SlideShare a Scribd company logo
1 of 35
Download to read offline
STORAGE ARCHITECTURES FOR THE MODERN DATA
CENTRE
Chris	
  M	
  Evans
Langton	
  Blue	
  Ltd
Copyright	
  ©	
  2016	
  Langton	
  Blue	
  Ltd 1
A	
  WORD ABOUT ME
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 2
§ 28	
  years	
  experience	
  in	
  the	
  IT	
  industry	
  across	
  all	
  
platforms,	
  including	
  IBM	
  mainframe,	
  Windows	
  
&	
  Open	
  Systems.	
  	
  
§ Co-­‐founder	
  and	
  independent	
  consultant	
  at	
  
Langton	
  Blue	
  Ltd,	
  a	
  specialist	
  consulting	
  
company	
  in	
  the	
  UK.
§ Blogger	
  and	
  part-­‐time	
  analyst	
  at	
  Architecting	
  IT
§ Twitter:	
  @chrismevans,	
  @architectingIT,	
  
@langtonblue
§ Web:	
  www.architecting.it,	
  
www.langtonblue.com
AGENDA
§ Evolution	
  of	
  The	
  Data	
  Centre
§ Understanding	
  Customer	
  Needs
§ Storage	
  Architectures
§ Product	
  Marketplace
§ Q&A
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 3
EVOLUTION	
  OF	
  THE	
  DATA	
  CENTRE
A	
  quick	
  primer	
  on	
  storage	
  heritage
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 4
20TH CENTURY DESIGN
§ Monolithic
§ Bespoke	
  hardware
§ Expensive	
  to	
  acquire,	
  maintain	
  &	
  
manage
§ Static	
  designs	
  and	
  configuration
§ (Think	
  time	
  taken	
  to	
  run	
  binfile
change)
§ Limited	
  or	
  no	
  data	
  services
§ No	
  RAID,	
  replication,	
  snapshots
§ All	
  features	
  that	
  developed	
  over	
  time
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 5
20TH CENTURY ARCHITECTURE
§ Custom	
  hardware	
  designs	
  &	
  components	
  still	
  reign
§ Hosts	
  own	
  LUNs	
  mapped	
  from	
  physical	
  RAID	
  groups
§ Performance	
  defined	
  by	
  RAID	
  group	
  size	
  and	
  disk	
  
speed
§ Wide	
  striping	
  for	
  performance
§ Throughput	
  &	
  I/O	
  latency	
  influenced	
  (and	
  restricted)	
  
by	
  controller	
  capabilities
§ Lots	
  of	
  manual	
  effort	
  to	
  rebalance	
  systems,	
  especially	
  
at	
  scale
§ LUN	
  migration,	
  rebuilds,	
  reliance	
  on	
  host-­‐level	
  LVMs	
  for	
  a	
  
lot	
  of	
  the	
  work
§ Lots	
  of	
  risk	
  fragmentation	
  and	
  orphan	
  storage
§ Wasted	
  space	
  in	
  RAID	
  groups
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 6
CONTROLLER	
  A CONTROLLER	
  B
BED BED BED BED BED BED BED BED
21ST CENTURY TECHNOLOGY
§ Vendors	
  have	
  moved	
  away	
  from	
  bespoke	
  hardware	
  designs
§ x86	
  Standardisation
§ Commodity,	
  reliable	
  components
§ High	
  performance	
  processors	
  &	
  memory	
  – Moore’s	
  Law
§ Interoperability
§ Components	
  generally	
  work	
  together	
  without	
  problems
§ Storage	
  protocols	
  are	
  robust	
  and	
  mature	
  (still	
  also	
  developing	
  – NVMe)
§ New	
  Technology
§ Flash
§ NVDIMM
§ 3D	
  Xpoint
§ Migration	
  of	
  “smarts”	
  to	
  software	
  	
