Techniques for Minimizing Cloud Footprint

Arun Kejariwal
Arun KejariwalStatistical Learning Principal at Machine Zone, Inc.
Techniques for Minimizing Cloud Footprint



                         Arun Kejariwal

                             March 2013




1              International Conference on Cloud Engineering 2013   © Arun Kejariwal
Overview

      Cloud Computing
        o  Ubiquitous



        o  IDC – Cloud Services revenue ~100 B by 2016
      Infrastructure-as-a-Service (IaaS)
        o  Large adoption
               Elasticity
                     Scale up and down
               Eliminates cycles for hardware procurement, datacenter maintenance
               Higher product development agility
               Fault Tolerance
                     Geographically distributed availability zones (e.g., AWS in US, Ireland, Singapore, Japan, Brazil)
        o  Growing vendors




2                                                International Conference on Cloud Engineering 2013                        © Arun Kejariwal
Capitalizing Cloud Elasticity

      Non-trivial to manage
        o  Aggressive scale down
              Potentially affect latency and throughput adversely
                    Degraded end-user experience - impacts bottomline

        o  Aggressive scale up
              Balloon cloud footprint
                    Adversely impact operational efficiency - impacts bottomline


        o  Which metric to use?
              Traffic
                    Incoming, Outgoing




              CPU
              Load


3                                              International Conference on Cloud Engineering 2013   © Arun Kejariwal
Capitalizing Cloud Elasticity (contd.)

      Non-trivial to manage
        o  Scaling policy
               Whether to scale up/down by fixed amount
               Whether to scale up/down by percentage of current capacity
               What should be the cooldown period?

      Efficient exploitation of cloud elasticity: Why Bother?
        o  Handle high traffic
               Deliver best end-user experience
        o  Contain overall footprint
               Reserved vs. On-Demand instances




4                                       International Conference on Cloud Engineering 2013   © Arun Kejariwal
Key Contributions

      Algorithms
       1.  Scale up/down by fixed amount
               Target applications: small (< 20-25 nodes) clusters
               Case Study
       2.  Scale up/down by percentage of current capacity
               Target applications: medium/large clusters
       3.  Implications of long application start up time?
               Example application: Netflix Recommendation Engine
                     Loading of user metadata (such as users preferences)
               Modified Algorithm 2 to capture effects of application start up time
               Case Study




5                                          International Conference on Cloud Engineering 2013   © Arun Kejariwal
Autoscaling in AWS: Quick review

      CloudWatch
        o  Real-time monitoring of EC2 instances
        o  Metrics
               CPU utilization, latency, network traffic, application specific etc.
        o  Alarm
               Change state when metric > threshold
               Action: when metric > threshold for a number of time periods

      Policy
        o  ScalingAdjustment
               # instances by which to scale
        o  Adjustment type
               ChangeInCapacity
               PercentChangeInCapacity

      Cooldown
        o  Allows effects of scaling activity to become visible

6                                          International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: Design Guidelines

      Avoid ping-pong effect




        o  High latency
        o  SLA violation
      Be proactive, not reactive
        o  Application start up

                                                                                       SLA driven need
                                                                                         to scale up

                                                                                       Autoscaling
                                                                                          event




7                                 International Conference on Cloud Engineering 2013         © Arun Kejariwal
Techniques: Design Guidelines (contd.)

      Aggressive upwards, conservative downwards
        o  Deliver best user experience
              Handle more than expected increase in traffic
              Handle slower ramp down of traffic
        o  Aggressive scale up
              Provides buffer for increase in traffic during the cooldown period
        o  Aggressive scale down
              May result in under-provisioning
                    Impact user experience

      Scalability Analysis
        o  Given an SLA, determine maximum throughput




8                                             International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: Properties

      RPS -> Requests per second

    1.  RPS/node after scale up > Scale down threshold

    2.  RPS/node after scale down > Scale up threshold




9                            International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: # 1

       Scale up by fixed amount
        o  Outline
                    : Scale up threshold (RPS per node)
                    : Scale up value
             
                   Proactive
                   Empirically determined




       Example applications
        o  Beaconserver, customer service
        o  It does add up!

10                                           International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: # 1 (contd.)

       Case Study
        Aggressive upwards
      Conservative downwards




11                                   International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: # 3

       Scale up/down by percentage of current capacity
        o  Account for application start up
        o  How to model it?
               Time series of RPS
                     : Application start up time, determine rolling RPS change time series




               Compute 99th percentile
               Compute effective scale up threshold
                     Consistent with being proactive design guideline




12                                              International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: # 3 (contd.)


