Introduction to Cloud Computing

(1)

January 25, 2014

1

Our team: UndergradTim Hegeman, Stefan Hugtenburg, Jesse Donkevliet …

GradSiqi Shen, Guo Yong, Nezih Yigitbasi StaffHenk Sips, Dick Epema, Alexandru Iosup, Otto Visser CollaboratorsIon Stoica and the Mesos team (UC Berkeley), Thomas Fahringer, Radu Prodan, Vlad Nae (U. Innsbruck), Nicolae Tapus, Mihaela Balint, Vlad Posea, Alexandru Olteanu (UPB), Derrick Kondo (INRIA), Assaf Schuster, Mark Silberstein, Orna Ben-Yehuda (Technion), Kefeng Deng (NUDT, CN), ...

Introduction to

Cloud Computing

Alexandru Iosup

Parallel and Distributed Systems Group Delft University of Technology The Netherlands

SPEC RG Cloud Meeting

January 25, 2014

2

January 25, 2014

3

What is Cloud Computing?

3. A Useful IT Service

“Use only when you want! Pay only for what you use!”

January 25, 2014

4

Q:What do youuse?

Q:Why not thislevel? Q:Why not thislevel?

January 25, 2014

5

Agenda

1. What is Cloud Computing?

2. IaaS Clouds, the Core Idea

3. The IaaS Owner Perspective

4. The IaaS User Perspective

5. Reality Check

6. Conclusion

IaaS Cloud Computing

(2)

Joe Has an Idea ($$$)

(Source: A. Antoniou, MSc Defense, TU Delft, 2012. Original idea: A. Iosup, 2011.)

MusicWave

• Big up-front commitment

• Load variability: NOT supported

Solution #1

Buy or Rent

…

10%

Solution #2

Deploy on IaaS Cloud

(Source: A. Antoniou, MSc Defense, TU Delft, 2012. Original idea: V. Nae, 2008.)

Q:So are we just shifting the problem to somebody else, that is, the IaaS cloud owner?

• NO

big up-front commitment

• Load variability: supported

Inside an IaaS Cloud Data Center

Time and Cost Sharing Among Users

User C

User B MusicWave

(Source: A. Antoniou, MSc Defense, TU Delft, 2012.)

Main Characteristics of IaaS Clouds

1. On-Demand Pay-per-Use

2. Elasticity (cloud concept of Scalability)

3. Resource Pooling

4. Fully automated IT services

5. Quality of Service

January 25, 2014

(3)

January 25, 2014

13

Agenda

1. What is Cloud Computing?

2. IaaS Clouds, the Core Idea

3. The IaaS Owner Perspective:

How to Deploy a Cloud?

4. The IaaS User Perspective

5. Reality Check

6. Conclusion

IaaS Cloud Deployment Models

Private

On-premises

Public

Off-premises

Hybrid

(Source: A. Antoniou, MSc Defense, TU Delft, 2012. Original idea: Mell and Grance, NIST Spec.Pub. 800-145, Sep 2011.)

Resource Sharing Models

MusicWave January 25, 2014 15 MusicWave OtherApp

Space-Sharing

Time-Sharing

IaaS Clouds

MusicWave OtherApp

Q:Which one is better?

Grids

Host OS Host OS OtherApp

Virtualization

January 25, 2014 16 Virtualization Host OS

MusicWave OtherApp OtherApp

Q:What to do now? Guest OS Virtual Resources VM Instance Applications Guest OS Virtual Resources VM Instance Applications

Q:What is the problem?

January 25, 2014

Virtualization and The Full IaaS Stack

17 Guest OS Virtual Resources VM Instance Applications Physical Infrastructure Virtual Infrastructure Manager

Virtual Machine Manager

Guest OS Virtual Resources

VM Instance Applications

Virtual Machine Manager

Guest OS Virtual Resources

VM Instance Applications

The Virtual Machine Lifecycle

January 25, 2014

18 (Source: A. Antoniou, MSc Defense, TU Delft, 2012.)

(4)

Use Case:

Amazon Elastic Compute Cloud (EC2)

• Prominent IaaS provider

• Datacenters all over the world

• Many VM instance types

• Per-hour

charging

January 25, 2014

19

Instance Capacity US$/hour

m1.small 0.10 m1.large 0.38 c1.xlarge 0.76 January 25, 2014 20

Agenda

1. What is Cloud Computing?

2. IaaS Clouds, the Core Idea

3. The IaaS Owner Perspective

4. The IaaS User Perspective:

How to Use Clouds? How to Choose Clouds?

