SlideShare uma empresa Scribd logo
1 de 30
Baixar para ler offline
The Secret Guide to
Cloud Performance
Jan 30 2014 @ DevCon TLV
Gideon Greenspan
Co-Founder, Cloudlook
Site/DB

Site/DB

Site/DB

OS etc. Site/DB

OS + web
server + DBMS

OS etc. Site/DB

Website +
Database

OS etc. Site/DB

Types of hosting

OS + web
server + DBMS

Hardware

Hypervisor
Host OS
Hardware

Hardware

Dedicated

Cloud / VPS

Shared

Secret Guide to Cloud Performance

Slide 2

Cloudlook
“The Cloud” vs Old-School VPS
•  Deploy/decommission in minutes
•  Scale up/down in minutes
•  Management APIs
•  Automatic scaling
•  Billing by the hour
•  Migration on hardware failure
Secret Guide to Cloud Performance

Slide 3

Cloudlook
Some public cloud providers

Secret Guide to Cloud Performance

Slide 4

Cloudlook
Cloud specifications — AWS EC2

Secret Guide to Cloud Performance

Slide 5

Cloudlook
Cloud specs — Rackspace

Secret Guide to Cloud Performance

Slide 6

Cloudlook
The trouble with the cloud
•  Vague system specifications
•  Variation between instances
–  Noisy neighbors?
–  Diverse underlying hardware

•  Variation over time within an instance
•  One-time benchmarks insufficient
Secret Guide to Cloud Performance

Slide 7

Cloudlook
Cloudlook: continuous benchmarking
•  6 cloud providers, 35 instance types
•  15,000 instances per month
•  50 samples per instance over 1 hour
•  All 64-bit CentOS (even on Azure!)
•  $2,000 per month
•  Available free for your server too!
Secret Guide to Cloud Performance

Slide 8

Cloudlook
What does Cloudlook measure?
•  Integer CPU performance
•  Concurrency
•  Memory bandwidth
•  Sequential disk access

All done with…

•  Random disk access
•  Host disk cache effect
Secret Guide to Cloud Performance

Slide 9

Cloudlook
The Secret Guide
•  Introduction
•  Benchmarks
•  Conclusions

Secret Guide to Cloud Performance

Slide 10

Cloudlook
Measuring integer CPU
•  PHP scripts run as interpreted bytecode
•  Built-in PHP functions are compiled C
•  We benchmarked 14 heavy functions
across 66 versions of PHP ≥ 5.0.4
•  Two performed consistently (±0.2%)
–  sha1() — hash function
–  levenshtein() — deep string comparison
Secret Guide to Cloud Performance

Slide 11

Cloudlook
CPU results
Core i7
equivalent

Between
instances

Within
instances

1

GoGrid (all)

2.05 GHz

± 6%

± 4%

2

HP Cloud (all)

2.01 GHz

± 12%

± 10%

3

Rackspace Performance 1 (all)

2.01 GHz

± 4%

± 4%

4

Rackspace Standard (all)

1.93 GHz

± 6%

± 5%

5

Windows Azure (all)

1.74 GHz

± 3%

± 11%

6

Digital Ocean ≥1GB

1.62 GHz

± 19%

± 12%

7

Amazon m1.medium

1.48 GHz

± 6%

± 5%

Secret Guide to Cloud Performance

Slide 12

Cloudlook
CPU variation per instance
Rackspace Cloud Performance 1 2GB

Digital Ocean 2GB (sometimes)
Secret Guide to Cloud Performance

Slide 13

Cloudlook
Measuring concurrency
•  Parallel CPU-heavy HTTP requests
•  Precise client-side synchronization
–  microtime() + usleep()

•  Avoid CPU temperature bias
Parallel
HTTP
requests

1
2
3
4

Secret Guide to Cloud Performance

Slide 14

Cloudlook
Concurrency results
Measured concurrency

5

4
Windows Azure
Digital Ocean + HP Cloud
Rackspace Performance 1
GoGrid
Amazon EC2
Rackspace Standard

3

2

1
1

2

3

4

5

6

7

8

Specified vCPUs / cores
Secret Guide to Cloud Performance

Slide 15

Cloudlook
Measuring memory bandwidth
•  Create a 64 MB string in memory
–  Largest L1+L2+L3 cache currently ~41 GB

•  substr() to read successive windows
–  memcpy() in C does the actual work

•  Window length = 128 kb
–  Too short ⇒ time in bytecode interpreter
–  Too long ⇒ copy too big for L1+L2 cache
Secret Guide to Cloud Performance

Slide 16

Cloudlook
Memory results
Memory
bandwidth

Between
instances

Within
instances

1

GoGrid

6.7 GB/s

± 10%

± 7%

2

Rackspace Performance 1

6.7 GB/s

± 7%

± 4%

3

Digital Ocean ≥1GB

6.3 GB/s

± 10%

± 8%

4

HP Cloud

5.9 GB/s

± 9%

± 12%

5

Amazon m1.medium

5.2 GB/s

± 25%

± 4%

6

Rackspace Standard

4.1 GB/s

± 16%

± 7%

7

Windows Azure

3.6 GB/s

± 5%

± 6%

Secret Guide to Cloud Performance

Slide 17

Cloudlook
Memory performance variation

HP Cloud Standard Medium

Rackspace Cloud Standard 4GB

Digital Ocean 4GB

Amazon m1.large

Secret Guide to Cloud Performance

Slide 18

Cloudlook
Measuring disk performance
•  Write benchmark files in /tmp
•  Avoid operating system cache
–  Total size > instance memory
–  Explicitly clear cache (our instances)
–  Read each block once (our instances)

•  “Random” is pseudo-random
–  Fibonacci hashing scheme
Secret Guide to Cloud Performance

