic Applications in the Cloud Dinesh Rajan University of Notre Dame CCL Workshop, June 2012.
Software Pipelining for Stream Programs on Resource Constrained Multi-core Architectures IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEM 2012 Authors:
A Broker for Cost-efficient QoS aware Resource Allocation in EC2. Kurt Vermeersch Coordinator: Kurt Vanmechelen.
Fault-Tolerant Workflow Scheduling Using Spot Instances on Clouds Deepak Poola, Kotagiri Ramamohanarao, and Rajkumar Buyya Cloud Computing and Distributed.
An Autonomic Framework in Cloud Environment Jiedan Zhu Advisor: Prof. Gagan Agrawal.
1 Time & Cost Sensitive Data-Intensive Computing on Hybrid Clouds Tekin Bicer David ChiuGagan Agrawal Department of Compute Science and Engineering The.
Y. Kotani · F. Ino · K. Hagihara Springer Science + Business Media B.V Reporter: 李長霖.
A Framework for Elastic Execution of Existing MPI Programs Aarthi Raveendran Tekin Bicer Gagan Agrawal 1.
20 October 2006Workflow Optimization in Distributed Environments Dynamic Workflow Management Using Performance Data David W. Walker, Yan Huang, Omer F.
Xiao Liu CS3 -- Centre for Complex Software Systems and Services Swinburne University of Technology, Australia Key Research Issues in.
1 A Framework for Data-Intensive Computing with Cloud Bursting Tekin Bicer David ChiuGagan Agrawal Department of Compute Science and Engineering The Ohio.
A Framework for Elastic Execution of Existing MPI Programs Aarthi Raveendran Graduate Student Department Of CSE 1.
A Survey of Distributed Task Schedulers Kei Takahashi (M1)
1 Finding Constant From Change: Revisiting Network Performance Aware Optimizations on IaaS Clouds Yifan Gong, Bingsheng He, Dan Li Nanyang Technological.
Euro-Par, A Resource Allocation Approach for Supporting Time-Critical Applications in Grid Environments Qian Zhu and Gagan Agrawal Department of.
Issues in (Financial) High Performance Computing John Darlington Director Imperial College Internet Centre Fast Financial Algorithms and Computing 4th.
Autonomic scheduling of tasks from data parallel patterns to CPU/GPU core mixes Published in: High Performance Computing and Simulation (HPCS), 2013 International.
Resource Mapping and Scheduling for Heterogeneous Network Processor Systems Liang Yang, Tushar Gohad, Pavel Ghosh, Devesh Sinha, Arunabha Sen and Andrea.
MROrder: Flexible Job Ordering Optimization for Online MapReduce Workloads School of Computer Engineering Nanyang Technological University 30 th Aug 2013.
Shanjiang Tang, Bu-Sung Lee, Bingsheng He, Haikun Liu School of Computer Engineering Nanyang Technological University Long-Term Resource Fairness Towards.
Dynamic Slot Allocation Technique for MapReduce Clusters School of Computer Engineering Nanyang Technological University 25 th Sept 2013 Shanjiang Tang,
DynamicMR: A Dynamic Slot Allocation Optimization Framework for MapReduce Clusters Nanyang Technological University Shanjiang Tang, Bu-Sung Lee, Bingsheng.
OPERATING SYSTEMS CS 3530 Summer 2014 Systems and Models Chapter 03.
1 Approximate XML Query Answers Presenter: Hongyu Guo Authors: N. polyzotis, M. Garofalakis, Y. Ioannidis.
Workflow scheduling and optimization on clouds
1 Adaptive Parallelism for Web Search Myeongjae Jeon Rice University In collaboration with Yuxiong He (MSR), Sameh Elnikety (MSR), Alan L. Cox (Rice),
PDAC-10 Middleware Solutions for Data- Intensive (Scientific) Computing on Clouds Gagan Agrawal Ohio State University (Joint Work with Tekin Bicer, David.
Onlinedeeneislam.blogspot.com1 Design and Analysis of Algorithms Slide # 1 Download From
A stochastic scheduling algorithm for precedence constrained tasks on Grid Future Generation Computer Systems (2011) Xiaoyong Tang, Kenli Li, Guiping Liao,
Euro-Par, HASTE: An Adaptive Middleware for Supporting Time-Critical Event Handling in Distributed Environments ICAC 2008 Conference June 2 nd,
1 Performance Impact of Resource Provisioning on Workflows Gurmeet Singh, Carl Kesselman and Ewa Deelman Information Science Institute University of Southern.
18 May 2006CCGrid2006 Dynamic Workflow Management Using Performance Data Lican Huang, David W. Walker, Yan Huang, and Omer F. Rana Cardiff School of Computer.
Relational Query Processing on OpenCL-based FPGAs Zeke Wang, Johns Paul, Hui Yan Cheah (NTU, Singapore), Bingsheng He (NUS, Singapore), Wei Zhang (HKUST,
TensorFlow– A system for large-scale machine learning
Resource Elasticity for Large-Scale Machine Learning
AWS Batch Overview A highly-efficient, dynamically-scaled, batch computing service May 2017.
A Framework for Automatic Resource and Accuracy Management in A Cloud Environment Smita Vijayakumar.
EECS 582 Final Review Mosharaf Chowdhury EECS 582 – F16.
Automatic Physical Design Tuning: Workload as a Sequence
Optimizing MapReduce for GPUs with Effective Shared Memory Usage
Managing Computational Workflows in the Cloud
Data-Intensive Computing: From Clouds to GPU Clusters
rvGAHP – Push-Based Job Submission Using Reverse SSH Connections
Supporting Online Analytics with User-Defined Estimation and Early Termination in a MapReduce-Like Framework Yi Wang, Linchuan Chen, Gagan Agrawal The.
Presentation transcript:

1 General and Effective Monetary Optimizations for Workflows in IaaS Clouds Amelie Chi Zhou Xtra Computing Group Nanyang Technological University, Singapore presented by

Workflows for Scientific Applications Workflows are structured –Tasks have very different I/O and computational behavior. Real-world workflows –Montage, Ligo, Epigenomics, water-simulation Workflow ensembles [Malawski et al., SC’12] –Composition of workflows with similar structures and different parameters and priorities 2 Montage Ligo Epigenomics

Running Workflows on IaaS Clouds Define IaaS clouds –Provide fundamental computing resources for users to provision –Examples: Amazon EC2, Rackspace, OpenStack, Google Compute Engine … Example projects –Montage, Broadband, Epigenomics on Amazon EC2 [Juve et al., eScience’09] –Astronomy applications on Nimbus, Eucalyptus, and EC2 [Vöckler et al., ScienceCloud’11] –… 3

Workflows in IaaS Clouds Features of IaaS clouds –Pay as you go (e.g., hourly pricing scheme) –Rich and evolving cloud offerings Research problems –Monetary cost optimizations –Performance optimizations –Elasticity –Fault tolerance –… 4 Are the current solutions ideal/sufficient?

Monetary Cost Opportunities Instance types –Amazon EC2 provides 29 types of instances Instance reuse –Hourly charging scheme Pricing schemes –On-demand, spot and reserved pricing V.S. Tasks can have very different I/O and computational behavior. Workflows have different deadline and monetary constraints. Users may have various workflow application scenarios. 5

Current Solutions are Far From Ideal Problems of current approaches –Auto-scaling [Mao et al., SC’11] resource management More effective optimizations  29% less cost –Assume static cloud performance and pricing Cloud dynamics + spot instances  73% less cost –Heuristic-based cost and performance optimizations are specific. They are likely to be suboptimal in evolving and diversified workflow applications. 6 29% 73%

Our Research Efforts Effectiveness –Dyna: Minimize the monetary cost of workflows, addressing both the price and performance dynamics in clouds Generality –ToF: Define transformation operations to model common cost and performance optimizations –Deco: Design a declarative language called WLog to specify various workflow optimization problems 7 The focus of this presentation.

Overall Design We design general workflow optimization frameworks to fully explore the optimization opportunities that lie in workflows 8 Wlog programs Transformation- based Optimizer Problem specification layer Optimization layer Execution layer Deco ToF

9 Outline Related Work Generalized Optimization Frameworks –General transformations for cost and performance optimizations –A declarative language for workflow optimization problems Conclusions

Related Work Performance and monetary cost optimization heuristics –Auto-scaling [Mao et al., SC’11] Fixed sequence of workflow optimizations –Workflow scheduling with performance and cost constraints [Kllapi et al., SIGMOD’11] Consider only one on-demand instance type 10 The heuristics are specifically designed for specific optimization problems and the optimization opportunities are not fully explored.

Related Work (cont’d) Generalized optimization frameworks: overhead is a problem –Generalized bin-ball abstraction for resource allocation [Rai et al., SoCC’12] GPU acceleration Not always convenient to model a problem with the bin-ball model –Declarative language to model a wide range of COPs [Liu et al., VLDB’12] Distributed systems Ignorant to the special features and optimization opportunities in workflows 11 There is no general optimization framework for workflows.