      Rolling RPS Change Time Series
       (Application start up: 30 mins)




13                  International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: # 3 (contd.)

       Scale up percentage of current capacity
           Outline
                     : Scale up threshold (RPS per node)
                     : Scale up value
              
                    Proactive
                    Empirically determined




       Example applications
         o  Merchweb, simsservice, recommendation service, API
         o  Savings up to 50%


14                                            International Conference on Cloud Engineering 2013   © Arun Kejariwal
Techniques: # 3 (contd.)

       Case Study




15                        International Conference on Cloud Engineering 2013   © Arun Kejariwal
Wrapping up …

       Summary
        o  Improve operational efficiency in the cloud
             o  Benefits both small and large clusters
             o  Up to 50% reduction in operational footprint


       Future work
        o  How to handle spikes?




        o  Capture interaction between different services in a SOA
        o  Autoscale across IaaS vendors


16                                       International Conference on Cloud Engineering 2013   © Arun Kejariwal
Q&A




17   International Conference on Cloud Engineering 2013   © Arun Kejariwal
1 of 17

Recommended

A Systematic Approach to Capacity Planning in the Real World by
A Systematic Approach to Capacity Planning in the Real WorldA Systematic Approach to Capacity Planning in the Real World
A Systematic Approach to Capacity Planning in the Real WorldArun Kejariwal
5.5K views23 slides
Clear sailing for allinea ddt at 700,000+ mpi tasks on blue waters by
Clear sailing for allinea ddt at 700,000+ mpi tasks on blue watersClear sailing for allinea ddt at 700,000+ mpi tasks on blue waters
Clear sailing for allinea ddt at 700,000+ mpi tasks on blue watersAkash Sharma
189 views2 slides
Distributed Trace & Log Analysis using ML by
Distributed Trace & Log Analysis using MLDistributed Trace & Log Analysis using ML
Distributed Trace & Log Analysis using MLJorge Cardoso
438 views25 slides
Hadoop technology by
Hadoop technologyHadoop technology
Hadoop technologyDhanashriDeokar1
16 views14 slides
Pivotal Greenplum in Action on AWS, Azure, and GCP - Greenplum Summit 2018 by
Pivotal Greenplum in Action on AWS, Azure, and GCP - Greenplum Summit 2018Pivotal Greenplum in Action on AWS, Azure, and GCP - Greenplum Summit 2018
Pivotal Greenplum in Action on AWS, Azure, and GCP - Greenplum Summit 2018VMware Tanzu
888 views23 slides
OW2con'16 Keynote address: Kubernetes, the rising tide of systems administrat... by
OW2con'16 Keynote address: Kubernetes, the rising tide of systems administrat...OW2con'16 Keynote address: Kubernetes, the rising tide of systems administrat...
OW2con'16 Keynote address: Kubernetes, the rising tide of systems administrat...OW2
171 views65 slides

More Related Content

What's hot

App Performance Tip: Sharing Flash Across Virtualized Workloads by
App Performance Tip: Sharing Flash Across Virtualized WorkloadsApp Performance Tip: Sharing Flash Across Virtualized Workloads
App Performance Tip: Sharing Flash Across Virtualized WorkloadsDataCore Software
590 views33 slides
Cloud Reliability: Decreasing outage frequency using fault injection by
Cloud Reliability: Decreasing outage frequency using fault injectionCloud Reliability: Decreasing outage frequency using fault injection
Cloud Reliability: Decreasing outage frequency using fault injectionJorge Cardoso
595 views29 slides
Going Server-less for Web-Services that need to Crunch Large Volumes of Data by
Going Server-less for Web-Services that need to Crunch Large Volumes of DataGoing Server-less for Web-Services that need to Crunch Large Volumes of Data
Going Server-less for Web-Services that need to Crunch Large Volumes of DataDenis C. Bauer
176 views29 slides
Machine Learning Model Deployment: Strategy to Implementation by
Machine Learning Model Deployment: Strategy to ImplementationMachine Learning Model Deployment: Strategy to Implementation
Machine Learning Model Deployment: Strategy to ImplementationDataWorks Summit
3.3K views39 slides
HPC in higher education by
HPC in higher educationHPC in higher education
HPC in higher educationKishor Satpathy
900 views54 slides
CUDA Sessions You Won't Want to Miss at GTC 2019 by
CUDA Sessions You Won't Want to Miss at GTC 2019CUDA Sessions You Won't Want to Miss at GTC 2019
CUDA Sessions You Won't Want to Miss at GTC 2019NVIDIA
16.9K views14 slides

What's hot(8)