5. Reality Check

6. Conclusion

Workload

January 25, 2014 21 MusicWave OtherApp Time MusicWave OtherApp OtherApp OtherApp Load = 4 RunTime= 6

Use Case: Workloads of Zynga

(Massively Social Gaming)

January 25, 2014

22 Sources: CNN, Zynga.

Source: InsideSocialGames.com

“Zynga made more than $600M in 2010 from selling in-game virtual goods.” S. Greengard, CACM, Apr 2011 Selling in-game virtual

goods: “Zynga made est. $270M in 2009 from.”

http://techcrunch.com/2010/ 05/03/zynga-revenue/

Use Case: Workloads of Zynga

(Massively Social Gaming)

• Load can grow very quickly

January 25, 2014 23 L o a d

Provisioning and Allocation of Resources

January 25, 2014 24 L o a d Time

Provisioning

Allocation

(5)

Provisioning and Allocation of Resources

January 25, 2014 25 L o a d Time

Provisioning

Allocation

Q:What is the interplaybetween provisioning and allocation?

Provisioning and Allocation

Policies

January 25, 2014 26

Where?

When?

How many?

Time L o a d

Provisioning

Allocation

From where?

Which type?

etc.

When?

etc.

Q:How manypolicies exist? Q: How to selecta policy?

(Source: A. Antoniou, MSc Defense, TU Delft, 2012.)

Use Case:

Two Provisioning Policies, Compared

January 25, 2014

27

Startup

OnDemand

Villegas, Antoniou, Sadjadi, Iosup. An Analysis of Provisioning and Allocation Policies for Infrastructure-as-a-Service Clouds, (submitted). PDS Tech.Rep.2011-009

Use Case:

Two Provisioning Policies, Compared

Metrics for comparison

• Job Slowdown (

JSD

): Ratio of actual runtime in the

cloud and the runtime in a dedicated non-virtualized

environment

• Charged Cost (

C

c

)

• Utility (

U

)

January 25, 2014

28

Q:Charged cost vs Total RunTime?

Use Case:

Two Provisioning Policies, Compared

Workloads

January 25, 2014 29 0.0 0.5 1.0 1.5 2.0 0 20 40 60 CPU Load (%) Time (min) 0.0 0.5 1.0 1.5 2.0 0 20 40 60 Time (min) 0.0 0.5 1.0 1.5 2.0 0 20 40 60 Time (min)

Uniform Increasing Bursty

System Hardware VIM Hypervisor Max VMs

DAS4/Delft 20 Dual quad-core 2.4 GHz 24 GB RAM 2x1 TB storage 64 FIU 7 Pentium 4 3.0 GHz 5 GB RAM 340 GB Storage 7

Amazon EC2 unkown/various - 20

Use Case:

Two Provisioning Policies, Compared

Environments

January 25, 2014

30

(6)

Use Case:

Many Provisioning Policies, Compared

Job Slowdown (JSD)

January 25, 2014 31 0 5 10 15 20 25 30 35 40 Job Slowdown DAS4 0 5 10 15 20 25 30 Job Slowdown FIU 0 5 10

Job Slowdown Workload EC2 Startup OnDemand ExecTime ExecAvg ExecKN QueueWait Q:Why is OnDemand worsethan Startup?

A:waiting for machines to boot

Use Case:

Many Provisioning Policies, Compared

Charged Cost (C

c

)

January 25, 2014 32 0 50 100 150 200 250 300 Hours DAS4 0 5 10 15 20 25 30 Hours FIU 0 5 10 15 20 25 30 35 40 45

Hours Workload EC2 Startup OnDemand ExecTime ExecAvg ExecKN QueueWait Q:Why is OnDemand worsethan Startup?

A:VM thrashing

Q:Why no OnDemand on Amazon EC2?

Use Case:

Many Provisioning Policies, Compared

Utility (U

)

33 0 1 2 3 4 5 6 7 8 Utility DAS4 0 1 2 3 4 5 Utility FIU 0 1 2 3 4 5

Utility

Workload EC2

Startup

OnDemand ExecTimeExecAvg QueueWaitExecKN January 25, 2014

34

Agenda

1. What is Cloud Computing?

2. IaaS Clouds, the Core Idea

3. The IaaS Owner Perspective

4. The IaaS User Perspective

5. Reality Check:

Who Uses Public Commercial Clouds?

