In Cloud, Can Scien/fic Communi/es Benefit from the Economies of Scale?

Size: px
Start display at page:

Download "In Cloud, Can Scien/fic Communi/es Benefit from the Economies of Scale?"

Transcription

1 In Cloud, Can Scien/fic Communi/es Benefit from the Economies of Scale? By: Lei Wang, Jianfeng Zhan, Weisong Shi, Senior Member, IEEE, and Yi Liang Presented By: Ramneek (MT )

2 Introduc/on The power of cloud compu/ng is elas/city the ability to pay for resources only when they are needed and to scale infrastructure up and down on demand, without addi/onal wai/ng /me. In this paper, we intend to answer one key ques/on to the success of cloud compu/ng: in cloud, can small- to- medium scale scien0fic communi0es benefit from the economies of scale? Economy of Scale: refers to reduc/ons in unit cost as the size of a facility increases.

3 Introduc/on Differences between scien/fic compu/ng workloads and the ini/al target workload of clouds: in workload characteris/cs in resource consump/on in performance goals In scien/fic communi/es, more and more research groups show great interests in private clouds or proposing hybrid cloud models to augment their local compu/ng resources with external public clouds. However, this paper focuses on the public clouds.

4 Introduc/on Types of clouds: Public cloud: Hosted at service provider s site. Supports mul/ple customers. Supports connec/vity over internet. Suited for informa/on that is not sensi/ve. Can be cheaper than the private cloud. Private cloud: Hosted at the enterprise or at the service provider s site. Supports a single customer. Connec/vity over private network/fiber or internet. Suited for informa/on that needs a high level of security.

5 Introduc/on Hybrid cloud: A hybrid cloud is a composi/on of at least one private cloud and at least one public cloud. For e.g., an organiza/on might use a public cloud service, such as Amazon Simple Storage Service (Amazon S3) for shared data but con/nue to maintain in- house storage for personal data.

6 This paper focuses on public clouds because: They are not restricted within an enterprise. They offer wide availability and economies of scale. Cheaper. There is a need to propose innova/ve cloud usage models, and build enabling systems that support scien/fic communi/es to benefit from the economies of scale of public clouds.

7 Overview This paper addresses three main issues: First, it proposes an innova/ve cloud usage model, called an Enhanced Scien/fic Public cloud (ESP) model. Second, on a basis of the ESP model, it presents the design of the Dawning- Cloud system. Third, it presents the evalua/on of the system comprehensively using both emula/on and real experiments.

8 Overview Deelman s System: each client/user in the organiza/on directly leases virtual machine resources from public cloud provider in a specified period. Evangelinos s System: the organiza/on as a whole rents resources with the fixed size from a public cloud to create a virtual cluster system. Dawning Cloud system: the common management service framework (CSF) is pre- deployed at the resource provider s site and thin run/me environment (TRE) layer is built on top of it according to service provider s workload requirements.

9 Enhanced Scien/fic Public Cloud Model (ESP) Features: Developing a new service management layer (TRE) for another different workload is lightweight. Second, the service management tasks become simpler, e.g., for a TRE, we only need to deploy fewer modules, since other ones are delegated to the CSF, which is pre- deployed on the cloud site Third, this model provides flexible system controls, and a service provider can ac/vate, safe- deac/vate, deac/vate, suspend, and resume TRE at its own need.

10 ESP TABLE 1: THE COMPARISONS OF DIFFERENT USAGE MODELS item a DS PC HC DP EP ESP SP vs. RP b 1:1 1:1 n:1 n:1 n:1 n:1 Separation No No No No Yes Yes of concerns resources local local local + rented rented rented rented resource provisioning fixed fixed fixed + elastic elastic fixed initial + dynamic SP s system control No No No limited limited flexible a. DS: dedicated system; PC: private cloud; HC: hybrid cloud; DP: Deelman s public cloud; EP: Evangelinos s public cloud; ESP : our model. b. SP: service provider; RP: resource provider

11 ESP

12 ESP As a high- level management en/ty, a TRE has five different states: ini/al, deployed, running, deac/vated, and suspended: The ini/al state indicates that the service provider and the resource provider are in the process of planning the TRE. At the request of a service provider, the resource provider can create a TRE. The crea/ng opera/on turns the state of TRE from ini/al to deployed. The deployed state indicates that the TRE is configured and deployed.

13 ESP When the service provider ac/vates the TRE that is deployed, the laaer turns to the running state. The deac/va/on opera/on has two- fold effects: the running jobs will be killed and the user data and programs will be permanently kept in the home directories unless the service provider explicitly destroys its TRE. The suspended state indicates that the TRE will reject the submissions of new jobs, however it will finish the submiaed jobs.

14 Dawning Cloud: Architecture Thin Run/me Environment (TRE) TRE Management and Programming API Common management Service Framework (CSF) A Public Cloud Site

15 Dawning Cloud: Architecture Layered Architecture Built using boaom- up approach Lower layer: CSF facilitates building thin service management layers- TRE for heterogeneous workloads: Provides a uniform view of virtualized or physical resources Lifecycle management mechanisms of TRE Enforcing resource provisioning policy Virtual machines or physical nodes provisioning Middle Layer: Management and Programming API for building TREs. Upper Later: TRE implements core func/ons for a specific workload.

16 Reservoir: Architecture Service Abstrac/on Layer (Claudia) OCCI Interface Open Nebula Core EC2 plugins Elas/cHost plugins Xen Drivers EC2 EC2 Elas/cHost Internal Datacenter

17 Reservoir Architecture The service abstrac/on layer interacts with the service providers to receive their Service Manifests, nego/ate pricing, and handle billing. Claudia: it is an IaaS solu/on created by Telefonica R&D aimed at managing services by providing an abstrac/on layer on top of both private and public Virtualiza/on Providers. OCCI (Open Cloud Compu/ng Interface) is a Protocol and API for all kinds of Management tasks. Open Nebula Core: The core consists of a set of components to control and monitor virtual machines, virtual networks, storage and hosts. The core performs its ac/ons (e.g. monitor a host, or cancel a VM) by invoking a suitable driver.