App Performance Tip: Sharing Flash Across Virtualized Workloads by DataCore Software
App Performance Tip: Sharing Flash Across Virtualized WorkloadsApp Performance Tip: Sharing Flash Across Virtualized Workloads
App Performance Tip: Sharing Flash Across Virtualized Workloads
DataCore Software590 views
Cloud Reliability: Decreasing outage frequency using fault injection by Jorge Cardoso
Cloud Reliability: Decreasing outage frequency using fault injectionCloud Reliability: Decreasing outage frequency using fault injection
Cloud Reliability: Decreasing outage frequency using fault injection
Jorge Cardoso595 views
Going Server-less for Web-Services that need to Crunch Large Volumes of Data by Denis C. Bauer
Going Server-less for Web-Services that need to Crunch Large Volumes of DataGoing Server-less for Web-Services that need to Crunch Large Volumes of Data
Going Server-less for Web-Services that need to Crunch Large Volumes of Data
Denis C. Bauer176 views
Machine Learning Model Deployment: Strategy to Implementation by DataWorks Summit
Machine Learning Model Deployment: Strategy to ImplementationMachine Learning Model Deployment: Strategy to Implementation
Machine Learning Model Deployment: Strategy to Implementation
DataWorks Summit3.3K views
CUDA Sessions You Won't Want to Miss at GTC 2019 by NVIDIA
CUDA Sessions You Won't Want to Miss at GTC 2019CUDA Sessions You Won't Want to Miss at GTC 2019
CUDA Sessions You Won't Want to Miss at GTC 2019
NVIDIA16.9K views
How novel compute technology transforms life science research by Denis C. Bauer
How novel compute technology transforms life science researchHow novel compute technology transforms life science research
How novel compute technology transforms life science research
Denis C. Bauer350 views

Viewers also liked

Integracio continguts sabadell_web by
Integracio continguts sabadell_webIntegracio continguts sabadell_web
Integracio continguts sabadell_webCESIRE - Dept d'Educació - GENCAT
3.7K views14 slides
Days In Green : Forecasting the Life of a Healthy Service @Twitter by
Days In Green : Forecasting the Life of a Healthy Service @TwitterDays In Green : Forecasting the Life of a Healthy Service @Twitter
Days In Green : Forecasting the Life of a Healthy Service @TwitterVibhav Garg
2K views32 slides
思薇爾Swear型錄 by
思薇爾Swear型錄思薇爾Swear型錄
思薇爾Swear型錄julia chuang
881 views20 slides
Ob型錄 by
Ob型錄Ob型錄
Ob型錄julia chuang
361 views28 slides
Matematiques donar respostes by
Matematiques donar respostesMatematiques donar respostes
Matematiques donar respostesCESIRE - Dept d'Educació - GENCAT
1K views65 slides
A Tool for Practical Garbage Collection Analysis In the Cloud by
A Tool for Practical Garbage Collection Analysis In the CloudA Tool for Practical Garbage Collection Analysis In the Cloud
A Tool for Practical Garbage Collection Analysis In the CloudArun Kejariwal
3.4K views15 slides

Viewers also liked(19)

Days In Green : Forecasting the Life of a Healthy Service @Twitter by Vibhav Garg
Days In Green : Forecasting the Life of a Healthy Service @TwitterDays In Green : Forecasting the Life of a Healthy Service @Twitter
Days In Green : Forecasting the Life of a Healthy Service @Twitter
Vibhav Garg2K views
思薇爾Swear型錄 by julia chuang
思薇爾Swear型錄思薇爾Swear型錄
思薇爾Swear型錄
julia chuang881 views
A Tool for Practical Garbage Collection Analysis In the Cloud by Arun Kejariwal
A Tool for Practical Garbage Collection Analysis In the CloudA Tool for Practical Garbage Collection Analysis In the Cloud
A Tool for Practical Garbage Collection Analysis In the Cloud
Arun Kejariwal3.4K views
Lasten ja nuorten osallisuus lapsivaikutusten arvioinnissa by THL
Lasten ja nuorten osallisuus lapsivaikutusten arvioinnissaLasten ja nuorten osallisuus lapsivaikutusten arvioinnissa
Lasten ja nuorten osallisuus lapsivaikutusten arvioinnissa
THL288 views
Frederick Grant Banting, descubridor de la insulina, historieta completa Novaro by Martin Alberto Belaustegui
Frederick Grant Banting, descubridor de la insulina, historieta completa NovaroFrederick Grant Banting, descubridor de la insulina, historieta completa Novaro
Frederick Grant Banting, descubridor de la insulina, historieta completa Novaro
электронное портфолио by LLANDERA80
электронное портфолиоэлектронное портфолио
электронное портфолио
LLANDERA80143 views
Kuule nuoria - Sokran Kemin kiertoajelun aineisto 14.3.2017 by THL
Kuule nuoria - Sokran Kemin kiertoajelun aineisto 14.3.2017Kuule nuoria - Sokran Kemin kiertoajelun aineisto 14.3.2017
Kuule nuoria - Sokran Kemin kiertoajelun aineisto 14.3.2017
THL1.3K views
Amalan terbaik dalam pembangunan sosial by Aini Rahman
Amalan terbaik dalam pembangunan sosialAmalan terbaik dalam pembangunan sosial
Amalan terbaik dalam pembangunan sosial
Aini Rahman680 views
Lmcp 1552 pembangunan mapan dalam islam by NUR SYAFAWANI
Lmcp 1552 pembangunan mapan dalam islamLmcp 1552 pembangunan mapan dalam islam
Lmcp 1552 pembangunan mapan dalam islam
NUR SYAFAWANI646 views
Gosaikund tour bsc 3rd gg 2014 student pics by amulya123
Gosaikund tour bsc 3rd gg 2014 student picsGosaikund tour bsc 3rd gg 2014 student pics
Gosaikund tour bsc 3rd gg 2014 student pics
amulya123192 views
мо вихователів та класних керівників І ступеню by Надежда Сорока
мо вихователів та класних керівників І ступенюмо вихователів та класних керівників І ступеню
мо вихователів та класних керівників І ступеню
Loadแนวข้อสอบ พนักงานสมทบตรวจเงินแผ่นดิน ชั้น 3 (ด้านบัญชี) สำนักงานการตรวจเ... by nawaporn khamseanwong
 Loadแนวข้อสอบ พนักงานสมทบตรวจเงินแผ่นดิน ชั้น 3 (ด้านบัญชี) สำนักงานการตรวจเ... Loadแนวข้อสอบ พนักงานสมทบตรวจเงินแผ่นดิน ชั้น 3 (ด้านบัญชี) สำนักงานการตรวจเ...
Loadแนวข้อสอบ พนักงานสมทบตรวจเงินแผ่นดิน ชั้น 3 (ด้านบัญชี) สำนักงานการตรวจเ...