  -­‐ Software	
  Defined	
  Storage
§ Features	
  implemented	
   in	
  code,	
  not	
  microcode
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 7
FLASH &	
  NEW MEDIA
§ Designs
§ SLC	
  (Single	
  Level	
  Cell),	
  MLC	
  (Multi-­‐level	
  Cell),	
  TLC	
  (Triple	
  Level	
  Cell)
§ 3D-­‐NAND	
  – 3-­‐D	
  rather	
  than	
  planar	
  (2D)	
  technology	
  for	
  higher	
  density
§ 3D-­‐Xpoint	
  – faster	
  and	
  more	
  scalable	
  persistent	
  storage	
  than	
  Flash,	
  
slower	
  than	
  DRAM	
  and	
  in	
  between	
  on	
  price
§ Intel/Micron	
  joint	
  development
§ NVRAM/NVDIMM	
  – DIMM	
  form	
  factor	
  products	
  delivering	
  
persistent	
  flash	
  memory	
  on	
  the	
  motherboard
§ NVMe – new	
  connectivity	
  standard	
  for	
  writing	
  to	
  persistent	
  storage
§ PCIe and	
  SDD	
  type	
  form	
  factors	
  (not	
  yet	
  widely	
  adopted)
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 8
PROCESSOR &	
  MEMORY PERFORMANCE
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 9
§ Processor,	
  memory	
  and	
  bus	
  speeds	
  are	
  
increasing
§ Processor	
  performance	
  and	
  bus	
  speeds	
  
follow	
  Moore’s	
  Law
§ Bus	
  speeds	
  double	
  with	
  each	
  release
§ PCIe	
  v1.x	
  =	
  250MB/s	
  (per	
  lane)	
  -­‐ 2003
§ PCIe	
  v2.x	
  =	
  500MB/s	
  (per	
  lane)	
  -­‐ 2007
§ PCIe	
  v3.0	
  =	
  985MB/s	
  (per	
  lane)	
  -­‐ 2010
§ PCIe	
  v4.0	
  =	
  1969MB/s	
  (per	
  lane)	
  – 2014
§ Features	
  much	
  easier	
  to	
  deliver	
  from	
  
software	
  on	
  x86
§ New	
  instruction	
  sets	
  like	
  SSSE3	
  and	
  AVX2	
  
allow	
  storage	
  computations	
  like	
  erasure	
  
coding/Reed	
  Solomon	
  to	
  be	
  done	
  in	
  software
§ Fewer	
  reasons	
  to	
  develop	
  bespoke	
  ASICs	
  and	
  
FGPAs
SOFTWARE DEFINED STORAGE(KEY FEATURES)
§ Abstraction – I/O	
  services	
  should	
  be	
  delivered	
  
independent	
  of	
  the	
  underlying	
  hardware,	
  through	
  logical	
  
constructs	
  like	
  LUNs,	
  volumes,	
  file	
  shares	
  and	
  
repositories.
§ Automation – resources	
  should	
  be	
  consumed	
  using	
  CLIs	
  
and	
  APIs	
  rather	
  than	
  manually	
  allocated	
  through	
  a	
  GUI.
§ Policy/Service	
  Driven	
  – the	
  service	
  received	
  (IOPS,	
  
latency)	
  should	
  be	
  established	
  by	
  policies	
  that	
  
implement	
  Quality	
  of	
  Service,	
  availability	
  and	
  resiliency.
§ Scalable – solutions	
  should	
  enable	
  performance	
  &	
  
capacity	
  scaling	
  independent	
  of	
  I/O	
  delivery.
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 10
MAJOR STORAGE THEMES
§ More	
  choice	
  in	
  products	
  (both	
  hardware	
  and	
  software	
  
than	
  ever	
  before)
§ Divergence	
  – storage	
  platforms	
  for	
  specific	
  purposes	
  
(no	
  single	
  platform	
  for	
  all	
  requirements)
§ Object	
  storage,	
  cloud,	
  backup,	
  primary
§ Convergence	
  – collapsing	
  the	
  storage/server	
  hardware	
  
into	
  a	
  single	
  unit	
  integrating	
  new	
  technology	
  
(converged	
  &	
  hyper-­‐converged)
§ Integration	
  – VVOLs,	
  APIs,	
  platform	
  drivers	
  (e.g.	
  
Cinder),	
  hypervisor	
  extensions	
  (VASA,	
  VAAI)
§ New	
  Media
§ Flash	
  and	
  its	
  successors
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 11
UNDERSTANDING	
  CUSTOMER	
  NEEDS
The	
  changing	
  face	
  of	
  application	
  deployment
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 12
WHAT ARE CUSTOMERS’	
  PAIN POINTS
§ Cost	
  