Slide 19

Cloudlook
OS etc. Site/DB

OS etc. Site/DB

OS etc. Site/DB

Host cache interference

Hypervisor
Host OS
Hardware
Secret Guide to Cloud Performance

•  Results can be skewed
•  Big instances: write lots of
other data inbetween
•  Small instances: wait a
week and pray…
•  But still worth measuring
because it’s a good thing!
Slide 20

Cloudlook
Disk results (no host cache)
10,000

EC2 m1.medium Primary

Disk Seeks — ops/s

EC2 m1.medium Secondary
Rackspace Standard >= 2GB
Rackspace Performance 1 >= 2GB Both

1,000

Azure >= 1.5 GB Primary
Azure >= 1.5 GB Secondary
HP Cloud >= 2GB Both
Digital Ocean >= 1GB

100

GoGrid >= 2 GB

10
10

100

1000

Disk Throughput — MB/s
Secret Guide to Cloud Performance

Slide 21

Cloudlook
Who has a host cache?
Primary disk rate
(no host cache)

Primary disk rate
(host cache)

Host cache
speedup

HP Cloud Small

109 MB/s

528 MB/s

4.8 ⤬

HP Cloud Medium

124 MB/s

320 MB/s

2.6 ⤬

HP Cloud Large

110 MB/s

117 MB/s

1.1 ⤬

Digital Ocean 1 GB

244 MB/s

709 MB/s

2.9 ⤬

Digital Ocean 2 GB

267 MB/s

703 MB/s

2.6 ⤬

Digital Ocean 4 GB

260 MB/s

709 MB/s

2.7 ⤬

Digital Ocean 8 GB

318 MB/s

474 MB/s

1.5 ⤬

Secret Guide to Cloud Performance

Slide 22

Cloudlook
The Secret Guide
•  Introduction
•  Benchmarks
•  Conclusions

Secret Guide to Cloud Performance

Slide 23

Cloudlook
Same instance, different stats
EC2 m1.medium

GoGrid 4 GB

Rackspace
Performance 1
4 GB

Integer CPU

1.4 : 1

1.2 : 1

1.5 : 1

Concurrency

1.1 : 1

1.5 : 1

1.5 : 1

Memory

3.8 : 1

1.6 : 1

1.7 : 1

Disk throughput

3.0 : 1

10.0 : 1

3.0 : 1

Disk seeks

15.8 : 1

4.2 : 1

3.3 : 1

Secret Guide to Cloud Performance

Slide 24

Cloudlook
Dedicated vs Cloud
Hetzner EX-40

GoGrid 2 GB

GoGrid 8 GB

Integer CPU

3.49 GHz

2.04 GHz

2.09 GHz

Concurrency

6.82 ⤬

1.83 ⤬

4.61 ⤬

Memory

12.9 GB/s

6.6 GB/s

6.9 GB/s

Disk throughput

155 MB/s

387 MB/s

381 MB/s

Disk seeks

134 ops/s

127 ops/s

112 ops/s

32 GB

2 GB

8 GB

Disk size

1000 GB

100 GB

400 GB

Price

$67/mo

$66/mo

$263/mo

Memory size

Secret Guide to Cloud Performance

Slide 25

Cloudlook
Bad reasons to host in the cloud
•  Hardware abstraction (peace of mind)
–  Last week Rackspace Cloud lost my data

•  Easy resizing
–  Dedicated is at top of scale anyway
–  Can’t resize Rackspace Performance servers

•  Lower costs
–  No. That’s why hosting companies like it…
Secret Guide to Cloud Performance

Slide 26

Cloudlook
Some good reasons
•  Small instances
•  Fast deploy/decommission
•  Tools and APIs
•  Other cloud services
–  Storage and CDN
–  Databases and messaging

Secret Guide to Cloud Performance

Slide 27

Cloudlook
Image from http://www.kapowwowobjects.com

So: Beware of fashions…

Secret Guide to Cloud Performance

Slide 28

Cloudlook
… and measure your own

Secret Guide to Cloud Performance

Slide 29

Cloudlook
Thank you!

www.cloudlook.com

Secret Guide to Cloud Performance

Slide 30

Cloudlook

Mais conteúdo relacionado

Mais procurados

RedisConf17 - Home Depot - Turbo charging existing applications with Redis
RedisConf17 - Home Depot - Turbo charging existing applications with RedisRedisConf17 - Home Depot - Turbo charging existing applications with Redis
RedisConf17 - Home Depot - Turbo charging existing applications with RedisRedis Labs
 
Persistent Storage for Containerized Applications
Persistent Storage for Containerized ApplicationsPersistent Storage for Containerized Applications
Persistent Storage for Containerized ApplicationsColleen Corrice
 
Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!
Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!
Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!ScyllaDB
 
Stratoscale Latest and Greatest
Stratoscale Latest and GreatestStratoscale Latest and Greatest
Stratoscale Latest and GreatestZach Lanksbury
 
Critical Attributes for a High-Performance, Low-Latency Database
Critical Attributes for a High-Performance, Low-Latency DatabaseCritical Attributes for a High-Performance, Low-Latency Database
Critical Attributes for a High-Performance, Low-Latency DatabaseScyllaDB
 
Back to the future with C++ and Seastar
Back to the future with C++ and SeastarBack to the future with C++ and Seastar
Back to the future with C++ and SeastarTzach Livyatan
 
RedisConf17 - Lyft - Geospatial at Scale - Daniel Hochman
RedisConf17 - Lyft - Geospatial at Scale - Daniel HochmanRedisConf17 - Lyft - Geospatial at Scale - Daniel Hochman
RedisConf17 - Lyft - Geospatial at Scale - Daniel HochmanRedis Labs
 