Similar to Techniques for Minimizing Cloud Footprint

SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM... by
SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM...SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM...
SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM...ijgca
281 views8 slides
Multi-level Elasticity Control of Cloud Services -- ICSOC 2013 by
Multi-level Elasticity Control of Cloud Services -- ICSOC 2013Multi-level Elasticity Control of Cloud Services -- ICSOC 2013
Multi-level Elasticity Control of Cloud Services -- ICSOC 2013Georgiana Copil
2.3K views19 slides
Deploy Microservices in the Real World by
Deploy Microservices in the Real WorldDeploy Microservices in the Real World
Deploy Microservices in the Real WorldElana Krasner
520 views31 slides
Hands-On Lab: Monitor Modern Applications in the Cloud by
Hands-On Lab: Monitor Modern Applications in the CloudHands-On Lab: Monitor Modern Applications in the Cloud
Hands-On Lab: Monitor Modern Applications in the CloudCA Technologies
182 views16 slides
CloudSpurt customer by
CloudSpurt customerCloudSpurt customer
CloudSpurt customerCloudSpurt Systems Private Limited
200 views15 slides
internship paper by
internship paperinternship paper
internship paperBandhana Harlalka
199 views9 slides

Similar to Techniques for Minimizing Cloud Footprint(20)

SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM... by ijgca
SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM...SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM...
SERVICE LEVEL AGREEMENT BASED FAULT TOLERANT WORKLOAD SCHEDULING IN CLOUD COM...
ijgca281 views
Multi-level Elasticity Control of Cloud Services -- ICSOC 2013 by Georgiana Copil
Multi-level Elasticity Control of Cloud Services -- ICSOC 2013Multi-level Elasticity Control of Cloud Services -- ICSOC 2013
Multi-level Elasticity Control of Cloud Services -- ICSOC 2013
Georgiana Copil2.3K views
Deploy Microservices in the Real World by Elana Krasner
Deploy Microservices in the Real WorldDeploy Microservices in the Real World
Deploy Microservices in the Real World
Elana Krasner520 views
Hands-On Lab: Monitor Modern Applications in the Cloud by CA Technologies
Hands-On Lab: Monitor Modern Applications in the CloudHands-On Lab: Monitor Modern Applications in the Cloud
Hands-On Lab: Monitor Modern Applications in the Cloud
CA Technologies182 views
A survey paper on an improved scheduling algorithm for task offloading on cloud by Aditya Tornekar
A survey paper on an improved scheduling algorithm for task offloading on cloudA survey paper on an improved scheduling algorithm for task offloading on cloud
A survey paper on an improved scheduling algorithm for task offloading on cloud
Aditya Tornekar56 views
Evaluating paas scalability and improving performance using scalability impro... by eSAT Journals
Evaluating paas scalability and improving performance using scalability impro...Evaluating paas scalability and improving performance using scalability impro...
Evaluating paas scalability and improving performance using scalability impro...
eSAT Journals38 views
Evaluating paas scalability and improving performance using scalability impro... by eSAT Publishing House
Evaluating paas scalability and improving performance using scalability impro...Evaluating paas scalability and improving performance using scalability impro...
Evaluating paas scalability and improving performance using scalability impro...
The Enterprise Adoption of Cloud Technology - Infographic by RapidValue by RapidValue