§ Performance	
  
§ Operational	
  Complexity	
  
§ Reliability
§ Floor	
  Space
§ Power	
  Consumption
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 13
Source:	
  Tintri State	
  of	
  Storage	
  Report,	
  March	
  2015
https://www.tintri.com/sites/default/files/fie ld/pdf/whitepapers/state_of_storage_ infographic_150513t10216.pdf
WHAT DO CUSTOMERS CARE ABOUT
§ Efficiency
§ Cost	
  ($/GB,	
  effective	
  or	
  raw)
§ Physical	
  media
§ Performance
§ Latency,	
  IOPS,	
  throughput,	
  Quality	
  of	
  Service
§ Management
§ Multi-­‐tenancy,	
  APIs	
  (Cinder),	
  Automation,	
  Abstraction	
  (VVOLs)
§ Reliability
§ Resiliency,	
  Availability
§ Features
§ Data	
  Services,	
  Data	
  Protection
§ Breaking	
  the	
  buying	
   cycle
§ No	
  forklift	
  upgrades	
  or	
  3-­‐4	
  year	
  refresh	
  projects
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 14
PREDICTABLYUNPREDICTABLE
§ Workloads	
  becoming	
  more	
  random	
  in	
  nature
§ I/O	
  blender
§ Peaks	
  and	
  troughs	
  in	
  demand
§ VDI	
  (Bootstorms,	
  logoff	
  storms)
§ Increase	
  in	
  I/O	
  Density
§ Driven	
  by	
  virtualisation	
  and	
  soon	
  containers
§ Agility
§ Desire	
  to	
  move	
  data	
  around,	
  spin	
  up	
  copies,	
  create	
  test	
  dev environments,	
  destroy	
  
and	
  recreate
§ Persistence
§ Applications	
  expecting	
  to	
  be	
  transient	
  in	
  nature,	
  but	
  more	
  need	
  for	
  storage	
  to	
  be	
  
100%	
  available	
  across	
  the	
  entire	
  infrastructure
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 15
VIRTUAL FIRST STRATEGY
§ Provided	
  the	
  ability	
  to	
  consolidate	
  over-­‐provisioned	
   server	
  resources
§ Systems	
  typically	
  25%	
  allocated
§ Reduced	
  delivery	
  time	
  for	
  (virtual)	
  servers	
  from	
  days/months	
   to	
  hours
§ Assuming	
  sufficient	
  capacity	
   available
§ Enabled	
  high	
  availability	
  through	
  infrastructure
§ No	
  need	
  to	
  deploy	
  clustered	
  systems
§ Centralised	
  data	
  protection	
  (e.g.	
   VADP)
§ Centralised	
  storage	
  and	
  virtualisation	
  don’t	
  play	
  well	
  together
§ Storage	
  doesn’t	
  understand	
  VMs
§ Storage	
  sees	
  LUNs	
  &	
  volumes
§ Virtualisation	
  sees	
  VMs
§ LUNs	
  encapsulate	
   many	
  volumes	
  – no	
  visibility	
  of	
  VMs
§ Problems	
  being	
  worked	
  around	
  with	
  VVOLs	
  and	
  other	
  solutions	
  that	
  integrate	
  directly	
  
into	
  hyper-­‐converged	
  platforms.
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 16
TECHNICAL REQUIREMENTS– QUALITY OF SERVICE
§ Eliminate	
  the	
  “noisy	
  neighbour”	
  problem
§ Provide	
  storage	
  resources	
  based	
  on	
  policy
§ Ensure	
  storage	
  isn’t	
  under-­‐delivering
§ Guarantee	
  IOPS,	
  performance	
  minima
§ Ensure	
  storage	
  isn’t	
  over-­‐delivering
§ 1st on	
  the	
  array	
  gets	
  best	
  performance	
  (until	
  
there	
  are	
  more)
§ Set	
  limits	
  and	
  prioritisation
§ Apply	
  to	
  VM/VMDK	
  where	
  possible
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 17
TECHNICAL REQUIREMENTS-­‐ API	
  DRIVEN
§ Admin-­‐based	
  functionality	
  driven	
  by	
  API
§ Ability	
  to	
  create,	
  manage,	
  destroy	
  LUNs	
  
from	
  code
§ Integrate	
  storage	
  functions	
  into	
  
workflow
§ Cloud,	
  Automation,	
  Self	
  Service
§ Reduce	
  dependence	
  on	
  manual	
  process
§ Agility
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 18
TECHNICAL REQUIREMENT – SUPPORT VIRTUAL
§ Reduce	
  waste	
  from	
  over-­‐provisioning
§ Support	
  – object	
  (e.g.	
  VM-­‐based)	
  
management
§ Apply	
  policies	
  at	
  the	
  VM	
  level
§ Data	
  location,	
  performance,	
  latency,	
  backup
§ VM	
  friendly	
  for	
  data	
  protection	
  