12 Outline Related Work Generalized Optimization Frameworks –General transformations for cost and performance optimizations –A declarative language for workflow optimization problems Conclusions

ToF: A Transformation-based Optimization Framework Outline –Main contributions of this work –System overview –Design details –Evaluation results 13

Main Contributions This study has two major contributions –We define a series of common transformations for the performance and cost optimizations of workflows. –We design a light-weight optimizer to guide the transformation process. 14

Workflow Transformation Definitions –Instance assignment graph Each node represents instance configuration for a task. Same structure as the workflow DAG –Transformation operation Structural change in the instance assignment graph Transformations 0 1, , , ,2,3

System Overview Design ideas –Two types of transformations Main schemes: reduce cost Auxiliary schemes: help main schemes to reduce cost –Use cost model to guide the transformation optimization –Periodical batch optimization Maximize instance sharing and reuse Reduce optimizer overhead 16 Main Schemes Auxiliary Schemes Termin ation? Output Cost model No Yes Optimization process in one plan period

Design Details Transformation operations –Main schemes: Merge, Demote –Auxiliary schemes: Move, Promote, Split, Co- scheduling –Transformations can combine with each other 17

Using Transformations Example of using Move and Merge operations 18 Only transform shape Reduces cost

Experimental Setup Workload –Montage, Ligo and Mixed –Workflow submission rate follows Poisson distribution Comparisons –ToF –Baseline: only implement the initial instance configuration –Auto-scaling [Mao et al., SC’11] –Greedy: randomly select the transformation during optimization All results are normalized to Baseline 19

Evaluation Results on Cost Optimizations 20 Optimization results under the pricing scheme of Amazon EC2. ToF obtains the lowest monetary cost on all workflows. Over Auto-scaling by 29% Over Baseline by 27% Over Greedy by 17% 29% 17% 21% 16% 28% 15%

12% Evaluation Results on Performance Optimizations 21 Performance optimization results. ToF obtains the lowest average execution time on all workflows. Over Auto-scaling by 21% Over Baseline by 21% Over Greedy by 18% 21% 18% 21% 8% 16%

22 Outline Related Work Generalized Optimization Frameworks –General transformations for cost and performance optimizations –A declarative language for workflow optimization problems Conclusions

Deco: A Declarative Optimization Framework Outline –Main contributions of this work –System overview –A declarative language for workflows –GPU-accelerated search engine –Evaluation results 23

Main Contributions This work has three main contributions –A declarative language for resource provisioning of scientific workflows in IaaS clouds –A generalized optimization framework to serve a wide range of optimization problems –Fast GPU-based implementation for low optimization overhead 24

Motivating Ideas Why declarative language? –Declarative languages like HTML, SQL, Prolog –Concise and clear –Focus on what to do rather than how to do it Why GPU acceleration? –Generic search has large runtime overhead –Monte Carlo method is used for probabilistic approximation [Raedt et al. 2007] which is suitable for GPU acceleration 25

System Overview Overview of the Deco system –WLog, a declarative language for workflows –GPU-Accelerated search engine 26

WLog – A Declarative Language for Workflows WLog is designed based on Prolog A WLog program describing a workflow scheduling problem 27 goal minimize Ct in totalcost(Ct). cons deadline(95%, 10h). var configs(Tid, Vid) forall task(Tid) and Vm(Vid). r1 import(amazonec2). r2 import(montage). r3 path(X,Y,Y,C) :- edge(X,Y), exetime(X,Vid,T), C is T. r4 path(X,Y,Z,C) :- edge(X,Z), Zn==Y, path(Z,Y,Z2,C1), exetime(X,Vid,T), C is T+C1. r5 maxtime(Path,T) :- setof([Z,C],path(root,tail,Z,C),Set), max(Set,[Path,T]). r6 cost(Tid,Vid,C) :- price(Vid,Up), exetime(Tid,Vid,T), C is ceil(T/60.0)*Up. r7 totalcost(Ct) :- findall(C,cost(Tid,Vid,C),Bag), sum(Bag,Ct). problem specific keywords: goal Optimization goal defined by the user. cons Problem constraint defined by the user. var Problem variable to be optimized. deadline(P, D) A probabilistic deadline requirement that D is at the P-th percentile of workflow execution time. import(cloud) Import the cloud-related facts from the cloud metadata. import(daxfile) Import the workflow-related facts generated from a DAX file.