6. Conclusion

January 25, 2014

35 35

The Real IaaS Cloud

• “The path to abundance”

• On-demand capacity

• Cheap for short-term tasks

• Great for web apps (EIP, web

crawl, DB ops, I/O)

• “The killer cyclone”

• Not so great performance

for scientific applications

(compute- or data-intensive)

http://www.flickr.com/photos/dimitrisotiropoulos/4204766418/ Tropical Cyclone Nargis (NASA, ISSS, 04/29/08)

VS

January 25, 2014 January 25, 2014

36 (Source: http://www.cca08.org/files/slides/w_vogel.pdf)

(7)

Zynga zCloud: Hybrid Self-Hosted/EC2

• After Zynga had large scale

• More efficient self-hosted servers

• Run at high utilization

• Use EC2 for unexpected demand

January 25, 2014

37 (Sources: http://seekingalpha.com/article/609141-how-amazon-s-aws-can-attract-ugly-economics and http://www.undertheradarblog.com/blog/3-reasons-zynga-is-moving-to-a-private-cloud/)

Other Cloud Customers

• 218 virtual CPUs

• 9TB/2TB block/S3 storage

• 6.5TB/2TB I/O per month

January 25, 2014 38 (Source: http://markbuhagiar.com/technical/businessinthecloud/) January 25, 2014 39

Agenda

1. What is Cloud Computing?

2. IaaS Clouds, the Core Idea

3. The IaaS Owner Perspective

4. The IaaS User Perspective

5. Reality Check

6. Conclusion

January 25, 2014

40

Conclusion Take-Home Message

• Cloud Computing = IaaS + PaaS + SaaS

• Core idea = lease vs self-own

• On-Demand, Pay-per-Use, Elastic, Pooled, Automated, QoS

• The Owner Perspective

• Time-Sharing

• Virtualization

• The User Perspective

• Variable workloads

• Provisioning and Allocation policies

• Reality Check: 100s of users

http://www.flickr.com/photos/dimitrisotiropoulos/4204766418/

January 25, 2014

41

Thank you for your attention!

Questions? Suggestions? Observations?

Alexandru Iosup

A.Iosup@tudelft.nl

http://www.pds.ewi.tudelft.nl/~iosup/

(or google “iosup”)

Parallel and Distributed Systems Group

Delft University of Technology

-http://www.st.ewi.tudelft.nl/~iosup/research.html -http://www.st.ewi.tudelft.nl/~iosup/research_cloud.html -http://www.pds.ewi.tudelft.nl/

More Info:

Do not hesitate to contact me…

… And Now For Something Different

• Ongoing projects in Cloud Computing at TUD

January 25, 2014

(8)

Massivizing Social Games: Distributed Computing

Challenges and High Quality Time – A. Iosup 43

Cloudification: PaaS for MSGs

(Platform Challenge) Build Social Gaming

platform that uses (mostly) cloud resources

• Close to players

• No upfront costs, no maintenance

• Compute platforms: multi-cores, GPUs, clusters, all-in-one!

• Performance guarantees

• Hybrid deployment model

• Code for various compute platforms—platform profiling

• Load prediction miscalculation costs real money

• What are the services?

• Vendor lock-in?

• My

data

Data, Data, Data

Understanding the Behavior of

Players in Online Social Games

January 25, 2014 44 460 73 66 596 871 198 1075 1004 32 127 51 240 (w/ E.Dias, A. Olteanu, F. Kuipers)

Iosup et al. An Analysis of Implicit Social Networks in Multiplayer Online Games.IEEE Internet Computing

45

• Using data centers for

dynamic

resource allocation

• Main advantages:

1. Significantly lower over-provisioning

2. Efficient coverage of the world is possible

Proposed hosting model: dynamic

Massive join

Massive join Massive leave

Nae, Iosup, Prodan. Dynamic Resource Provisioning in Massively Multiplayer Online Games. IEEE TPDS 2011

46

Resource Provisioning and Allocation

Static

vs.

Dynamic

Provisioning

250%

25%

[Source: Nae, Iosup, and Prodan, ACM SC 2008]

O v e r-p ro v is io n in g = W A S T E (% )

Nae, Iosup, Prodan. Dynamic Resource Provisioning in Massively Multiplayer Online Games. IEEE TPDS 2011

• Old scheduling aspects

• Workloads evolve over time and exhibit periods of distinct characteristics

• No one-size-fits-all policy: hundreds exist, each good for specific conditions

• Data centers increasingly popular (also not new)

• Constant deployment since mid-1990s

• Users moving their computation to IaaS-cloud data centers

• Consolidation efforts in mid- and large-scale companies

• New scheduling aspects

• New workloads

• New data center architectures

• New cost models

• Developing a scheduling policy is risky and ephemeral • Selecting a scheduling policy for your data center is difficult • Combining the strengths of multiple scheduling policies is …

Why Portfolio Scheduling?

What is Portfolio Scheduling?