18 Difference between Reservoir and the Dawning Cloud 1. As compared to the Reservoir, the CSF of Dawning Cloud facilitates building lightweight service management layers for heterogeneous workloads. The following figure shows a typical dawning cloud system:

19 Difference between Reservoir and the Dawning Cloud Reservoir project integrates sohware packages from different partners: OpenNebula a virtual infrastructure manager Claudia a service management layer. 2. Unlike Reservoir, the implementa/on of TRE is not bound to virtual machines, e.g., XEN or VMware.

20 Dynamic Resource Nego/a/on Mechanism e dynamic resource negotiation mechanism.

21 Dynamic Resource Nego/a/on Mechanism Resource management policy: Service provider specifies its requirement for resource management in a resource management policy. A resource management policy defines the behavior specifica/on of the HTC or MTC server i.e., the server resizes resources to what an extent according to what criterion. Resource provision service: It is responsible for provisioning resources to different TRE. Resource provision policy: Used by a resource provider specifies its requirement for resource provisioning.

22 Dynamic Resource Nego/a/on Mechanism Setup policy: It determines when and how to do the setup work. Each /me a node is assigned or reclaimed, a setup policy is triggered. Life cycle management service: responsible for managing life cycles of TRE and performs the setup work.

23 Resource Management and Provisioning Policies In Dawning Cloud, we dis/nguish two types of resources provisioned to a TRE: Ini/al resources: ini/al resources, once allocated, will not be reclaimed by the resource provider un/l the TRE is destroyed. Dynamic resources: dynamic resources assigned to a TRE may be reclaimed by the resource provider.

24 Resource Management and Provisioning Policies In Dawning Cloud, a service provider and a resource provider need to set four parameters: 1. The size of ini/al resources. 2. The /me unit of leasing resources. 3. The checking resource cycle. It is a periodical /mer that the HTC or MTC server checks jobs in the queue. 4. The threshold ra/o of obtaining dynamic resources.

25 Start yes No yes

26 Resource Management and Provisioning Policies This paper proposes a resource management policy for an HTC or MTC service provider as follows: At the startup of a TRE, the service provider will request ini/al resources with the specified size. Ra/o of obtaining dynamic resources = (accumulated resource demands of all jobs in the queue )/( current resources owned by a TRE) If this ra/o> 1, the server will request Dynamic Resources (DR) as follows: DR = (the accumulated resources demand of all jobs in the queue) (the current resources owned by the TRE).

27 Resource Management and Provisioning Policies Aher obtaining DR, the server registers a new periodical /mer and checks idle DR per /me unit of leasing resources. If size of idle resources < DR, the server release resources with size DR = (DR- idle DR). If size of idle resources >= DR, the server will release resources with size=dr and deregister the /mer.

28 Evalua/on Methodologies and Experiments Workloads Evalua/on Methodologies Evalua/on Metrics Emula/on Experiment Configura/ons Resource Configura/ons Scheduling policies Resource management and provisioning policies System Level Evalua/on

29 Evalua/on Methodologies and Experiments: Workloads HTC workload traces: Lower load: NASA ipse trace (46.6% u/liza/on). For the NASA trace, the /me dura/on is three months, the average execu/on /me is 764 seconds, and the total number of jobs is Higher loads: SDSC BLUE trace (76.2% u/liza/on). For the SDSC trace, the /me dura/on is 32 months, the average execu/on /me is 4381 seconds, and the total number of jobs is LLNL Thunder trace (86.5% u/liza/on). For the LLNL trace, the /me dura/on is five months, the average execu/on /me is 2227 seconds, and the total number of jobs is

30 Evalua/on Methodologies and Experiments: Workloads MTC workload traces: Montage workflow Montage is an astronomy workflow applica/on, created by NASA/IPAC Infrared Science Archive for gathering mul/ple input images to create custom mosaics of the sky. The chosen Montage workload includes 9 types of task with the total amount of 1,000 jobs. Each job requests one node for running, and the average execu/on /me of jobs is seconds.

31 Evalua/on Methodologies and Experiments: evalua/on methodology Mainly concerned with the public cloud Emula/on methodology is used. Resource provider deploys Dawning Cloud, Deelman s system, Evangelinos s system, and a dedicated cluster system. Emula/on is deployed on testbed with nodes of following configura/on: 1.6 GH processor 2 GB memory CentOS 5.0 Opera/ng System

32 Evalua/on Methodologies and Experiments: Evalua/on Metrics For the service provider, we choose the resource consump/on in terms of node*hours. In Deelman s system, there is no role of a service provider, so we calculate the ac- cumulated resource consump/on of all end users, which amounts to the cost of a service provider in other models. For the dedicated cluster system, since a service provider owns resources, we calculate the resource consump/on of a service provider as the product of the configura/on size of a dedicated cluster system and the dura/on of a certain period.

33 Evalua/on Methodologies and Experiments: Scheduling policies The aim of this paper is not to evaluate the effect of different scheduling policies. For Dawning Cloud, Evangelinos, and dedicated cluster systems, following policies are used: For HTC server: First Fit policy is used For MTC server: First Come First Serve (FCFS) is used. Deelman s system uses no scheduling policy, since all jobs run immediately without queuing.

34 Evalua/on Methodologies and Experiments: Evalua/on The Metrics of the Service Provider Using Four Systems for the NASA Trace.

35 Evalua/on Methodologies and Experiments: Evalua/on For the NASA, SDSC, and LLNL traces, in comparison with the dedicated cluster system or Evangelinos s system, service providers in Dawning Cloud save the resource consump/on maximally by 25.6% and minimally by 7.9%. This is because service providers in Dawning Cloud can resize resources according to varying workload status. We also find that the the saved resource consump/on is inversely propor/onal to the resource u/liza/on reported in the traces, this is because that high u/liza/on implies less space for saving resources.

36 Evalua/on Methodologies and WORKFLOW. Configuration Experiments: Evalua/on resource consumption (node*hour) saved resources tasks per second dedicated cluster / system Evangelinos s system Deelman s system % DawningCloud % THE METRICS OF THE SERVICE PROVIDER RUNNING FOUR SYSTEMS FOR THE MONTAGE WORKFLOW.

37 Evalua/on Methodologies and Experiments: Evalua/on In comparison with the dedicated cluster system or Evangelinos s system, the service provider in the Dawning Cloud saves the resource consump/on by 67.5%. This is because Dawning Cloud owns ini/al resources with the smaller size, and resizes dynamic resources driven by the change of workload status.