The Enterprise Adoption of Cloud Technology - Infographic by RapidValueThe Enterprise Adoption of Cloud Technology - Infographic by RapidValue
The Enterprise Adoption of Cloud Technology - Infographic by RapidValue
RapidValue93 views
Application Architecture for Cloud Computing by white paper
Application Architecture for Cloud Computing Application Architecture for Cloud Computing
Application Architecture for Cloud Computing
white paper757 views
Cloud Testing : An Overview by QA InfoTech
Cloud Testing : An OverviewCloud Testing : An Overview
Cloud Testing : An Overview
QA InfoTech216 views
Cloud computing (pdf) by Steven Habuda
Cloud computing   (pdf)Cloud computing   (pdf)
Cloud computing (pdf)
Steven Habuda29.1K views
What does performance mean in the cloud by Michael Kopp
What does performance mean in the cloudWhat does performance mean in the cloud
What does performance mean in the cloud
Michael Kopp772 views
Overview of Sensors project by Shan Guan
Overview of Sensors projectOverview of Sensors project
Overview of Sensors project
Shan Guan395 views
Best Data Center Service Provider in India - Best Hybrid Cloud Hosting Servi... by NetData Vault
Best Data Center Service Provider in India -  Best Hybrid Cloud Hosting Servi...Best Data Center Service Provider in India -  Best Hybrid Cloud Hosting Servi...
Best Data Center Service Provider in India - Best Hybrid Cloud Hosting Servi...
NetData Vault92 views
A New Approach to Continuous Monitoring in the Cloud by NETSCOUT
A New Approach to Continuous Monitoring in the CloudA New Approach to Continuous Monitoring in the Cloud
A New Approach to Continuous Monitoring in the Cloud
NETSCOUT1.3K views
WeLab Reaps Advantages of Multi-Cloud Capabilities. You Can Too. by NuoDB
WeLab Reaps Advantages of Multi-Cloud Capabilities. You Can Too.WeLab Reaps Advantages of Multi-Cloud Capabilities. You Can Too.
WeLab Reaps Advantages of Multi-Cloud Capabilities. You Can Too.
NuoDB296 views
G-SLAM:OPTIMIZING ENERGY EFFIIENCY IN CLOUD by Alfiya Mahmood
G-SLAM:OPTIMIZING ENERGY EFFIIENCY IN CLOUDG-SLAM:OPTIMIZING ENERGY EFFIIENCY IN CLOUD
G-SLAM:OPTIMIZING ENERGY EFFIIENCY IN CLOUD
Alfiya Mahmood73 views

More from Arun Kejariwal

Anomaly Detection At The Edge by
Anomaly Detection At The EdgeAnomaly Detection At The Edge
Anomaly Detection At The EdgeArun Kejariwal
581 views54 slides
Serverless Streaming Architectures and Algorithms for the Enterprise by
Serverless Streaming Architectures and Algorithms for the EnterpriseServerless Streaming Architectures and Algorithms for the Enterprise
Serverless Streaming Architectures and Algorithms for the EnterpriseArun Kejariwal
2.8K views227 slides
Sequence-to-Sequence Modeling for Time Series by
Sequence-to-Sequence Modeling for Time SeriesSequence-to-Sequence Modeling for Time Series
Sequence-to-Sequence Modeling for Time SeriesArun Kejariwal
3.2K views64 slides
Sequence-to-Sequence Modeling for Time Series by
Sequence-to-Sequence Modeling for Time SeriesSequence-to-Sequence Modeling for Time Series
Sequence-to-Sequence Modeling for Time SeriesArun Kejariwal
1.9K views45 slides
Model Serving via Pulsar Functions by
Model Serving via Pulsar FunctionsModel Serving via Pulsar Functions
Model Serving via Pulsar FunctionsArun Kejariwal
1.7K views44 slides
Designing Modern Streaming Data Applications by
Designing Modern Streaming Data ApplicationsDesigning Modern Streaming Data Applications
Designing Modern Streaming Data ApplicationsArun Kejariwal
2.6K views227 slides

More from Arun Kejariwal(20)