In a Nutshell, for Data Centers

• Create a set of scheduling policies

• Resource provisioning and allocation policies, in this work

• Online selection of the active policy, at important moments

• Periodic selection, in this work

• Same principle for other changes: pricing model, system, …

Deng, Song, Ren, Iosup: Exploring portfolio scheduling for long-term execution of scientific workloads in IaaS clouds. SC 2013: 55

(9)

Performance Evaluation

1) Effect of Portfolio Scheduling (1)

• Portfolio scheduling is 8%, 11%, 45%, and 30%better than the best constituent policy

A portfolio scheduler can be better than

any of its constituent policies

Q:

What can prevent a portfolio scheduler from being

better than any of its constituent policies?

Deng, Song, Ren, Iosup: Exploring portfolio scheduling for long-term execution of scientific workloads in IaaS clouds. SC 2013: 55

Performance Evaluation

1) Effect of Portfolio Scheduling (2)

• Job selection policies such as UNICEF and LXF that favor

short jobs have the best performance

_{(provisioning, job selection, VM selection) policy}

Q:

How well do you think a single

would perform? Will it be dominant? (Rhetorical)

Massivizing Social Games: Distributed Computing

Challenges and High Quality Time – A. Iosup 51

Mobile Social Gaming and the SuperServer

(Platform Challenge) Support

Social Gaming on mobiles

• Mobiles everywhere (2bn+ users)

• Gaming industry for mobiles is new

Growing Market

• SuperServer to generate content for

low-capability devices?

• Battery for 3D/Networked games?

• Where is my server? (Ad-hoc mobile

gaming networks?)

• Security, cheat-prevention

Extending the Capabilities of Mobile

Devices through Cloud Offloading …

with Application to Online Social Games

Metrics:

• processing time

• packet size

• inter-arrival rate

Design & Implementation:

• based on OpenTTD

• repeatability

• offloading mechanisms

• instrumentation for metrics

53

Social Everything! So Analytics

• Social Network=undirected graph, relationship=edge

• Community=sub-graph, density of edges between its nodes higher

than density of edges outside sub-graph

(Analytics Challenge)

Improve gaming experience

• Ranking / Rating

• Matchmaking / Recommendations

• Play Style/Tutoring

Self-Organizing Gaming Communities

• Player Behavior

Asterix B-tree

How to Analyze? Ecosystems of

Big-Data Programming Models

Dremel Service Tree

SQL JAQL Hive Pig

MapReduce Model Algebrix

PACT MPI/ Erlang L F S

Nephele Hadoop/ Dryad Hyracks

YARN Haloop DryadLINQ Scope Pregel HDFS AQL CosmosFS Azure Engine Tera Data Engine

Adapted from: Dagstuhl Seminar on Information Management in the Cloud,

http://www.dagstuhl.de/program/calendar/partlist/?semnr=11321&SUOG Azure Data Store Tera Data Store Storage Engine Execution Engine Voldemort High-Level Language Programming Model GFS BigQuery Flume Flume Engine S3 Dataflow Giraph Sawzall Meteor

(10)

January 25, 2014

55

Benchmarking suite

Platforms and Process

• Platforms

• Process

• Evaluate baseline (out of the box) and tuned performance

• Evaluate performance on fixed-size system

• Future: evaluate performance on elastic-size system

• Evaluate scalability

YARN

Giraph

Guo, Biczak, Varbanescu, Iosup, Martella, Willke. How Well do Graph-Processing Platforms Perform? An Empirical Performance Evaluation and Analysis

January 25, 2014

56

BFS: results for all platforms, all data sets

• No platform can runs fastest of every graph

• Not all platforms can process all graphs

• Hadoop is the worst performer

January 25, 2014

57

Giraph: results for

all algorithms, all data sets

• Storing the whole graph in memory helps Giraph perform well

• Giraph may crash when

graphs

or messages

become larger

January 25, 2014

58

Conclusion and ongoing work

• Performance is f(Data set, Algorithm, Platform, Deployment)

• Cannot tell yet which of (Data set, Algorithm, Platform) the

most important (also depends on Platform)

• Platforms have their own drawbacks

• Some platforms can scale up reasonably with cluster size

(horizontally) or number of cores (vertically)

• Ongoing work

• Benchmarking

suite

• Build a performance boundary model

• Explore performance variability

Guo, Biczak, Varbanescu, Iosup, Martella, Willke. How Well do Graph-Processing Platforms Perform? An Empirical Performance Evaluation and Analysis.IPDPS

http://bit.ly/10hYdIU

DotA communities

• Players are loosely organised in communities

•Operate game servers

•Maintain lists of tournaments and results

•Publish statistics and rankings on websites

•Dota-League: players join a queue and matchmaking forms teams •DotAlicious: players can choose which match/team to join R. van de Bovenkamp, S. Shen, A. Iosup, F. A. Kuipers:

Understanding and recommending play relationships in online social gaming. COMSNETS 2013: 1-10