38 Evalua/on Methodologies and Experiments: Evalua/on The total resource consump/ons of the resource provider using four different systems.

39 Evalua/on Methodologies and Experiments: Evalua/on The management overhead of the resource provider.

40 Evalua/on Methodologies and Experiments: Evalua/on For the dedicated cluster system, since the resource provider owns resource, it has no management overhead in terms of obtaining dynamic resources. Evangelinos s system has the lowest management overhead, since it leases resources with the fixed dura/on. Dawning Cloud has smaller management overhead than that of Deelman s system, since the ini/al resources will not be reclaimed un/l a TRE is destroyed.

41 Conclusion For HTC and MTC workloads, we can conclude that that small- to- medium scale scien/fic communi/es indeed can benefit from the economies of scale of public clouds with the support of the enabling system Dawning Cloud.

42 Conclusion Scien/sts may prefer the ownership of virtual resources, since this reduces the uncertainty concerning access when needed and to have a beaer control over the virtualized resources which become difficult due to mul/- tenant architecture in case of public clouds. Studies have shown that private clouds like Nimbus incurred less variability and showed a beaer compu/ng performance due to the underlying hardware. The public cloud providers may not easily agree to implement these resource management architectures as they incur a large management overhead.

In Cloud, Can Scientific Communities Benefit from the Economies of Scale?

In Cloud, Can Scientific Communities Benefit from the Economies of Scale? PRELIMINARY VERSION IS PUBLISHED ON SC-MTAGS 09 WITH THE TITLE OF IN CLOUD, DO MTC OR HTC SERVICE PROVIDERS BENEFIT FROM THE ECONOMIES OF SCALE? 1 In Cloud, Can Scientific Communities Benefit from the

More information

In Cloud, Do MTC or HTC Service Providers Benefit from the Economies of Scale?

In Cloud, Do MTC or HTC Service Providers Benefit from the Economies of Scale? In Cloud, Do MTC or HTC Service Providers Benefit from the Economies of Scale? Lei Wang, Jianfeng Zhan Chinese Academy of Sciences {wl, jfzhan}@ncic.ac.cn Weisong Shi Wayne State University weisong@wayne.edu

More information

Failure Predic,on of Jobs in Compute Clouds: A Google Cluster Case Study

Failure Predic,on of Jobs in Compute Clouds: A Google Cluster Case Study Failure Predic,on of Jobs in Compute Clouds: A Google Cluster Case Study Xin Chen, and Karthik Pa0abiraman University of Bri:sh Columbia (UBC) Charng- da Lu, Unaffiliated Compute Clouds } Infrastructure

More information

Cloud Platforms. Various types and their properties. Prof. Balwinder Sodhi. 1 Computer Science and Engineering, IIT Ropar

Cloud Platforms. Various types and their properties. Prof. Balwinder Sodhi. 1 Computer Science and Engineering, IIT Ropar Cloud Platforms Various types and their properties Prof. Balwinder Sodhi 1 Computer Science and Engineering, IIT Ropar Cloud Classification Service model based Depends on the cloud services being offered

More information

IBM ICE (Innovation Centre for Education) Welcome to: Unit 1 Overview of delivery models in Cloud Computing. Copyright IBM Corporation

IBM ICE (Innovation Centre for Education) Welcome to: Unit 1 Overview of delivery models in Cloud Computing. Copyright IBM Corporation Welcome to: Unit 1 Overview of delivery models in Cloud Computing 9.1 Unit Objectives After completing this unit, you should be able to: Understand cloud history and cloud computing Describe the anatomy

More information

Experience with Adap0ng a WS BPEL Run0me for escience Workflows

Experience with Adap0ng a WS BPEL Run0me for escience Workflows Experience with Adap0ng a WS BPEL Run0me for escience Workflows Thilina Gunarathne, Chathura Herath, Eran Chinthaka, Suresh Marru Pervasive Technology Ins0tute Indiana University Introduc0on Scien0st communi0es

More information

Enterprise APM version 4.2 FAQ

Enterprise APM version 4.2 FAQ Meridium Enterprise Asset Performance Management (APM) version 4.2 is the next generation APM solution that helps your company simply and easily connect disparate systems and use that data to create and

More information

BUILDING A PRIVATE CLOUD

BUILDING A PRIVATE CLOUD INNOVATION CORNER BUILDING A PRIVATE CLOUD How Platform Computing s Platform ISF* Can Help MARK BLACK, CLOUD ARCHITECT, PLATFORM COMPUTING JAY MUELHOEFER, VP OF CLOUD MARKETING, PLATFORM COMPUTING PARVIZ

More information

Process Modeling Best Practices. Raphael Derbier Nicolas Marzin

Process Modeling Best Practices. Raphael Derbier Nicolas Marzin Process Modeling Best Practices Raphael Derbier Nicolas Marzin SAFE HARBOR DISCLOSURE During the course of this presenta1on TIBCO or its representa1ves may make forward-looking statements regarding future

More information

11435 CICS Pla,orm and Applica6ons Basics

11435 CICS Pla,orm and Applica6ons Basics 11435 CICS Pla,orm and Applica6ons Basics Ma:hew Webster ma:hew_webster@uk.ibm.com Cloud- style CICS development, deployment, and opera6ons Sessions SHARE Lunch & Learn CICS Transac6on Server V5.1 open

More information

Case Study BONUS CHAPTER 2

Case Study BONUS CHAPTER 2 BONUS CHAPTER 2 Case Study ABC is a large accounting firm with customers in five countries across North America and Europe. Its North American headquarters is located in Miami, Florida, where it hosts

More information

A Examcollection.Premium.Exam.35q

A Examcollection.Premium.Exam.35q A2030-280.Examcollection.Premium.Exam.35q Number: A2030-280 Passing Score: 800 Time Limit: 120 min File Version: 32.2 http://www.gratisexam.com/ Exam Code: A2030-280 Exam Name: IBM Cloud Computing Infrastructure

More information

Understanding Business Value of Life Cycle Management. Sumeet Mehra Application & Management Solution Specialist 6 th November 2008

Understanding Business Value of Life Cycle Management. Sumeet Mehra Application & Management Solution Specialist 6 th November 2008 Understanding Business Value of Life Cycle Management Sumeet Mehra Application & Management Solution Specialist 6 th November 2008 Disclaimer This session may contain product features that are currently