Anomaly Detection At The Edge by Arun Kejariwal
Anomaly Detection At The EdgeAnomaly Detection At The Edge
Anomaly Detection At The Edge
Arun Kejariwal581 views
Serverless Streaming Architectures and Algorithms for the Enterprise by Arun Kejariwal
Serverless Streaming Architectures and Algorithms for the EnterpriseServerless Streaming Architectures and Algorithms for the Enterprise
Serverless Streaming Architectures and Algorithms for the Enterprise
Arun Kejariwal2.8K views
Sequence-to-Sequence Modeling for Time Series by Arun Kejariwal
Sequence-to-Sequence Modeling for Time SeriesSequence-to-Sequence Modeling for Time Series
Sequence-to-Sequence Modeling for Time Series
Arun Kejariwal3.2K views
Sequence-to-Sequence Modeling for Time Series by Arun Kejariwal
Sequence-to-Sequence Modeling for Time SeriesSequence-to-Sequence Modeling for Time Series
Sequence-to-Sequence Modeling for Time Series
Arun Kejariwal1.9K views
Model Serving via Pulsar Functions by Arun Kejariwal
Model Serving via Pulsar FunctionsModel Serving via Pulsar Functions
Model Serving via Pulsar Functions
Arun Kejariwal1.7K views
Designing Modern Streaming Data Applications by Arun Kejariwal
Designing Modern Streaming Data ApplicationsDesigning Modern Streaming Data Applications
Designing Modern Streaming Data Applications
Arun Kejariwal2.6K views
Correlation Analysis on Live Data Streams by Arun Kejariwal
Correlation Analysis on Live Data StreamsCorrelation Analysis on Live Data Streams
Correlation Analysis on Live Data Streams
Arun Kejariwal321 views
Deep Learning for Time Series Data by Arun Kejariwal
Deep Learning for Time Series DataDeep Learning for Time Series Data
Deep Learning for Time Series Data
Arun Kejariwal1.7K views
Correlation Analysis on Live Data Streams by Arun Kejariwal
Correlation Analysis on Live Data StreamsCorrelation Analysis on Live Data Streams
Correlation Analysis on Live Data Streams
Arun Kejariwal2.1K views
Modern real-time streaming architectures by Arun Kejariwal
Modern real-time streaming architecturesModern real-time streaming architectures
Modern real-time streaming architectures
Arun Kejariwal7.2K views
Anomaly detection in real-time data streams using Heron by Arun Kejariwal
Anomaly detection in real-time data streams using HeronAnomaly detection in real-time data streams using Heron
Anomaly detection in real-time data streams using Heron
Arun Kejariwal4.7K views
Data Data Everywhere: Not An Insight to Take Action Upon by Arun Kejariwal
Data Data Everywhere: Not An Insight to Take Action UponData Data Everywhere: Not An Insight to Take Action Upon
Data Data Everywhere: Not An Insight to Take Action Upon
Arun Kejariwal1.5K views
Real Time Analytics: Algorithms and Systems by Arun Kejariwal
Real Time Analytics: Algorithms and SystemsReal Time Analytics: Algorithms and Systems
Real Time Analytics: Algorithms and Systems
Arun Kejariwal23K views
Finding bad apples early: Minimizing performance impact by Arun Kejariwal
Finding bad apples early: Minimizing performance impactFinding bad apples early: Minimizing performance impact
Finding bad apples early: Minimizing performance impact
Arun Kejariwal1.1K views
Statistical Learning Based Anomaly Detection @ Twitter by Arun Kejariwal
Statistical Learning Based Anomaly Detection @ TwitterStatistical Learning Based Anomaly Detection @ Twitter
Statistical Learning Based Anomaly Detection @ Twitter
Arun Kejariwal5.1K views
Days In Green (DIG): Forecasting the life of a healthy service by Arun Kejariwal
Days In Green (DIG): Forecasting the life of a healthy serviceDays In Green (DIG): Forecasting the life of a healthy service
Days In Green (DIG): Forecasting the life of a healthy service
Arun Kejariwal793 views
Gimme More! Supporting User Growth in a Performant and Efficient Fashion by Arun Kejariwal
Gimme More! Supporting User Growth in a Performant and Efficient FashionGimme More! Supporting User Growth in a Performant and Efficient Fashion
Gimme More! Supporting User Growth in a Performant and Efficient Fashion
Arun Kejariwal2.3K views
Isolating Events from the Fail Whale by Arun Kejariwal
Isolating Events from the Fail WhaleIsolating Events from the Fail Whale
Isolating Events from the Fail Whale
Arun Kejariwal2K views