(snapshots/replication)
§ Offload	
  heavy	
  lifting	
  tasks	
  (e.g.	
  VAAI)
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 19
STORAGE	
  ARCHITECTURES
A	
  few	
  words	
  about	
  design
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 20
DIVERSITY OF STORAGE PLATFORMS
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 21
HYBRID
COMMODITY
OPEN-­‐
SOURCE
APP/DATA	
  
AWARE
SMART	
  
STORAGE
OBJECT	
  
STORES
SCALE-­‐OUT	
  
SAN
SCALE-­‐OUT	
  
NAS
ALL-­‐FLASH
SOFTWARE	
  
DEFINED
SCALE-­‐OUT	
  
BACKUP
ULTRA-­‐FAST	
  
FLASH
HIGH-­‐CAPACITY	
  
FLASH
HPC
INFRASTRUCTURE CONSUMPTION CONTINUUM
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 22
As-­‐a-­‐Service
Hyper-­‐
converged
Converged
Proprietary	
  
Appliance
Software	
  on	
  
Commodity
SMB/Start-­‐ups Large	
  Enterprise Hyperscale &	
  SP
Ease	
  of	
  Implementation
Implementation	
  Flexibility
Degree	
  of	
  Vendor	
  Lock-­‐in
Cost	
  Efficiency
Harder
More	
  Flexible
Less	
  Lock-­‐in
At	
  Large	
  Scale
Easier
Less	
  Flexible
More	
  Lock-­‐in
At	
  Small	
  Scale
SCALE UP
§ Typically	
  Dual	
  controller	
  implementation
§ Active/Active	
  &	
  Active/Passive
§ Throughput	
   limited	
  by	
  controller	
  performance
§ More	
  throughput	
  means	
  better	
  controller	
  
capability	
  (CPU/memory)
§ I/O	
  performance	
  limited	
  by	
  media
§ Flash	
  based	
  implementations	
   failed	
  to	
  use	
  
flash	
  to	
  fullest	
  capability
§ Older	
  designs	
  don’t	
  cater	
  for	
  write	
  endurance	
  
issues
§ Automated	
  tiering capabilities	
   to	
  place	
  data	
  on	
  
most	
  appropriate	
  technology
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 23
CONTROLLER	
  A CONTROLLER	
  B
SCALE OUT (CLOSELY COUPLED)
§ Scale	
  out	
  through	
  adding	
  nodes	
  
and/or	
  capacity
§ Tight	
  node	
  integration	
  either	
  paired	
  
or	
  with	
  bespoke	
  backplane	
  (e.g.	
  
PCIe/Infiniband)
§ Symmetric	
  design	
  – scale	
  with	
  nodes	
  
of	
  similar	
  size/capacity
§ LUNs/Volumes	
  delivered	
  from	
  
specific	
  nodes	
  – not	
  fully	
  load	
  
balanced
§ Scale	
  out	
  limited	
  by	
  backplane	
  traffic
§ Add/remove	
  nodes	
  dynamically
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 24
CTRL	
  A CTRL	
  B CTRL	
  C CTRL	
  D CTRL	
  E CTRL	
  F
SCALE OUT (LOOSELY COUPLED)
§ Fully	
  distributed	
  “Shared	
  Nothing”	
  architecture
§ Loose	
  node	
  coupling	
   (usually	
  IP-­‐based)
§ All	
  nodes	
  contribute	
  storage	
  (no	
  disk	
  shelves)
§ At	
  least	
  one	
  node	
  of	
  “spare”	
  capacity	
  required	
  across	
  the	
  cluster
§ More	
  nodes	
  means	
  more	
  efficiency	
  in	
  redundancy
§ Data	
  resiliency	
  based	
  on	
  replicating	
  data	
  between	
  nodes
§ Standard	
  hardware	
  (typically	
  1U	
  server)	
  design
§ Node	
  power	
  efficiency	
  &	
  size	
  become	
  important
§ High	
  scalability
§ Smaller	
  increments	
  in	
  adding	
  capacity/performance
§ More	
  complex	
  design	
  