More information

Cisco Intelligent Automation for Cloud

Cisco Intelligent Automation for Cloud Data Sheet Cisco Intelligent Automation for Cloud Introduction IT is under increasing pressure to deliver services to the business more quickly and inexpensively than ever before. Fortunately, a new solution,

More information

Autonomic Provisioning and Application Mapping on Spot Cloud Resources

Autonomic Provisioning and Application Mapping on Spot Cloud Resources Autonomic Provisioning and Application Mapping on Spot Cloud Resources Daniel J. Dubois, Giuliano Casale Imperial College London, Department of Computing 2015 International Conference on Cloud and Autonomic

More information

Azure Stack. Unified Application Management on Azure and Beyond

Azure Stack. Unified Application Management on Azure and Beyond Azure Stack Unified Application Management on Azure and Beyond Table of Contents Introduction...3 Deployment Models...4 Dedicated On-Premise Cloud... 4 Shared Application Hosting... 4 Extended Hosting

More information

Bluemix Overview. Last Updated: October 10th, 2017

Bluemix Overview. Last Updated: October 10th, 2017 Bluemix Overview Last Updated: October 10th, 2017 Agenda Overview Architecture Apps & Services Cloud Computing An estimated 85% of new software is being built for cloud deployment Cloud Computing is a

More information

Getting Started with vrealize Operations First Published On: Last Updated On:

Getting Started with vrealize Operations First Published On: Last Updated On: Getting Started with vrealize Operations First Published On: 02-22-2017 Last Updated On: 07-30-2017 1 Table Of Contents 1. Installation and Configuration 1.1.vROps - Deploy vrealize Ops Manager 1.2.vROps

More information

WHITE PAPER. CA Nimsoft APIs. keys to effective service management. agility made possible

WHITE PAPER. CA Nimsoft APIs. keys to effective service management. agility made possible WHITE PAPER CA Nimsoft APIs keys to effective service management agility made possible table of contents Introduction 3 CA Nimsoft operational APIs 4 Data collection APIs and integration points Message

More information

Cloud Management Platform Overview First Published On: Last Updated On:

Cloud Management Platform Overview First Published On: Last Updated On: Cloud Management Platform Overview First Published On: 06-09-2016 Last Updated On: 07-25-2017 1 Table of Contents 1. Cloud Management Platform Overview 1.1.Cloud Consumer Request/Catalog 1.2.Cloud Admin

More information

HPC Work)low Performance

HPC Work)low Performance Slide 1 HPC Work)low Performance Karen L. Karavanic New Mexico Consortium & Portland State University David Montoya (LANL) August 2, 2016 What is an HPC Work)low? Applica'on View Slide 2 Run$me system

More information

PRIORITY BASED SCHEDULING IN CLOUD COMPUTING BASED ON TASK AWARE TECHNIQUE

PRIORITY BASED SCHEDULING IN CLOUD COMPUTING BASED ON TASK AWARE TECHNIQUE RESEARCH ARTICLE OPEN ACCESS PRIORITY BASED SCHEDULING IN CLOUD COMPUTING BASED ON TASK AWARE TECHNIQUE Jeevithra.R *, Karthikeyan.T ** * (M.Phil Computer Science Student Department of Computer Science)

More information

HP Cloud Maps for rapid provisioning of infrastructure and applications

HP Cloud Maps for rapid provisioning of infrastructure and applications Technical white paper HP Cloud Maps for rapid provisioning of infrastructure and applications Table of contents Executive summary 2 Introduction 2 What is an HP Cloud Map? 3 HP Cloud Map components 3 Enabling

More information

Tetris: Optimizing Cloud Resource Usage Unbalance with Elastic VM

Tetris: Optimizing Cloud Resource Usage Unbalance with Elastic VM Tetris: Optimizing Cloud Resource Usage Unbalance with Elastic VM Xiao Ling, Yi Yuan, Dan Wang, Jiahai Yang Institute for Network Sciences and Cyberspace, Tsinghua University Department of Computing, The

More information

SAP Journey to Virtualization and Cloud. by Ganesh Radhakrishnan, CEO WFT

SAP Journey to Virtualization and Cloud. by Ganesh Radhakrishnan, CEO WFT SAP Journey to Virtualization and Cloud by Ganesh Radhakrishnan, CEO WFT AGENDA v About Presenter (Ganesh Radhakrishnan) v CIO s Pain Point v SAP Virtualiza=on / Cloud Decision v Is SAP Virtualiza=on or

More information

Preemptive, Low Latency Datacenter Scheduling via Lightweight Virtualization

Preemptive, Low Latency Datacenter Scheduling via Lightweight Virtualization Preemptive, Low Latency Datacenter Scheduling via Lightweight Virtualization Wei Chen, Jia Rao*, and Xiaobo Zhou University of Colorado, Colorado Springs * University of Texas at Arlington Data Center

More information

1 Copyright 2011, Oracle and/or its affiliates. All rights reserved.

1 Copyright 2011, Oracle and/or its affiliates. All rights reserved. 1 Copyright 2011, Oracle and/or its affiliates. All rights ORACLE PRODUCT LOGO Virtualization and Cloud Deployments of Oracle E-Business Suite Ivo Dujmović, Director, Applications Development 2 Copyright

More information

[Header]: Demystifying Oracle Bare Metal Cloud Services

[Header]: Demystifying Oracle Bare Metal Cloud Services [Header]: Demystifying Oracle Bare Metal Cloud Services [Deck]: The benefits and capabilities of Oracle s next-gen IaaS By Umair Mansoob Introduction As many organizations look to the cloud as a way to

More information

CLOUD computing and its pay-as-you-go cost structure

CLOUD computing and its pay-as-you-go cost structure IEEE TRANSACTIONS ON PARALLEL AND DISTRIBUTED SYSTEMS, VOL. 26, NO. 5, MAY 2015 1265 Cost-Effective Resource Provisioning for MapReduce in a Cloud Balaji Palanisamy, Member, IEEE, Aameek Singh, Member,

More information

Cloud Computing Concept, Technology & Architecture

Cloud Computing Concept, Technology & Architecture Cloud Computing Concept, Technology & Architecture Chapter 09 Cloud Management Mechanics 課程名稱 : 雲端管理系統 授課教師 : 高勝助 Contents Cloud-based IT resources need to be set up, configured, maintained, and monitored.

