Upload
databarracks
View
192
Download
0
Tags:
Embed Size (px)
DESCRIPTION
Mark Thomas from Databarracks and Tim Pitcher from SolidFire demonstrate how to keep cloud performance consistent at Cloud Expo Europe 2014.
Citation preview
Demonstrating how to keep your cloud performance consistent
www.databarracks.com | 2www.databarracks.com | 2
WHO WE ARE
Tim PitcherVice President, International
Responsible for expanding SolidFire's presence globally. Tim has significant international experience, most recently serving as Senior Director of Global Account Storage at Hewlett Packard after its acquisition of 3PAR and was VP of global accounts and VP of Northern Europe for NetApp.
Mark ThomasSolutions Architect
Formerly Director of Cloud Professional Services, EMEA at Virtustream, Mark is the Solutions Architect at Databarracks. An expert in cloud technology, data centre infrastructure and virtualisation, Mark has worked with major clients such as HSBC, Field Fisher Waterhouse and Allied Irish Bank.
www.databarracks.com | 3
About Databarracks
• Nuclear bunker data centre, certified & accredited
Secure & Compliant
• Pedigree and understanding of storage
High Performance & Flexibility
www.databarracks.com | 4
Databarracks customers
www.databarracks.com | 5
About SolidFire
• Storage systems built for the next generation data center
• All-flash architecture, with volume-level Quality of Service (QoS) controls
• Guaranteed storage performance to thousands of applications within a shared infrastructure
• In-line data reduction techniques and system-wide automation for capital and operating cost savings relative to traditional storage systems
The cloud storage problem
67% Of all
workloads will be run in the
cloud by 2016
$150B
In total cloud revenue By 2016
Up from $46b in 2008
5x more
Growth in cloud than in
Traditional I.T.through 2016 $12.2b
In public cloud storage sales in
2016Up from $5.6b in 2012
www.databarracks.com | 9www.databarracks.com | 9
The Cloud Evolution
High
Med
Low
IOPS
ApplicationsTest / DevelopmentBackup / ArchiveStartups
Cloud today
Cloud 1.0
Cloud 2.0
Performance Sensitive AppsOracle / SAP / Private CloudHadoop / NoSQLMS Exchange, VDIERP,CRM
Cloud Evolution
$$$
$$
$
• High Performance
• QoS / Hard SLAs• Massive scale• Reliability• Security
Early cloud
www.databarracks.com | 10www.databarracks.com | 10
Why don’t you run performance apps in the cloud?
• Current storage architectures delivers inconsistent and
variable performance (‘noisy neighbour’ effect)
• Inability to efficiently scale performance
• Unable to throttle performance independent of capacity
• Low levels of transparency (no visibility into systems)
• Dedicated storage array costs are prohibitive
• Perception of unreliability
www.databarracks.com | 11www.databarracks.com | 11
Enterprise IT lacks storage agility, and is under significant pressure
• Deploy new applications and capabilities faster
• Provide more agile and scalable infrastructure
• Increase application performance and predictability
• Enable automation and end-user self-service
• Raise operational efficiency and reduce cost
www.databarracks.com | 12www.databarracks.com | 12
Traditional enterprise storage falls short of these requirements
xDeploy new applications and capabilities faster
xProvide more agile and scalable infrastructure
xIncrease application performance and predictability
xEnable automation and end-user self-service
xRaise operational efficiency and reduce cost
www.databarracks.com | 13www.databarracks.com | 13
The Cloud Needs Better Storage
Performance
• Unable to manage performance independent of capacity
• Can not guarantee storage performance
Efficiency
• Low and inefficient utilization rates
• Lack of high performance in-line data reduction
Management
• Complex, manual, lacks automation
Scale
• Limited scalability of both capacity and performance
• Manage multiple islands of storage
www.databarracks.com | 14www.databarracks.com | 14
Key Differentiators
In-lineefficiencyIn-line data reduction and 85% utilization requires less purchased capacity
Guaranteed Quality of Service (QoS)Fine-grain performance management on a per volume basis
Cloud ScalabilitySimultaneous scaling of both capacity and performance
Complete automationREST-based API for complete control
Demonstration:How it actually works
www.databarracks.com | 16www.databarracks.com | 16
www.databarracks.com | 17www.databarracks.com | 17
Eliminating Noisy Neighbours
The Noisy Neighbour Effect Individual tenant impacts other applications
Unsuitable for performance sensitive apps
Quality of Service in Practice Create fine-grained tiers of performance
Application performance is isolated
Performance SLAs enforced
Noisy Neighbour
Tier 0
Tier 1
Tier 2
Tier 3
Decreased Performance
Predictability
Consistent storageperformance for
critical enterprise applications
QoSGuaranteedWith SLA’s
Flexibility
To scale. No Risk.
No Migrations. ££ PriceDedicated Flash
storage performance with
a cloud priced value proposition
www.databarracks.com | 19www.databarracks.com | 19
Resou
rces c
on
su
med
Baseline capacity
100% reservation
x
0%
Over timeVM A
SecureInstance™
VM operates within configured resources.
100%
VM resource usage is constrained by configuration. Fixed cost per VM configuration
Fixed Cost
www.databarracks.com | 20www.databarracks.com | 20
Baseline capacity
Reserved capacity
xx x
VMs operate within configured resources. Aggregate usage as VMs peak and trough
Usage over time is aggregated. Reserved threshold can be moved to adjust to standard business usage
100%
100%
100%
0%0%0% Client A
SecurePool™
Resou
rces c
on
su
med
Over time
Fixed Cost
PAYG
Thank you