Recently uploaded

CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue by
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueCloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueShapeBlue
137 views13 slides
Updates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBIT by
Updates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBITUpdates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBIT
Updates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBITShapeBlue
208 views8 slides
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ... by
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...ShapeBlue
171 views28 slides
Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha... by
Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha...Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha...
Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha...ShapeBlue
183 views18 slides
What’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlue by
What’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlueWhat’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlue
What’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlueShapeBlue
265 views23 slides
Business Analyst Series 2023 - Week 4 Session 7 by
Business Analyst Series 2023 -  Week 4 Session 7Business Analyst Series 2023 -  Week 4 Session 7
Business Analyst Series 2023 - Week 4 Session 7DianaGray10
146 views31 slides

Recently uploaded(20)

CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue by ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlueCloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
CloudStack Managed User Data and Demo - Harikrishna Patnala - ShapeBlue
ShapeBlue137 views
Updates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBIT by ShapeBlue
Updates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBITUpdates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBIT
Updates on the LINSTOR Driver for CloudStack - Rene Peinthor - LINBIT
ShapeBlue208 views
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ... by ShapeBlue
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...
How to Re-use Old Hardware with CloudStack. Saving Money and the Environment ...
ShapeBlue171 views
Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha... by ShapeBlue
Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha...Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha...
Mitigating Common CloudStack Instance Deployment Failures - Jithin Raju - Sha...
ShapeBlue183 views
What’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlue by ShapeBlue
What’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlueWhat’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlue
What’s New in CloudStack 4.19 - Abhishek Kumar - ShapeBlue
ShapeBlue265 views
Business Analyst Series 2023 - Week 4 Session 7 by DianaGray10
Business Analyst Series 2023 -  Week 4 Session 7Business Analyst Series 2023 -  Week 4 Session 7
Business Analyst Series 2023 - Week 4 Session 7
DianaGray10146 views
The Power of Generative AI in Accelerating No Code Adoption.pdf by Saeed Al Dhaheri
The Power of Generative AI in Accelerating No Code Adoption.pdfThe Power of Generative AI in Accelerating No Code Adoption.pdf
The Power of Generative AI in Accelerating No Code Adoption.pdf
Saeed Al Dhaheri39 views
Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ... by ShapeBlue
Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ...Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ...
Import Export Virtual Machine for KVM Hypervisor - Ayush Pandey - University ...
ShapeBlue120 views
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ... by ShapeBlue
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
Live Demo Showcase: Unveiling Dell PowerFlex’s IaaS Capabilities with Apache ...
ShapeBlue129 views
Webinar : Desperately Seeking Transformation - Part 2: Insights from leading... by The Digital Insurer
Webinar : Desperately Seeking Transformation - Part 2:  Insights from leading...Webinar : Desperately Seeking Transformation - Part 2:  Insights from leading...
Webinar : Desperately Seeking Transformation - Part 2: Insights from leading...
The Role of Patterns in the Era of Large Language Models by Yunyao Li
The Role of Patterns in the Era of Large Language ModelsThe Role of Patterns in the Era of Large Language Models
The Role of Patterns in the Era of Large Language Models
Yunyao Li91 views
Initiating and Advancing Your Strategic GIS Governance Strategy by Safe Software
Initiating and Advancing Your Strategic GIS Governance StrategyInitiating and Advancing Your Strategic GIS Governance Strategy
Initiating and Advancing Your Strategic GIS Governance Strategy
Safe Software184 views
Redefining the book supply chain: A glimpse into the future - Tech Forum 2023 by BookNet Canada
Redefining the book supply chain: A glimpse into the future - Tech Forum 2023Redefining the book supply chain: A glimpse into the future - Tech Forum 2023
Redefining the book supply chain: A glimpse into the future - Tech Forum 2023
BookNet Canada44 views
KVM Security Groups Under the Hood - Wido den Hollander - Your.Online by ShapeBlue
KVM Security Groups Under the Hood - Wido den Hollander - Your.OnlineKVM Security Groups Under the Hood - Wido den Hollander - Your.Online
KVM Security Groups Under the Hood - Wido den Hollander - Your.Online
ShapeBlue225 views
"Surviving highload with Node.js", Andrii Shumada by Fwdays
"Surviving highload with Node.js", Andrii Shumada "Surviving highload with Node.js", Andrii Shumada
"Surviving highload with Node.js", Andrii Shumada
Fwdays58 views
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue by ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlueElevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
Elevating Privacy and Security in CloudStack - Boris Stoyanov - ShapeBlue
ShapeBlue224 views
NTGapps NTG LowCode Platform by Mustafa Kuğu
NTGapps NTG LowCode Platform NTGapps NTG LowCode Platform
NTGapps NTG LowCode Platform
Mustafa Kuğu437 views
Zero to Cloud Hero: Crafting a Private Cloud from Scratch with XCP-ng, Xen Or... by ShapeBlue
Zero to Cloud Hero: Crafting a Private Cloud from Scratch with XCP-ng, Xen Or...Zero to Cloud Hero: Crafting a Private Cloud from Scratch with XCP-ng, Xen Or...
Zero to Cloud Hero: Crafting a Private Cloud from Scratch with XCP-ng, Xen Or...
ShapeBlue199 views