More information

Autonomic Provisioning and Application Mapping on Spot Cloud Resources

Autonomic Provisioning and Application Mapping on Spot Cloud Resources Autonomic Provisioning and Application Mapping on Spot Cloud Resources Daniel J. Dubois, GiulianoCasale 2015 IEEE International Conference on Cloud and Autonomic Computing (ICCAC), Cambridge, Massachusetts,

More information

Automating Capacity Management in Virtual, Cloud & Software Defined Datacenters

Automating Capacity Management in Virtual, Cloud & Software Defined Datacenters Automating Capacity Management in Virtual, Cloud & Software Defined Datacenters Andrew Hillier CTO, CiRBA Inc. Capacity Challenges Increasingly Dynamic Access to Alternatives Complex Low Utilization Application

More information

How to develop Data Scientist Super Powers! Using Azure from R to scale and persist analytic workloads.. Simon Field

How to develop Data Scientist Super Powers! Using Azure from R to scale and persist analytic workloads.. Simon Field How to develop Data Scientist Super Powers! Using Azure from R to scale and persist analytic workloads.. Simon Field Topics Why cloud Managing cloud resources from R Highly parallelised model training

More information

Top six performance challenges in managing microservices in a hybrid cloud

Top six performance challenges in managing microservices in a hybrid cloud Top six performance challenges in managing microservices in a hybrid cloud Table of Contents Top six performance challenges in managing microservices in a hybrid cloud Introduction... 3 Chapter 1: Managing

More information

Cloud OS Customer-Ready Services

Cloud OS Customer-Ready Services Cloud OS Customer-Ready Services ON-PREMISES CONSISTENT 1PLATFORM MICROSOFT SERVICE PROVIDER Web Platform application Services (PaaS) Infrastructure Services (IaaS) Reliable messaging Virtual Networking

More information

An Econocom Group company. Your partner in the transi8on towards Mobile IT

An Econocom Group company. Your partner in the transi8on towards Mobile IT E c o n o c o m Te l e c o m S e r v i c e s An Econocom Group company Your partner in the transi8on towards Mobile IT E c o n o c o m Te l e c o m S e r v i c e s A few key figures 40 000 mobile terminals

More information

10 Ways Oracle Cloud Is Better Than AWS

10 Ways Oracle Cloud Is Better Than AWS 10 Ways Oracle Cloud Is Better Than AWS BY UMAIR MANSOOB Who Am I Oracle Certified Administrator from Oracle 7 12c Exadata Certified Implementation Specialist since 2011 Oracle Database Performance Tuning

More information

TIM158 Business Informa3on Strategy

TIM158 Business Informa3on Strategy TIM158 Business Informa3on Strategy Instructor: Safwan Shah Teaching Assistant: Paul Vroomen To maintain consistency. Lectures throughout TIM158 adapted or borrowed from Kevin Ross. Addi3onal material

More information

Cisco Intelligent Automation for Cloud

Cisco Intelligent Automation for Cloud Lifecycle Management Data Sheet Cisco Intelligent Automation for Cloud Introduction Enterprise IT leaders are under increasing pressures to deliver services to the business more quickly and inexpensively

More information

ENTERPRISE HYBRID CLOUD 4.1.1

ENTERPRISE HYBRID CLOUD 4.1.1 ENTERPRISE HYBRID CLOUD 4.1.1 EMC Solutions Abstract This document describes how to deploy an SAP environment on an Enterprise Hybrid Cloud with VMware vrealize Automation as its core. This solution addresses

More information

Supporting Data-intensive Workflows in Software-defined Federated Multi-Clouds

Supporting Data-intensive Workflows in Software-defined Federated Multi-Clouds 1 Supporting Data-intensive Workflows in Software-defined Federated Multi-Clouds Javier Diaz-Montes 1, Manuel Diaz-Granados 1, Mengsong Zou 1, Shu Tao 2 and Manish Parashar 1 1 Rutgers Discovery Informatics

More information

Monitoring- based Commissioning: Con2nuous Performance Inves2ga2on and Evalua2on

Monitoring- based Commissioning: Con2nuous Performance Inves2ga2on and Evalua2on Monitoring- based Commissioning: Con2nuous Performance Inves2ga2on and Evalua2on Jessica Granderson PhD Lawrence Berkeley Na2onal Laboratory Energy Informa2on Systems and the Energy Informa-on Handbook

More information

A hybrid auto-scaling technique for clouds processing applications with service level agreements

A hybrid auto-scaling technique for clouds processing applications with service level agreements Biswas et al. Journal of Cloud Computing: Advances, Systems and Applications (2017) 6:29 DOI 10.1186/s13677-017-0100-5 Journal of Cloud Computing: Advances, Systems and Applications RESEARCH A hybrid auto-scaling

More information

Understanding Cloud. #IBMDurbanHackathon. Presented by: Britni Lonesome IBM Cloud Advisor

Understanding Cloud. #IBMDurbanHackathon. Presented by: Britni Lonesome IBM Cloud Advisor Understanding Cloud #IBMDurbanHackathon Presented by: Britni Lonesome IBM Cloud Advisor What is this thing called cloud? Cloud computing is a new consumption and delivery model inspired by consumer internet

More information

The Cloud at Your Service

The Cloud at Your Service C 1 The Cloud at Your Service loud computing is a way to use and share hardware, operating systems, storage, and network capacity over the Internet. Cloud service providers rent virtualized servers, storage,

More information

IT Business Management Standard Edition User's Guide

IT Business Management Standard Edition User's Guide IT Business Management Standard Edition User's Guide VMware IT Business Management Suite 1.0.1 This document supports the version of each product listed and supports all subsequent versions until the document

More information

CLOUD COMPUTING- A NEW EDGE TO TECHNOLOGY

CLOUD COMPUTING- A NEW EDGE TO TECHNOLOGY CLOUD COMPUTING- A NEW EDGE TO TECHNOLOGY Prof. Pragati Goel Asso. Prof. MCA Dept., Sterling Institute of Management Studies, Nerul, Navi Mumbai. Navi Mumbai, India Email: goelpragati78@gmail.com The Cloud