Introduction to AWS Outposts
Introduction to AWS OutpostsIntroduction to AWS Outposts
Introduction to AWS OutpostsScyllaDB
 
Micro-batching: High-performance writes
Micro-batching: High-performance writesMicro-batching: High-performance writes
Micro-batching: High-performance writesInstaclustr
 
12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES
12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES
12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETESZalando adtech lab
 
Cassandra on Docker @ Walmart Labs
Cassandra on Docker @ Walmart LabsCassandra on Docker @ Walmart Labs
Cassandra on Docker @ Walmart LabsDataStax Academy
 
RedisConf17 - Redis in High Traffic Adtech Stack
RedisConf17 - Redis in High Traffic Adtech StackRedisConf17 - Redis in High Traffic Adtech Stack
RedisConf17 - Redis in High Traffic Adtech StackRedis Labs
 
Introducing Scylla Open Source 4.0
Introducing Scylla Open Source 4.0Introducing Scylla Open Source 4.0
Introducing Scylla Open Source 4.0ScyllaDB
 
AWS to Bare Metal: Motivation, Pitfalls, and Results
AWS to Bare Metal: Motivation, Pitfalls, and ResultsAWS to Bare Metal: Motivation, Pitfalls, and Results
AWS to Bare Metal: Motivation, Pitfalls, and ResultsMongoDB
 
Apache Cassandra Management
Apache Cassandra ManagementApache Cassandra Management
Apache Cassandra ManagementInstaclustr
 
Introduction to Container Storage Interface (CSI)
Introduction to Container Storage Interface (CSI)Introduction to Container Storage Interface (CSI)
Introduction to Container Storage Interface (CSI)Idan Atias
 
Scylla Summit 2019 Keynote - Avi Kivity
Scylla Summit 2019 Keynote - Avi KivityScylla Summit 2019 Keynote - Avi Kivity
Scylla Summit 2019 Keynote - Avi KivityScyllaDB
 
Cloud Price Comparison - AWS vs Azure vs Google
Cloud Price Comparison - AWS vs Azure vs GoogleCloud Price Comparison - AWS vs Azure vs Google
Cloud Price Comparison - AWS vs Azure vs GoogleRightScale
 
Cassandra serving netflix @ scale
Cassandra serving netflix @ scaleCassandra serving netflix @ scale
Cassandra serving netflix @ scaleVinay Kumar Chella
 
Scaling DataStax in Docker
Scaling DataStax in DockerScaling DataStax in Docker
Scaling DataStax in DockerDataStax
 

Mais procurados (20)

RedisConf17 - Home Depot - Turbo charging existing applications with Redis
RedisConf17 - Home Depot - Turbo charging existing applications with RedisRedisConf17 - Home Depot - Turbo charging existing applications with Redis
RedisConf17 - Home Depot - Turbo charging existing applications with Redis
 
Persistent Storage for Containerized Applications
Persistent Storage for Containerized ApplicationsPersistent Storage for Containerized Applications
Persistent Storage for Containerized Applications
 
Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!
Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!
Scylla Summit 2018: Keeping Your Latency SLAs No Matter What!
 
Stratoscale Latest and Greatest
Stratoscale Latest and GreatestStratoscale Latest and Greatest
Stratoscale Latest and Greatest
 
Critical Attributes for a High-Performance, Low-Latency Database
Critical Attributes for a High-Performance, Low-Latency DatabaseCritical Attributes for a High-Performance, Low-Latency Database
Critical Attributes for a High-Performance, Low-Latency Database
 
Back to the future with C++ and Seastar
Back to the future with C++ and SeastarBack to the future with C++ and Seastar
Back to the future with C++ and Seastar
 
RedisConf17 - Lyft - Geospatial at Scale - Daniel Hochman
RedisConf17 - Lyft - Geospatial at Scale - Daniel HochmanRedisConf17 - Lyft - Geospatial at Scale - Daniel Hochman
RedisConf17 - Lyft - Geospatial at Scale - Daniel Hochman
 
Introduction to AWS Outposts
Introduction to AWS OutpostsIntroduction to AWS Outposts
Introduction to AWS Outposts
 
Micro-batching: High-performance writes
Micro-batching: High-performance writesMicro-batching: High-performance writes
Micro-batching: High-performance writes
 
12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES
12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES
12.07.2017 Docker Meetup - POSTGRE SQL ON KUBERNETES
 
Cassandra on Docker @ Walmart Labs
Cassandra on Docker @ Walmart LabsCassandra on Docker @ Walmart Labs
Cassandra on Docker @ Walmart Labs
 
RedisConf17 - Redis in High Traffic Adtech Stack
RedisConf17 - Redis in High Traffic Adtech StackRedisConf17 - Redis in High Traffic Adtech Stack
RedisConf17 - Redis in High Traffic Adtech Stack
 
Introducing Scylla Open Source 4.0
Introducing Scylla Open Source 4.0Introducing Scylla Open Source 4.0
Introducing Scylla Open Source 4.0
 
AWS to Bare Metal: Motivation, Pitfalls, and Results
AWS to Bare Metal: Motivation, Pitfalls, and ResultsAWS to Bare Metal: Motivation, Pitfalls, and Results
AWS to Bare Metal: Motivation, Pitfalls, and Results
 
Apache Cassandra Management
Apache Cassandra ManagementApache Cassandra Management
Apache Cassandra Management
 
Introduction to Container Storage Interface (CSI)
Introduction to Container Storage Interface (CSI)Introduction to Container Storage Interface (CSI)
Introduction to Container Storage Interface (CSI)
 