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 25
NODE	
  A
NODE	
  B
NODE	
  C
NODE	
  D
NODE	
  E
NODE	
  F
HYPER-­‐CONVERGED
§ Single	
  hardware	
  device	
  for	
  storage	
  and	
  server
§ Node-­‐based	
  scale-­‐out
§ Integrated	
  hypervisor	
  &	
  storage	
  services
§ Storage	
  implemented	
  in	
  VM	
  or	
  hypervisor	
  kernel
§ Distributed	
  file	
  system	
  (scale-­‐out	
  shared	
  nothing)
§ Efficient	
  use	
  of	
  resources
§ Simplified	
  management
§ At	
  scale	
  probably	
  more	
  expensive	
  than	
  separate	
  
components
§ Reduced	
  VM	
  density
§ Issues	
  managing	
  performance	
  vs	
  capacity	
  scaling
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 26
NODE	
  A
NODE	
  B
NODE	
  C
NODE	
  D
NODE	
  E
NODE	
  F
HYPER-­‐CONVERGENCE -­‐ TCO
§ Simplified
§ Less	
  “plan”	
  time	
  -­‐ no	
  requirement	
  to	
  certify	
  and	
  test	
  components
§ Less	
  “build”	
  time	
  – deploy	
  a	
  new	
  node,	
  power	
  up	
  and	
  add	
  to	
  the	
  
configuration
§ Less	
  “operate”	
  time	
  – single	
  management	
  console,	
  typically	
  
integrated	
  with	
  the	
  hypervisor,	
  tight	
  hypervisor	
  awareness
§ Efficient
§ Scale	
  per	
  node;	
  less	
  wastage,	
  no	
  up-­‐front	
  hardware	
  acquisition
§ “Decommission”	
  time	
  almost	
  eliminated;	
   simply	
  add	
  new	
  nodes	
  
and	
  evacuate/remove	
  old	
  nodes	
  over	
  time
§ Provides	
  overall	
  TCO	
  savings	
  (not	
  storage	
  specific)
§ Not	
  typically	
  suited	
  to	
  high-­‐scale	
  environments
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 27
STORAGE VIRTUALISATION-­‐ EVOLUTION
§ Monolithic	
   – single	
  central	
  controller
§ Inline	
  in	
  the	
  data	
  path
§ Controller	
  maps	
  logical	
  to	
  “physical”	
  storage
§ data	
  management	
  features	
  built	
  into	
  the	
  controller
§ Not	
  highly	
  scalable
§ Centralised Metadata	
  – Distributed	
  Data
§ Central	
  metadata	
  functions
§ Data	
  distributed/replicated	
  across	
  many	
  devices
§ System	
  not	
  in	
  the	
  data	
  path
§ Separation	
  of	
  control	
  and	
  data	
  planes
§ Totally	
  Distributed
§ No	
  central	
  metadata	
  or	
  data
§ Data	
  distributed	
  across	
  many	
  devices
§ Fully	
  scalable	
  architecture
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 28
ADVANCED SERVICES
§ Space	
  Efficiency
§ Compression
§ De-­‐duplication
§ Thin	
  Provisioning
§ Quality	
  of	
  Service
§ Object	
  abstraction	
  (VVOLS,	
  per	
  VM	
  services)
§ Application	
  consistent	
  snapshots/backup
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 29
SPACE OPTIMISATION/EFFICIENCY
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 30
Feature Compression De-­‐duplication Thin	
  Provisioning
Physical	
  space	
  reduction ✔ ✔ ❌
Architecture/metadata	
  dependent ❌ ✔ ✔
CPU Friendly ❌ ✔ ✔
Traditional	
  OLTP	
  Database	
  friendly ✔ ❌ ❌
VDI/VSI Friendly ❌ ✔ ✔
HDD	
  Friendly ✔ ❌ ✔
Flash	
  Friendly ✔ ✔ ✔
QUALITY OF SERVICE (AGAIN)
§ Assign	
  service-­‐based	
  metrics	
  to	
  application	
  workloads
§ Reduce/eliminate	
  the	
  “noisy	
  neighbour”	
  effect
§ Set	
  application	
  owner	
  “expectations”	
  on	
  performance
§ Avoid	
  first/last	
  application	
  syndrome	
  from	
  legacy	
  arrays
§ Metrics
§ Latency
§ IOPS
§ Throughput
§ Establish	
  maximum/minimum/burst	
  values	
  &	
  priority
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 31
OBJECT ABSTRACTION
§ Implement	
  per-­‐VM	
  services
§ Abstract	
  away	
  from	
  LUNs/Volumes
§ Implement	
  object-­‐specific	
  services
§ Snapshots
§ Replication
§ Cloning
§ Application-­‐consistent	
  data	
  management
§ Examples
§ VVOLs
§ Object	
  Stores
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 32
VENDOR	
  LANDSCAPE
A	
  look	
  at	
  the	
  product	
  marketplace
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 33
STORAGE NASCAR	
  SLIDE!
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 34
PUTTING IT ALL TOGETHER
§ Start	
  with	
  Requirements
§ Don’t	
  buy	
  what	
  you	
  don’t	
  need
§ Focus	
  on	
  your	
  sensitivities	
  (cost,	
  performance,	
  scalability)
§ See	
  the	
  “bigger	
  picture”
§ Don’t	
  just	
  storage	
  in	
  isolation
§ Don’t	
  buy	
  a	
  technology
§ Look	
  past	
  the	
  glamour	
  of	
  shiny	
  boxes!
§ Evaluate	
  your	
  options
§ Use	
  tools	
  to	
  create	
  an	
  equitable	
  test	
  environment
§ Use	
  online	
  resources	
  and	
  other	
  views
§ Wisdom	
  of	
  the	
  crowd	
  applies
Copyright	
  ©	
  2009-­‐2016 Langton	
  Blue	
  Ltd 35