More information

Optimizing resource efficiency in Microsoft Azure

Optimizing resource efficiency in Microsoft Azure Microsoft IT Showcase Optimizing resource efficiency in Microsoft Azure By July 2017, Core Services Engineering (CSE, formerly Microsoft IT) plans to have 90 percent of our computing resources hosted in

More information

Deep Learning Acceleration with

Deep Learning Acceleration with Deep Learning Acceleration with powered by A Technical White Paper TABLE OF CONTENTS The Promise of AI The Challenges of the AI Lifecycle and Infrastructure MATRIX Powered by Bitfusion Flex Solution Architecture

More information

Challenges for making scalable security management for informa5on and communica5on infrastructure

Challenges for making scalable security management for informa5on and communica5on infrastructure Challenges for making scalable security management for informa5on and communica5on infrastructure Prof. Dr. Suguru Yamaguchi Graduate School of Informa5on Science, Nara Ins5tute of Science and Technology,

More information

CHAPTER. Introduction

CHAPTER. Introduction CHAPTER 1 In recent years, there has been a race by both traditional Service Providers (SPs) and public cloud providers such as Amazon to capture the cloud services market. SPs have identified the capability

More information

IT System Scope Development. Presented by Lourdes Coss, MPA, CPPO

IT System Scope Development. Presented by Lourdes Coss, MPA, CPPO IT System Scope Development Presented by Lourdes Coss, MPA, CPPO Objec4ves Discuss Key Components of an IT System Scope of Services Prac?ce the Development of the Document Discuss some of the laws of teamwork

More information

Table of Contents HOL CMP

Table of Contents HOL CMP Table of Contents Lab Overview - - vrealize Business for Cloud - Getting Started... 2 Lab Guidance... 3 Module 1 - Computing the Cost of your Private Cloud (30 Minutes)... 9 Introduction... 10 Overview

More information

SAS Applica?ons with Oracle Exadata and Big Data Appliance: Turning Data into Knowledge

SAS Applica?ons with Oracle Exadata and Big Data Appliance: Turning Data into Knowledge SAS Applica?ons with Oracle Exadata and Big Data Appliance: Turning Data into Knowledge May 3, 2016 Exadata SIG Virtual Conference Mathew Steinberg Exadata Product Management Oracle Database Development

More information

Special thanks to Chad Diaz II, Jason Montgomery & Micah Torres

Special thanks to Chad Diaz II, Jason Montgomery & Micah Torres Special thanks to Chad Diaz II, Jason Montgomery & Micah Torres Outline: What cloud computing is The history of cloud computing Cloud Services (Iaas, Paas, Saas) Cloud Computing Service Providers Technical

More information

Accelerate Insights with Topology, High Throughput and Power Advancements

Accelerate Insights with Topology, High Throughput and Power Advancements Accelerate Insights with Topology, High Throughput and Power Advancements Michael A. Jackson, President Wil Wellington, EMEA Professional Services May 2014 1 Adaptive/Cray Example Joint Customers Cray

More information

Splunking IT Data Is Great, Splunking Non- IT Data Is Awesome

Splunking IT Data Is Great, Splunking Non- IT Data Is Awesome Copyright 2015 Splunk Inc. Splunking IT Data Is Great, Splunking Non- IT Data Is Awesome Mathew Benwell Informa?on Security Specialist, The University of Adelaide Disclaimer During the course of this presenta?on,

More information

Contract Management COE Abbo1 Approach

Contract Management COE Abbo1 Approach Contract Management COE Abbo1 Approach September 20, 2017 Kedric Chamberlin Director, Legal Opera7ons Abbo: Kedric.Chamberlin@abbo:.com 1 Topics CMS @ Abbo: Centers of Excellence Key Concepts Abbo: COE

More information

An IBM Proof of Technology IBM Workload Deployer Overview

An IBM Proof of Technology IBM Workload Deployer Overview An IBM Proof of Technology IBM Workload Deployer Overview WebSphere Infrastructure: The Big Picture Vertically integrated and horizontally fit for purpose Operational Management & Efficiency IBM Workload

More information

Deep Learning Acceleration with MATRIX: A Technical White Paper

Deep Learning Acceleration with MATRIX: A Technical White Paper The Promise of AI The Challenges of the AI Lifecycle and Infrastructure MATRIX Solution Architecture Bitfusion Core Container Management Smart Resourcing Data Volumes Multitenancy Interactive Workspaces

More information

Law Department Strategic Planning. Moving from Vision to Execu;on

Law Department Strategic Planning. Moving from Vision to Execu;on Law Department Strategic Planning Moving from Vision to Execu;on 1 Welcome and Panel Introduc;ons Aaron Van Nice Chris6ne Juhasz Nancy Jessen Nikki Rahimzadeh Director, Legal Opera;ons Legal Opera;ons

More information

NVIDIA QUADRO VIRTUAL DATA CENTER WORKSTATION APPLICATION SIZING GUIDE FOR SIEMENS NX APPLICATION GUIDE. Ver 1.0

NVIDIA QUADRO VIRTUAL DATA CENTER WORKSTATION APPLICATION SIZING GUIDE FOR SIEMENS NX APPLICATION GUIDE. Ver 1.0 NVIDIA QUADRO VIRTUAL DATA CENTER WORKSTATION APPLICATION SIZING GUIDE FOR SIEMENS NX APPLICATION GUIDE Ver 1.0 EXECUTIVE SUMMARY This document provides insights into how to deploy NVIDIA Quadro Virtual

More information

D5.1 Inter-Layer Cloud Stack Adaptation Summary

D5.1 Inter-Layer Cloud Stack Adaptation Summary D5.1 Inter-Layer Cloud Stack Adaptation Summary The ASCETiC architecture focuses on providing novel methods and tools to support software developers aiming at optimising energy efficiency resulting from

More information

Cisco Workload Optimization Manager: Setup and Use Cases

Cisco Workload Optimization Manager: Setup and Use Cases Cisco Workload Optimization Manager: Setup and Use Cases 2017 Cisco and/or its affiliates. All rights reserved. This document is Cisco Public Information. Page 1 of 49 Contents Introduction Minimum requirements

More information

Engineering Cloud Applications 1

Engineering Cloud Applications 1 Engineering Cloud Applications 1 Cloud Computing Dr. Philipp Leitner University of Zurich, Switzerland software evolution & architecture lab What is Cloud Computing? (1) From a user perspective: Software