Scylla Summit 2019 Keynote - Avi Kivity
Scylla Summit 2019 Keynote - Avi KivityScylla Summit 2019 Keynote - Avi Kivity
Scylla Summit 2019 Keynote - Avi Kivity
 
Cloud Price Comparison - AWS vs Azure vs Google
Cloud Price Comparison - AWS vs Azure vs GoogleCloud Price Comparison - AWS vs Azure vs Google
Cloud Price Comparison - AWS vs Azure vs Google
 
Cassandra serving netflix @ scale
Cassandra serving netflix @ scaleCassandra serving netflix @ scale
Cassandra serving netflix @ scale
 
Scaling DataStax in Docker
Scaling DataStax in DockerScaling DataStax in Docker
Scaling DataStax in Docker
 

Destaque

User interface webpage
User interface webpageUser interface webpage
User interface webpagedhaval_tare
 
Localization and Internationalization 2013
Localization and Internationalization 2013Localization and Internationalization 2013
Localization and Internationalization 2013gidgreen
 
User Interface Design
User Interface DesignUser Interface Design
User Interface Designgidgreen
 
Analytics and Optimization 2013
Analytics and Optimization 2013Analytics and Optimization 2013
Analytics and Optimization 2013gidgreen
 
Web API Design 2013
Web API Design 2013Web API Design 2013
Web API Design 2013gidgreen
 
How-To Easily Deploy MongoDB in the Cloud
How-To Easily Deploy MongoDB in the CloudHow-To Easily Deploy MongoDB in the Cloud
How-To Easily Deploy MongoDB in the CloudGoGrid Cloud Hosting
 
Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...
Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...
Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...Max Lynam
 
Why GoGrid Wins Over Amazon Web Services (AWS)
Why GoGrid Wins Over Amazon Web Services (AWS)Why GoGrid Wins Over Amazon Web Services (AWS)
Why GoGrid Wins Over Amazon Web Services (AWS)GoGrid Cloud Hosting
 
IaaS Cloud Benchmarking: Approaches, Challenges, and Experience
IaaS Cloud Benchmarking: Approaches, Challenges, and ExperienceIaaS Cloud Benchmarking: Approaches, Challenges, and Experience
IaaS Cloud Benchmarking: Approaches, Challenges, and ExperienceAlexandru Iosup
 

Destaque (9)

User interface webpage
User interface webpageUser interface webpage
User interface webpage
 
Localization and Internationalization 2013
Localization and Internationalization 2013Localization and Internationalization 2013
Localization and Internationalization 2013
 
User Interface Design
User Interface DesignUser Interface Design
User Interface Design
 
Analytics and Optimization 2013
Analytics and Optimization 2013Analytics and Optimization 2013
Analytics and Optimization 2013
 
Web API Design 2013
Web API Design 2013Web API Design 2013
Web API Design 2013
 
How-To Easily Deploy MongoDB in the Cloud
How-To Easily Deploy MongoDB in the CloudHow-To Easily Deploy MongoDB in the Cloud
How-To Easily Deploy MongoDB in the Cloud
 
Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...
Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...
Secrets to Achieve Your Business Goals with Responsive Web Design & Marketi...
 
Why GoGrid Wins Over Amazon Web Services (AWS)
Why GoGrid Wins Over Amazon Web Services (AWS)Why GoGrid Wins Over Amazon Web Services (AWS)
Why GoGrid Wins Over Amazon Web Services (AWS)
 
IaaS Cloud Benchmarking: Approaches, Challenges, and Experience
IaaS Cloud Benchmarking: Approaches, Challenges, and ExperienceIaaS Cloud Benchmarking: Approaches, Challenges, and Experience
IaaS Cloud Benchmarking: Approaches, Challenges, and Experience
 

Semelhante a The Secret Guide to Cloud Performance - Cloudlook

Retour d'expérience d'un environnement base de données multitenant
Retour d'expérience d'un environnement base de données multitenantRetour d'expérience d'un environnement base de données multitenant
Retour d'expérience d'un environnement base de données multitenantSwiss Data Forum Swiss Data Forum
 
Data Scotland 2019: You can run SQL Server on AWS
Data Scotland 2019: You can run SQL Server on AWSData Scotland 2019: You can run SQL Server on AWS
Data Scotland 2019: You can run SQL Server on AWSJohn McCormack
 
Taking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout SessionTaking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout SessionSplunk
 
MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...
MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...
MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...MongoDB
 
Our Multi-Year Journey to a 10x Faster Confluent Cloud
Our Multi-Year Journey to a 10x Faster Confluent CloudOur Multi-Year Journey to a 10x Faster Confluent Cloud
Our Multi-Year Journey to a 10x Faster Confluent CloudHostedbyConfluent
 
Scaling with sync_replication using Galera and EC2
Scaling with sync_replication using Galera and EC2Scaling with sync_replication using Galera and EC2
Scaling with sync_replication using Galera and EC2Marco Tusa
 
Tuning Linux Windows and Firebird for Heavy Workload
Tuning Linux Windows and Firebird for Heavy WorkloadTuning Linux Windows and Firebird for Heavy Workload
Tuning Linux Windows and Firebird for Heavy WorkloadMarius Adrian Popa
 
Taking Splunk to the Next Level - Architecture
Taking Splunk to the Next Level - ArchitectureTaking Splunk to the Next Level - Architecture
Taking Splunk to the Next Level - ArchitectureSplunk
 
Colvin exadata mistakes_ioug_2014
Colvin exadata mistakes_ioug_2014Colvin exadata mistakes_ioug_2014
Colvin exadata mistakes_ioug_2014marvin herrera
 
Percona Live 2014 - Scaling MySQL in AWS
Percona Live 2014 - Scaling MySQL in AWSPercona Live 2014 - Scaling MySQL in AWS
Percona Live 2014 - Scaling MySQL in AWSPythian
 
PGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar Ahmed
PGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar AhmedPGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar Ahmed
PGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar AhmedEqunix Business Solutions
 
Taking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout SessionTaking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout SessionSplunk
 
The state of Hive and Spark in the Cloud (July 2017)
The state of Hive and Spark in the Cloud (July 2017)The state of Hive and Spark in the Cloud (July 2017)
The state of Hive and Spark in the Cloud (July 2017)Nicolas Poggi
 
Scaling AEM (CQ5) Gem Session
Scaling AEM (CQ5) Gem SessionScaling AEM (CQ5) Gem Session
Scaling AEM (CQ5) Gem SessionMichael Marth
 
Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...
Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...
Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...RightScale
 
DX12 & Vulkan: Dawn of a New Generation of Graphics APIs
DX12 & Vulkan: Dawn of a New Generation of Graphics APIsDX12 & Vulkan: Dawn of a New Generation of Graphics APIs
DX12 & Vulkan: Dawn of a New Generation of Graphics APIsAMD Developer Central
 
DrupalCampLA 2014 - Drupal backend performance and scalability
DrupalCampLA 2014 - Drupal backend performance and scalabilityDrupalCampLA 2014 - Drupal backend performance and scalability
DrupalCampLA 2014 - Drupal backend performance and scalabilitycherryhillco
 

Semelhante a The Secret Guide to Cloud Performance - Cloudlook (20)

Retour d'expérience d'un environnement base de données multitenant
Retour d'expérience d'un environnement base de données multitenantRetour d'expérience d'un environnement base de données multitenant
Retour d'expérience d'un environnement base de données multitenant
 
Shootout at the PAAS Corral
Shootout at the PAAS CorralShootout at the PAAS Corral
Shootout at the PAAS Corral
 
Data Scotland 2019: You can run SQL Server on AWS
Data Scotland 2019: You can run SQL Server on AWSData Scotland 2019: You can run SQL Server on AWS
Data Scotland 2019: You can run SQL Server on AWS
 
Taking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout SessionTaking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout Session
 
MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...
MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...
MongoDB World 2018: Solving Your Backup Needs Using MongoDB Ops Manager, Clou...
 
Our Multi-Year Journey to a 10x Faster Confluent Cloud
Our Multi-Year Journey to a 10x Faster Confluent CloudOur Multi-Year Journey to a 10x Faster Confluent Cloud
Our Multi-Year Journey to a 10x Faster Confluent Cloud
 
Scaling with sync_replication using Galera and EC2
Scaling with sync_replication using Galera and EC2Scaling with sync_replication using Galera and EC2
Scaling with sync_replication using Galera and EC2
 
Tuning Linux Windows and Firebird for Heavy Workload
Tuning Linux Windows and Firebird for Heavy WorkloadTuning Linux Windows and Firebird for Heavy Workload
Tuning Linux Windows and Firebird for Heavy Workload
 
Taking Splunk to the Next Level - Architecture
Taking Splunk to the Next Level - ArchitectureTaking Splunk to the Next Level - Architecture
Taking Splunk to the Next Level - Architecture
 
Colvin exadata mistakes_ioug_2014
Colvin exadata mistakes_ioug_2014Colvin exadata mistakes_ioug_2014
Colvin exadata mistakes_ioug_2014
 
Percona Live 2014 - Scaling MySQL in AWS
Percona Live 2014 - Scaling MySQL in AWSPercona Live 2014 - Scaling MySQL in AWS
Percona Live 2014 - Scaling MySQL in AWS
 
Drupal performance
Drupal performanceDrupal performance
Drupal performance
 
PGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar Ahmed
PGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar AhmedPGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar Ahmed
PGConf.ASIA 2019 Bali - Tune Your LInux Box, Not Just PostgreSQL - Ibrar Ahmed
 
Taking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout SessionTaking Splunk to the Next Level - Architecture Breakout Session
Taking Splunk to the Next Level - Architecture Breakout Session
 
The state of Hive and Spark in the Cloud (July 2017)
The state of Hive and Spark in the Cloud (July 2017)The state of Hive and Spark in the Cloud (July 2017)
The state of Hive and Spark in the Cloud (July 2017)
 
Shootout at the AWS Corral
Shootout at the AWS CorralShootout at the AWS Corral
Shootout at the AWS Corral
 
Scaling AEM (CQ5) Gem Session
Scaling AEM (CQ5) Gem SessionScaling AEM (CQ5) Gem Session
Scaling AEM (CQ5) Gem Session
 
Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...
Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...
Should You Move Between AWS, Azure, or Google Clouds? Considerations, Pros an...
 
DX12 & Vulkan: Dawn of a New Generation of Graphics APIs
DX12 & Vulkan: Dawn of a New Generation of Graphics APIsDX12 & Vulkan: Dawn of a New Generation of Graphics APIs
DX12 & Vulkan: Dawn of a New Generation of Graphics APIs
 
DrupalCampLA 2014 - Drupal backend performance and scalability
DrupalCampLA 2014 - Drupal backend performance and scalabilityDrupalCampLA 2014 - Drupal backend performance and scalability
DrupalCampLA 2014 - Drupal backend performance and scalability
 

Mais de gidgreen

Search Engine Visibility 2013
Search Engine Visibility 2013Search Engine Visibility 2013
Search Engine Visibility 2013gidgreen
 
Marketing for Startups 2013
Marketing for Startups 2013Marketing for Startups 2013
Marketing for Startups 2013gidgreen
 
Selling Advertising 2013
Selling Advertising 2013Selling Advertising 2013
Selling Advertising 2013gidgreen
 