GPU Accelerations Explore vs. exploit –By exploit, partial results are prioritized. –Exploration traverses the search tree level by level which offers GPU a opportunity to parallel the searching process. Memory optimizations –Minimize the usage of global memory –Reduce accesses to shared memory 28

Evaluation Settings Three use cases –Workflow scheduling problem –Workflow ensemble [Malawski et al., SC’12] Goal: execute more workflows with high priorities within given budget and deadline –Follow-the-cost: multiple workflows, multiple datacenters Comparison for workflow ensemble problem –Algorithms: Deco vs. SPSS [Malawski et al., SC’12] –Ensemble types: constant, Uniform(Un)sorted, Pareto(Un)sorted –Generate 5 budgets between [MinBudget, MaxBudget] All results are normalized to that of SPSS 29

Evaluation Results Under all ensemble types and budget constraints –Deco obtains better score metric value than SPSS 30 Obtained score results of SPSS and Deco with different ensemble types under budget 1 to 5 and fixed deadline. Workflow type is Ligo.

Evaluation Results (cont’d) Programmability of WLog in Deco (lines of codes) –Users (re-)implement the workflow application in C++. –With Deco, users implement in WLog. 31 Use CaseC++ Implementation WLog Workflow Scheduling Workflow Ensemble Follow-the-Cost Deco allows much lower coding complexity than manual implementation.

Performance Speedup of GPUs 32 Performance speedup of GPU implementation over CPU implementation on a single core for the three applications 437x 93x 31x

33 Outline Related Work Generalized Optimization Frameworks –General transformations for cost and performance optimizations –A declarative language for workflow optimization problems Conclusions

34 Conclusions IaaS clouds have become an attractive platform for hosting workflows. Despite recent efforts in monetary cost optimizations of workflows in the cloud, there is still a large room for further improvements. Due to the complex cloud offerings and problem specifications, we develop general optimization frameworks. –ToF achieves up to 29% improvement over the state-of- the-art algorithm. –Deco achieves up to 77% improvement over the state-of- the-art algorithm.

Future Work Energy-efficient Cloud –Reduce the investment cost of cloud provider to potentially reduce instance price with energy-efficient hardware/software Optimization opportunities in Multi-Cloud –Utilize different cloud offerings, e.g., instance types, to further reduce cost 35

References Maciej Malawski, Gideon Juve, Ewa Deelman, and Jarek Nabrzyski Cost- and deadline- constrained provisioning for scientific workflow ensembles in IaaS clouds. SC ' pages. Juve, G.; Deelman, E.; Vahi, K.; Mehta, G.; Berriman, B.; Berman, B.P.; Maechling, P., "Scientific workflow applications on Amazon EC2," E-Science Workshops, pp.59,66, 9-11 Dec Jens-Sönke Vöckler, Gideon Juve, Ewa Deelman, Mats Rynge, and Bruce Berriman Experiences using cloud computing for a scientific workflow application. ScienceCloud '11. P15-P Ming Mao, Marty Humphrey: Auto-scaling to minimize cost and meet application deadlines in cloud workflows. SC 2011: 49. Herald Kllapi, Eva Sitaridi, Manolis M. Tsangaris, and Yannis Ioannidis Schedule optimization for data processing flows on the cloud. SIGMOD ' Anshul Rai, Ranjita Bhagwan, and Saikat Guha Generalized resource allocation for the cloud. SoCC '12. Article 15, 12 pages. Changbin Liu, Lu Ren, Boon Thau Loo, Yun Mao, and Prithwish Basu Cologne: a declarative distributed constraint optimization platform. Proc. VLDB Endow. 5, L. De Raedt, A. Kimmig, and H. Toivonen, ProbLog: A probabilistic Prolog and its application in link discovery, IJCAI 2007, pages , Amelie Chi Zhou, Bingsheng He, Transformation-based Monetary Cost Optimizations for Workflows in the Cloud, accepted by TCC, Dec Amelie Chi Zhou, Bingsheng He, A declarative optimization framework for workflows in IaaS clouds, submitted to SC Amelie Chi Zhou, Bingsheng He, Cheng Liu, Monetary Cost Optimizations for Hosting Workflow-as-a- Service in IaaS Clouds, submitted to ToC,

37 Thank you! Amelie Chi Zhou Advisor: Bingsheng He Xtra Computing Group Nanyang Technological University, Singapore