More information

Accenture* Integrates a Platform Telemetry Solution for OpenStack*

Accenture* Integrates a Platform Telemetry Solution for OpenStack* white paper Communications Service Providers Service Assurance Accenture* Integrates a Platform Telemetry Solution for OpenStack* Using open source software and Intel Xeon processor-based servers, Accenture

More information

Realize Your Product Promise

Realize Your Product Promise Realize Your Product Promise ANSYS Enterprise Cloud is a complete simulation platform, delivered in your secure, dedicated environment on the public cloud. Complete and extensible, ANSYS Enterprise Cloud

More information

AN ENERGY EFFICIENT SCHEME FOR CLOUD RESOURCE PROVISIONING USING TASK SCHEDULING STRATEGY BASED ON VACATION QUEUING THEORY

AN ENERGY EFFICIENT SCHEME FOR CLOUD RESOURCE PROVISIONING USING TASK SCHEDULING STRATEGY BASED ON VACATION QUEUING THEORY AN ENERGY EFFICIENT SCHEME FOR CLOUD RESOURCE PROVISIONING USING TASK SCHEDULING STRATEGY BASED ON VACATION QUEUING THEORY M.Sarojinidevi Department of Computer Science and Engineering K.Ramakrishnan College

More information

Virtualization Manager 7.1 Comprehensive virtualization management for VMware vsphere and Microsoft Hyper-V

Virtualization Manager 7.1 Comprehensive virtualization management for VMware vsphere and Microsoft Hyper-V DATASHEET Virtualization Manager 7.1 Comprehensive virtualization management for VMware vsphere and Microsoft Hyper-V Download a free product trial and start monitoring your network typically in about

More information

Oracle on Google Cloud Platform: Pitfalls, Real Options & Best Practices

Oracle on Google Cloud Platform: Pitfalls, Real Options & Best Practices Oracle on Google Cloud Platform: Pitfalls, Real Options & Best Practices February 22, 2018 Copyright House of Brick Technologies, LLC This work is the intellectual property of House of Brick Technologies,

More information

Microsoft FastTrack For Azure Service Level Description

Microsoft FastTrack For Azure Service Level Description ef Microsoft FastTrack For Azure Service Level Description 2017 Microsoft. All rights reserved. 1 Contents Microsoft FastTrack for Azure... 3 Eligible Solutions... 3 FastTrack for Azure Process Overview...

More information

Table of Contents HOL CMP

Table of Contents HOL CMP Table of Contents Lab Overview - HOL-1834-CMP - vrealize Suite Lifecycle Manager... 2 Lab Guidance... 3 Module 1 - Introduction to vrealize Suite Lifecycle Manager (30 minutes)...10 Introduction... 11

More information

ORACLE DATABASE PERFORMANCE: VMWARE CLOUD ON AWS PERFORMANCE STUDY JULY 2018

ORACLE DATABASE PERFORMANCE: VMWARE CLOUD ON AWS PERFORMANCE STUDY JULY 2018 ORACLE DATABASE PERFORMANCE: VMWARE CLOUD ON AWS PERFORMANCE STUDY JULY 2018 Table of Contents Executive Summary...3 Introduction...3 Test Environment... 4 Test Workload... 6 Virtual Machine Configuration...

More information

Chapter 3: Automated self-service and catalogue service. Topics covered: 3.1 Datacenter architecture. 3.2 Iaas and its rental cost

Chapter 3: Automated self-service and catalogue service. Topics covered: 3.1 Datacenter architecture. 3.2 Iaas and its rental cost 1 Chapter 3: Automated self-service and catalogue service Topics covered: 3.1 Datacenter architecture 3.2 Iaas and its rental cost 3.3 Cloud web access architecture 3.4 Service catalogue 3.5 Change instances

More information

IDN Variant TLD Program 18 July 2013

IDN Variant TLD Program 18 July 2013 IDN Variant TLD Program 18 July 2013 Introduc

More information

Automated Service Builder

Automated Service Builder 1 Overview ASB is a platform and application agnostic solution for implementing complex processing chains over globally distributed processing and data ASB provides a low coding solution to develop a data

More information

enlight CLOUD CLOUD for On PREMISE Computing.

enlight CLOUD CLOUD for On PREMISE Computing. enlight CLOUD CLOUD for On PREMISE Computing. Private Cloud Solution Target Segment NIC National Information Center. Government and PSU. Enterprises with on premise datacenter. Enterprises with collocated

More information

Experimental Analysis on Autonomic Strategies for Cloud Elasticity T HOMAS L E DOUX

Experimental Analysis on Autonomic Strategies for Cloud Elasticity T HOMAS L E DOUX 2015 International Conference on Cloud Computing and Autonomic Computing Experimental Analysis on Autonomic Strategies for Cloud Elasticity SIMON DUPONT, J O NAT HAN LEJEUNE, FREDERICO ALVARES T HOMAS

More information

World Leading Storage Cloud at ETH Zürich

World Leading Storage Cloud at ETH Zürich Felix Sutter Dr. Tilo Steiger IT Architect, IBM Switzerland Ltd Head of Storage Services, ETH Zürich Informatikdienste World Leading Storage Cloud at ETH Zürich Agenda The challenges From IT Silos to Cloud

More information

IBM United States Hardware Announcement , dated September 2, 2014

IBM United States Hardware Announcement , dated September 2, 2014 IBM United States Hardware Announcement 114-125, dated September 2, 2014 IBM PureFlex System Table of contents 1 Overview 3 Product number 2 Key prerequisites 4 Publications 2 Planned availability date

More information

Redefining Perspectives A thought leadership forum for technologists interested in defining a new future

Redefining Perspectives A thought leadership forum for technologists interested in defining a new future Redefining Perspectives A thought leadership forum for technologists interested in defining a new future Session 1 The Past, Present and Future of Cloud Computing in Capital and Commodity Markets Dixit

More information

Cloud Computing mit mathematischen Anwendungen

Cloud Computing mit mathematischen Anwendungen Cloud Computing mit mathematischen Anwendungen Dr. habil. Marcel Kunze Engineering Mathematics and Computing Lab (EMCL) Institut für Angewandte und Numerische Mathematik IV Karlsruhe Institute of Technology