Selling Products and Services 2013
Selling Products and Services 2013Selling Products and Services 2013
Selling Products and Services 2013gidgreen
 
User Interface Design 2013
User Interface Design 2013User Interface Design 2013
User Interface Design 2013gidgreen
 
User Interface Principles 2013
User Interface Principles 2013User Interface Principles 2013
User Interface Principles 2013gidgreen
 
The Software Entrepreneurship Process 2013
The Software Entrepreneurship Process 2013The Software Entrepreneurship Process 2013
The Software Entrepreneurship Process 2013gidgreen
 
Introduction to Software Products and Startups 2013
Introduction to Software Products and Startups 2013Introduction to Software Products and Startups 2013
Introduction to Software Products and Startups 2013gidgreen
 
Question2Answer - September 2012
Question2Answer - September 2012Question2Answer - September 2012
Question2Answer - September 2012gidgreen
 
Search Engine Visibility
Search Engine VisibilitySearch Engine Visibility
Search Engine Visibilitygidgreen
 
Marketing for Startups
Marketing for StartupsMarketing for Startups
Marketing for Startupsgidgreen
 
Analytics and Optimization
Analytics and OptimizationAnalytics and Optimization
Analytics and Optimizationgidgreen
 
Selling Products and Services
Selling Products and ServicesSelling Products and Services
Selling Products and Servicesgidgreen
 
Advertising as a Business Model
Advertising as a Business ModelAdvertising as a Business Model
Advertising as a Business Modelgidgreen
 
Localization and Internationalization
Localization and InternationalizationLocalization and Internationalization
Localization and Internationalizationgidgreen
 
User Interface Principles
User Interface PrinciplesUser Interface Principles
User Interface Principlesgidgreen
 
The Software Entrepreneurship Process
The Software Entrepreneurship ProcessThe Software Entrepreneurship Process
The Software Entrepreneurship Processgidgreen
 
Introduction to Software Products and Startups
Introduction to Software Products and StartupsIntroduction to Software Products and Startups
Introduction to Software Products and Startupsgidgreen
 

Mais de gidgreen (18)

Search Engine Visibility 2013
Search Engine Visibility 2013Search Engine Visibility 2013
Search Engine Visibility 2013
 
Marketing for Startups 2013
Marketing for Startups 2013Marketing for Startups 2013
Marketing for Startups 2013
 
Selling Advertising 2013
Selling Advertising 2013Selling Advertising 2013
Selling Advertising 2013
 
Selling Products and Services 2013
Selling Products and Services 2013Selling Products and Services 2013
Selling Products and Services 2013
 
User Interface Design 2013
User Interface Design 2013User Interface Design 2013
User Interface Design 2013
 
User Interface Principles 2013
User Interface Principles 2013User Interface Principles 2013
User Interface Principles 2013
 
The Software Entrepreneurship Process 2013
The Software Entrepreneurship Process 2013The Software Entrepreneurship Process 2013
The Software Entrepreneurship Process 2013
 
Introduction to Software Products and Startups 2013
Introduction to Software Products and Startups 2013Introduction to Software Products and Startups 2013
Introduction to Software Products and Startups 2013
 
Question2Answer - September 2012
Question2Answer - September 2012Question2Answer - September 2012
Question2Answer - September 2012
 
Search Engine Visibility
Search Engine VisibilitySearch Engine Visibility
Search Engine Visibility
 
Marketing for Startups
Marketing for StartupsMarketing for Startups
Marketing for Startups
 
Analytics and Optimization
Analytics and OptimizationAnalytics and Optimization
Analytics and Optimization
 
Selling Products and Services
Selling Products and ServicesSelling Products and Services
Selling Products and Services
 
Advertising as a Business Model
Advertising as a Business ModelAdvertising as a Business Model
Advertising as a Business Model
 
Localization and Internationalization
Localization and InternationalizationLocalization and Internationalization
Localization and Internationalization
 
User Interface Principles
User Interface PrinciplesUser Interface Principles
User Interface Principles
 
The Software Entrepreneurship Process
The Software Entrepreneurship ProcessThe Software Entrepreneurship Process
The Software Entrepreneurship Process
 
Introduction to Software Products and Startups
Introduction to Software Products and StartupsIntroduction to Software Products and Startups
Introduction to Software Products and Startups
 

Último

TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024Lonnie McRorey
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfRankYa
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostZilliz
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyAlfredo García Lavilla
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLScyllaDB
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):comworks
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 3652toLead Limited
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brandgvaughan
 
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfHyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfPrecisely
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxhariprasad279825
 
Take control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test SuiteTake control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test SuiteDianaGray10
 
Powerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time ClashPowerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time Clashcharlottematthew16
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsMiki Katsuragi
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebUiPathCommunity
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024BookNet Canada
 
How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity PlanDatabarracks
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsMark Billinghurst
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Commit University
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Enterprise Knowledge
 

Último (20)

TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024TeamStation AI System Report LATAM IT Salaries 2024
TeamStation AI System Report LATAM IT Salaries 2024
 
Search Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdfSearch Engine Optimization SEO PDF for 2024.pdf
Search Engine Optimization SEO PDF for 2024.pdf
 
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage CostLeverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
Leverage Zilliz Serverless - Up to 50X Saving for Your Vector Storage Cost
 
Commit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easyCommit 2024 - Secret Management made easy
Commit 2024 - Secret Management made easy
 
Developer Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQLDeveloper Data Modeling Mistakes: From Postgres to NoSQL
Developer Data Modeling Mistakes: From Postgres to NoSQL
 
CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):CloudStudio User manual (basic edition):
CloudStudio User manual (basic edition):
 
Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365Ensuring Technical Readiness For Copilot in Microsoft 365
Ensuring Technical Readiness For Copilot in Microsoft 365
 
WordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your BrandWordPress Websites for Engineers: Elevate Your Brand
WordPress Websites for Engineers: Elevate Your Brand
 
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdfHyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
Hyperautomation and AI/ML: A Strategy for Digital Transformation Success.pdf
 
DMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special EditionDMCC Future of Trade Web3 - Special Edition
DMCC Future of Trade Web3 - Special Edition
 
Artificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptxArtificial intelligence in cctv survelliance.pptx
Artificial intelligence in cctv survelliance.pptx
 
Take control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test SuiteTake control of your SAP testing with UiPath Test Suite
Take control of your SAP testing with UiPath Test Suite
 
Powerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time ClashPowerpoint exploring the locations used in television show Time Clash
Powerpoint exploring the locations used in television show Time Clash
 
Vertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering TipsVertex AI Gemini Prompt Engineering Tips
Vertex AI Gemini Prompt Engineering Tips
 
Dev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio WebDev Dives: Streamline document processing with UiPath Studio Web
Dev Dives: Streamline document processing with UiPath Studio Web
 
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
Transcript: New from BookNet Canada for 2024: BNC CataList - Tech Forum 2024
 
How to write a Business Continuity Plan
How to write a Business Continuity PlanHow to write a Business Continuity Plan
How to write a Business Continuity Plan
 
Human Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR SystemsHuman Factors of XR: Using Human Factors to Design XR Systems
Human Factors of XR: Using Human Factors to Design XR Systems
 
Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!Nell’iperspazio con Rocket: il Framework Web di Rust!
Nell’iperspazio con Rocket: il Framework Web di Rust!
 
Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024Designing IA for AI - Information Architecture Conference 2024
Designing IA for AI - Information Architecture Conference 2024
 