Techniques for Minimizing Cloud Footprint

  • 1. Techniques for Minimizing Cloud Footprint Arun Kejariwal March 2013 1 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 2. Overview   Cloud Computing o  Ubiquitous o  IDC – Cloud Services revenue ~100 B by 2016   Infrastructure-as-a-Service (IaaS) o  Large adoption   Elasticity   Scale up and down   Eliminates cycles for hardware procurement, datacenter maintenance   Higher product development agility   Fault Tolerance   Geographically distributed availability zones (e.g., AWS in US, Ireland, Singapore, Japan, Brazil) o  Growing vendors 2 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 3. Capitalizing Cloud Elasticity   Non-trivial to manage o  Aggressive scale down   Potentially affect latency and throughput adversely   Degraded end-user experience - impacts bottomline o  Aggressive scale up   Balloon cloud footprint   Adversely impact operational efficiency - impacts bottomline o  Which metric to use?   Traffic   Incoming, Outgoing   CPU   Load 3 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 4. Capitalizing Cloud Elasticity (contd.)   Non-trivial to manage o  Scaling policy   Whether to scale up/down by fixed amount   Whether to scale up/down by percentage of current capacity   What should be the cooldown period?   Efficient exploitation of cloud elasticity: Why Bother? o  Handle high traffic   Deliver best end-user experience o  Contain overall footprint   Reserved vs. On-Demand instances 4 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 5. Key Contributions   Algorithms 1.  Scale up/down by fixed amount   Target applications: small (< 20-25 nodes) clusters   Case Study 2.  Scale up/down by percentage of current capacity   Target applications: medium/large clusters 3.  Implications of long application start up time?   Example application: Netflix Recommendation Engine   Loading of user metadata (such as users preferences)   Modified Algorithm 2 to capture effects of application start up time   Case Study 5 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 6. Autoscaling in AWS: Quick review   CloudWatch o  Real-time monitoring of EC2 instances o  Metrics   CPU utilization, latency, network traffic, application specific etc. o  Alarm   Change state when metric > threshold   Action: when metric > threshold for a number of time periods   Policy o  ScalingAdjustment   # instances by which to scale o  Adjustment type   ChangeInCapacity   PercentChangeInCapacity   Cooldown o  Allows effects of scaling activity to become visible 6 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 7. Techniques: Design Guidelines   Avoid ping-pong effect o  High latency o  SLA violation   Be proactive, not reactive o  Application start up SLA driven need to scale up Autoscaling event 7 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 8. Techniques: Design Guidelines (contd.)   Aggressive upwards, conservative downwards o  Deliver best user experience   Handle more than expected increase in traffic   Handle slower ramp down of traffic o  Aggressive scale up   Provides buffer for increase in traffic during the cooldown period o  Aggressive scale down   May result in under-provisioning   Impact user experience   Scalability Analysis o  Given an SLA, determine maximum throughput 8 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 9. Techniques: Properties   RPS -> Requests per second 1.  RPS/node after scale up > Scale down threshold 2.  RPS/node after scale down > Scale up threshold 9 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 10. Techniques: # 1   Scale up by fixed amount o  Outline   : Scale up threshold (RPS per node)   : Scale up value     Proactive   Empirically determined   Example applications o  Beaconserver, customer service o  It does add up! 10 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 11. Techniques: # 1 (contd.)   Case Study Aggressive upwards Conservative downwards 11 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 12. Techniques: # 3   Scale up/down by percentage of current capacity o  Account for application start up o  How to model it?   Time series of RPS   : Application start up time, determine rolling RPS change time series   Compute 99th percentile   Compute effective scale up threshold   Consistent with being proactive design guideline 12 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 13. Techniques: # 3 (contd.) Rolling RPS Change Time Series (Application start up: 30 mins) 13 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 14. Techniques: # 3 (contd.)   Scale up percentage of current capacity   Outline   : Scale up threshold (RPS per node)   : Scale up value     Proactive   Empirically determined   Example applications o  Merchweb, simsservice, recommendation service, API o  Savings up to 50% 14 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 15. Techniques: # 3 (contd.)   Case Study 15 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 16. Wrapping up …   Summary o  Improve operational efficiency in the cloud o  Benefits both small and large clusters o  Up to 50% reduction in operational footprint   Future work o  How to handle spikes? o  Capture interaction between different services in a SOA o  Autoscale across IaaS vendors 16 International Conference on Cloud Engineering 2013 © Arun Kejariwal
  • 17. Q&A 17 International Conference on Cloud Engineering 2013 © Arun Kejariwal