More information

inteliscaler Workload and Resource Aware, Proactive Auto Scaler for PaaS Cloud

inteliscaler Workload and Resource Aware, Proactive Auto Scaler for PaaS Cloud inteliscaler Workload and Resource Aware, Proactive Auto Scaler for PaaS Cloud Paper #10368 RS Shariffdeen, UKJU Bandara, DTSP Munasinghe, HS Bhathiya, and HMN Dilum Bandara Dept. of Computer Science &

More information

Managed Cloud storage. Turning to Storage as a Service for flexibility

Managed Cloud storage. Turning to Storage as a Service for flexibility Managed Cloud storage Turning to Storage as a Service for flexibility Table of contents Encountering problems? 2 Get an answer 2 Check out cloud services 2 Getting started 3 Understand changing costs 4

More information

MS Integrating On-Premises Core Infrastructure with Microsoft Azure

MS Integrating On-Premises Core Infrastructure with Microsoft Azure MS-10992 Integrating On-Premises Core Infrastructure with Microsoft Azure COURSE OVERVIEW: This three-day instructor-led course covers a range of components, including Azure Compute, Azure Storage, and

More information

BARCELONA. 2015, Amazon Web Services, Inc. or its affiliates. All rights reserved

BARCELONA. 2015, Amazon Web Services, Inc. or its affiliates. All rights reserved BARCELONA 2015, Amazon Web Services, Inc. or its affiliates. All rights reserved Optimizing Cost and Efficiency on AWS Inigo Soto Practice Manager, AWS Professional Services 2015, Amazon Web Services,

More information

UAB Expanding the Campus Compu7ng Cloud

UAB Expanding the Campus Compu7ng Cloud Condor @ UAB Expanding the Campus Compu7ng Cloud Condor Week 2012 John- Paul Robinson jpr@uab.edu Poornima Pochana ppreddy@uab.edu Thomas Anthony tanthony@uab.edu Public University with 17,571 Student

More information

Example. You manage a web site, that suddenly becomes wildly popular. Performance starts to degrade. Do you?

Example. You manage a web site, that suddenly becomes wildly popular. Performance starts to degrade. Do you? Scheduling Main Points Scheduling policy: what to do next, when there are mul:ple threads ready to run Or mul:ple packets to send, or web requests to serve, or Defini:ons response :me, throughput, predictability

More information

Principles of Information Systems

Principles of Information Systems Principles of Information Systems Session 08 Systems Investigation and Analysis An Overview of Systems Development Today, users of informa0on systems are involved in their development Avoid costly failures

More information

AVANTUS TRAINING PTE LTD

AVANTUS TRAINING PTE LTD [MS10979]: Microsoft Azure Fundamentals Length : 2 Days Audience(s) : IT Professionals Level : 100 Technology : Azure Delivery Method : Instructor-led (Classroom) Course Overview This course provides the

More information

TECHNICAL WHITE PAPER. Rubrik and Microsoft Azure Technology Overview and How It Works

TECHNICAL WHITE PAPER. Rubrik and Microsoft Azure Technology Overview and How It Works TECHNICAL WHITE PAPER Rubrik and Microsoft Azure Technology Overview and How It Works TABLE OF CONTENTS THE UNSTOPPABLE RISE OF CLOUD SERVICES...3 CLOUD PARADIGM INTRODUCES DIFFERENT PRINCIPLES...3 WHAT

More information

1. DevOps 2. Application Hosting Options 3. Automating Environment Setup 4. Deployment Scripting 5. Application Monitoring 6. Continuous Deployment

1. DevOps 2. Application Hosting Options 3. Automating Environment Setup 4. Deployment Scripting 5. Application Monitoring 6. Continuous Deployment 1. DevOps 2. Application Hosting Options 3. Automating Environment Setup 4. Deployment Scripting 5. Application Monitoring 6. Continuous Deployment and Scrum Dev A Development Operations Development Working

More information

Customer Challenges SOLUTION BENEFITS

Customer Challenges SOLUTION BENEFITS SOLUTION BRIEF Matilda Cloud Solutions simplify migration of your applications to a public or private cloud, then monitor and control the environment for ongoing IT operations. Our solution empowers businesses

More information

The Hybrid PMO. How to increase you por5olio s transparency to meet the needs of an Agile enterprise

The Hybrid PMO. How to increase you por5olio s transparency to meet the needs of an Agile enterprise The Hybrid PMO How to increase you por5olio s transparency to meet the needs of an Agile enterprise Presenter: Alan Shefveland, Director - Product Strategy, Changepoint A recent Changepoint survey said

More information

Cloud Capacity Management

Cloud Capacity Management Cloud Capacity Management Defining Cloud Computing Cloud computing is a type of Internet based computing that provides shared computer processing resources and data to computers and other devices on demand.

More information

Oracle Analy+cs Data Visualiza+on and Discovery. November 10, 2016

Oracle Analy+cs Data Visualiza+on and Discovery. November 10, 2016 Oracle Analy+cs Data Visualiza+on and Discovery November 10, 2016 Data Visualiza+on - Abstract! For years the ability to analyze data to discover new insights through impressive visualiza@ons and rapid

More information

Cloud Cruiser for Cisco Intelligent Automation for Cloud

Cloud Cruiser for Cisco Intelligent Automation for Cloud Cloud Cruiser for Cisco Intelligent Automation for Cloud Cloud Financial Management 1 How Do Cloud Cruiser and Cisco Intelligent Automation Work Together? Cisco IAC provisions the service A tenant user

More information

Cost Management in the AWS Cloud. March 2018

Cost Management in the AWS Cloud. March 2018 Cost Management in the AWS Cloud March 2018 2018, Amazon Web Services, Inc. or its affiliates. All rights reserved. Notices This document is provided for informational purposes only. It represents AWS

More information

Commvault XaaS Solutions for Service Providers

Commvault XaaS Solutions for Service Providers Commvault XaaS Solutions for Service Providers DELIVERING AN EASILY SCALABLE, FEATURE RICH MULTI-TENNANT SOLUTION KEY BENEFITS Lower Infrastructure Costs Commvault software s singular platform shares heterogeneous

More information

Fast Innovation requires Fast IT

Fast Innovation requires Fast IT Fast Innovation requires Fast IT Trends and Data Centre Innovation Sudheesh Subhash Principal Solutions Architect Agenda Application trends Current data centre trends IT Cloud integration Automation and

More information