The Secret Guide to Cloud Performance - Cloudlook

  • 1. The Secret Guide to Cloud Performance Jan 30 2014 @ DevCon TLV Gideon Greenspan Co-Founder, Cloudlook
  • 2. Site/DB Site/DB Site/DB OS etc. Site/DB OS + web server + DBMS OS etc. Site/DB Website + Database OS etc. Site/DB Types of hosting OS + web server + DBMS Hardware Hypervisor Host OS Hardware Hardware Dedicated Cloud / VPS Shared Secret Guide to Cloud Performance Slide 2 Cloudlook
  • 3. “The Cloud” vs Old-School VPS •  Deploy/decommission in minutes •  Scale up/down in minutes •  Management APIs •  Automatic scaling •  Billing by the hour •  Migration on hardware failure Secret Guide to Cloud Performance Slide 3 Cloudlook
  • 4. Some public cloud providers Secret Guide to Cloud Performance Slide 4 Cloudlook
  • 5. Cloud specifications — AWS EC2 Secret Guide to Cloud Performance Slide 5 Cloudlook
  • 6. Cloud specs — Rackspace Secret Guide to Cloud Performance Slide 6 Cloudlook
  • 7. The trouble with the cloud •  Vague system specifications •  Variation between instances –  Noisy neighbors? –  Diverse underlying hardware •  Variation over time within an instance •  One-time benchmarks insufficient Secret Guide to Cloud Performance Slide 7 Cloudlook
  • 8. Cloudlook: continuous benchmarking •  6 cloud providers, 35 instance types •  15,000 instances per month •  50 samples per instance over 1 hour •  All 64-bit CentOS (even on Azure!) •  $2,000 per month •  Available free for your server too! Secret Guide to Cloud Performance Slide 8 Cloudlook
  • 9. What does Cloudlook measure? •  Integer CPU performance •  Concurrency •  Memory bandwidth •  Sequential disk access All done with… •  Random disk access •  Host disk cache effect Secret Guide to Cloud Performance Slide 9 Cloudlook
  • 10. The Secret Guide •  Introduction •  Benchmarks •  Conclusions Secret Guide to Cloud Performance Slide 10 Cloudlook
  • 11. Measuring integer CPU •  PHP scripts run as interpreted bytecode •  Built-in PHP functions are compiled C •  We benchmarked 14 heavy functions across 66 versions of PHP ≥ 5.0.4 •  Two performed consistently (±0.2%) –  sha1() — hash function –  levenshtein() — deep string comparison Secret Guide to Cloud Performance Slide 11 Cloudlook
  • 12. CPU results Core i7 equivalent Between instances Within instances 1 GoGrid (all) 2.05 GHz ± 6% ± 4% 2 HP Cloud (all) 2.01 GHz ± 12% ± 10% 3 Rackspace Performance 1 (all) 2.01 GHz ± 4% ± 4% 4 Rackspace Standard (all) 1.93 GHz ± 6% ± 5% 5 Windows Azure (all) 1.74 GHz ± 3% ± 11% 6 Digital Ocean ≥1GB 1.62 GHz ± 19% ± 12% 7 Amazon m1.medium 1.48 GHz ± 6% ± 5% Secret Guide to Cloud Performance Slide 12 Cloudlook
  • 13. CPU variation per instance Rackspace Cloud Performance 1 2GB Digital Ocean 2GB (sometimes) Secret Guide to Cloud Performance Slide 13 Cloudlook
  • 14. Measuring concurrency •  Parallel CPU-heavy HTTP requests •  Precise client-side synchronization –  microtime() + usleep() •  Avoid CPU temperature bias Parallel HTTP requests 1 2 3 4 Secret Guide to Cloud Performance Slide 14 Cloudlook
  • 15. Concurrency results Measured concurrency 5 4 Windows Azure Digital Ocean + HP Cloud Rackspace Performance 1 GoGrid Amazon EC2 Rackspace Standard 3 2 1 1 2 3 4 5 6 7 8 Specified vCPUs / cores Secret Guide to Cloud Performance Slide 15 Cloudlook
  • 16. Measuring memory bandwidth •  Create a 64 MB string in memory –  Largest L1+L2+L3 cache currently ~41 GB •  substr() to read successive windows –  memcpy() in C does the actual work •  Window length = 128 kb –  Too short ⇒ time in bytecode interpreter –  Too long ⇒ copy too big for L1+L2 cache Secret Guide to Cloud Performance Slide 16 Cloudlook
  • 17. Memory results Memory bandwidth Between instances Within instances 1 GoGrid 6.7 GB/s ± 10% ± 7% 2 Rackspace Performance 1 6.7 GB/s ± 7% ± 4% 3 Digital Ocean ≥1GB 6.3 GB/s ± 10% ± 8% 4 HP Cloud 5.9 GB/s ± 9% ± 12% 5 Amazon m1.medium 5.2 GB/s ± 25% ± 4% 6 Rackspace Standard 4.1 GB/s ± 16% ± 7% 7 Windows Azure 3.6 GB/s ± 5% ± 6% Secret Guide to Cloud Performance Slide 17 Cloudlook
  • 18. Memory performance variation HP Cloud Standard Medium Rackspace Cloud Standard 4GB Digital Ocean 4GB Amazon m1.large Secret Guide to Cloud Performance Slide 18 Cloudlook
  • 19. Measuring disk performance •  Write benchmark files in /tmp •  Avoid operating system cache –  Total size > instance memory –  Explicitly clear cache (our instances) –  Read each block once (our instances) •  “Random” is pseudo-random –  Fibonacci hashing scheme Secret Guide to Cloud Performance Slide 19 Cloudlook
  • 20. OS etc. Site/DB OS etc. Site/DB OS etc. Site/DB Host cache interference Hypervisor Host OS Hardware Secret Guide to Cloud Performance •  Results can be skewed •  Big instances: write lots of other data inbetween •  Small instances: wait a week and pray… •  But still worth measuring because it’s a good thing! Slide 20 Cloudlook
  • 21. Disk results (no host cache) 10,000 EC2 m1.medium Primary Disk Seeks — ops/s EC2 m1.medium Secondary Rackspace Standard >= 2GB Rackspace Performance 1 >= 2GB Both 1,000 Azure >= 1.5 GB Primary Azure >= 1.5 GB Secondary HP Cloud >= 2GB Both Digital Ocean >= 1GB 100 GoGrid >= 2 GB 10 10 100 1000 Disk Throughput — MB/s Secret Guide to Cloud Performance Slide 21 Cloudlook
  • 22. Who has a host cache? Primary disk rate (no host cache) Primary disk rate (host cache) Host cache speedup HP Cloud Small 109 MB/s 528 MB/s 4.8 ⤬ HP Cloud Medium 124 MB/s 320 MB/s 2.6 ⤬ HP Cloud Large 110 MB/s 117 MB/s 1.1 ⤬ Digital Ocean 1 GB 244 MB/s 709 MB/s 2.9 ⤬ Digital Ocean 2 GB 267 MB/s 703 MB/s 2.6 ⤬ Digital Ocean 4 GB 260 MB/s 709 MB/s 2.7 ⤬ Digital Ocean 8 GB 318 MB/s 474 MB/s 1.5 ⤬ Secret Guide to Cloud Performance Slide 22 Cloudlook
  • 23. The Secret Guide •  Introduction •  Benchmarks •  Conclusions Secret Guide to Cloud Performance Slide 23 Cloudlook
  • 24. Same instance, different stats EC2 m1.medium GoGrid 4 GB Rackspace Performance 1 4 GB Integer CPU 1.4 : 1 1.2 : 1 1.5 : 1 Concurrency 1.1 : 1 1.5 : 1 1.5 : 1 Memory 3.8 : 1 1.6 : 1 1.7 : 1 Disk throughput 3.0 : 1 10.0 : 1 3.0 : 1 Disk seeks 15.8 : 1 4.2 : 1 3.3 : 1 Secret Guide to Cloud Performance Slide 24 Cloudlook
  • 25. Dedicated vs Cloud Hetzner EX-40 GoGrid 2 GB GoGrid 8 GB Integer CPU 3.49 GHz 2.04 GHz 2.09 GHz Concurrency 6.82 ⤬ 1.83 ⤬ 4.61 ⤬ Memory 12.9 GB/s 6.6 GB/s 6.9 GB/s Disk throughput 155 MB/s 387 MB/s 381 MB/s Disk seeks 134 ops/s 127 ops/s 112 ops/s 32 GB 2 GB 8 GB Disk size 1000 GB 100 GB 400 GB Price $67/mo $66/mo $263/mo Memory size Secret Guide to Cloud Performance Slide 25 Cloudlook
  • 26. Bad reasons to host in the cloud •  Hardware abstraction (peace of mind) –  Last week Rackspace Cloud lost my data •  Easy resizing –  Dedicated is at top of scale anyway –  Can’t resize Rackspace Performance servers •  Lower costs –  No. That’s why hosting companies like it… Secret Guide to Cloud Performance Slide 26 Cloudlook
  • 27. Some good reasons •  Small instances •  Fast deploy/decommission •  Tools and APIs •  Other cloud services –  Storage and CDN –  Databases and messaging Secret Guide to Cloud Performance Slide 27 Cloudlook
  • 28. Image from http://www.kapowwowobjects.com So: Beware of fashions… Secret Guide to Cloud Performance Slide 28 Cloudlook
  • 29. … and measure your own Secret Guide to Cloud Performance Slide 29 Cloudlook
  • 30. Thank you! www.cloudlook.com Secret Guide to Cloud Performance Slide 30 Cloudlook