More Related Content

What's hot

Live CEO Interview and Webinar Update on the State of Deduplication
 Live CEO Interview and Webinar Update on the State of Deduplication Live CEO Interview and Webinar Update on the State of Deduplication
Live CEO Interview and Webinar Update on the State of DeduplicationStorage Switzerland
 
Q&A for TechWiseTV Workshop: HyperFlex
Q&A for TechWiseTV Workshop: HyperFlexQ&A for TechWiseTV Workshop: HyperFlex
Q&A for TechWiseTV Workshop: HyperFlexRobb Boyd
 
Le soluzioni tecnologiche per il Copy Data Management
Le soluzioni tecnologiche per il Copy Data ManagementLe soluzioni tecnologiche per il Copy Data Management
Le soluzioni tecnologiche per il Copy Data ManagementJürgen Ambrosi
 
TwinStrata CloudArray - Disaster Recovery as a Service
TwinStrata CloudArray - Disaster Recovery as a ServiceTwinStrata CloudArray - Disaster Recovery as a Service
TwinStrata CloudArray - Disaster Recovery as a Serviceinside-BigData.com
 
Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014
Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014
Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014Scalar Decisions
 
White paper: IBM FlashSystems in VMware Environments
White paper: IBM FlashSystems in VMware EnvironmentsWhite paper: IBM FlashSystems in VMware Environments
White paper: IBM FlashSystems in VMware EnvironmentsthinkASG
 
Veritas - Software Defined Storage
Veritas - Software Defined StorageVeritas - Software Defined Storage
Veritas - Software Defined StorageJürgen Ambrosi
 
EMC FAST VP for Unified Storage Systems
EMC FAST VP for Unified Storage Systems EMC FAST VP for Unified Storage Systems
EMC FAST VP for Unified Storage Systems EMC
 
Webinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di Utilizzo
Webinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di UtilizzoWebinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di Utilizzo
Webinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di UtilizzoNetgear Italia
 
IBM Storage Virtualization
IBM Storage VirtualizationIBM Storage Virtualization
IBM Storage VirtualizationIBM Danmark
 
All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price
All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price
All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price Western Digital
 
Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...
Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...
Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...inside-BigData.com
 
Evoluzione dello storage
Evoluzione dello storageEvoluzione dello storage
Evoluzione dello storageAndrea Mauro
 
NetApp FAS2200 Series Portfolio
NetApp FAS2200 Series PortfolioNetApp FAS2200 Series Portfolio
NetApp FAS2200 Series PortfolioNetApp
 
Le soluzioni tecnologiche a supporto del mondo OpenStack e Container
Le soluzioni tecnologiche a supporto del mondo OpenStack e ContainerLe soluzioni tecnologiche a supporto del mondo OpenStack e Container
Le soluzioni tecnologiche a supporto del mondo OpenStack e ContainerJürgen Ambrosi
 
Le soluzioni tecnologiche per il disaster recovery e business continuity
Le soluzioni tecnologiche per il disaster recovery e business continuityLe soluzioni tecnologiche per il disaster recovery e business continuity
Le soluzioni tecnologiche per il disaster recovery e business continuityJürgen Ambrosi
 
Guarantee Hyper-V App Performance With Hyper-V Software Defined Storage
Guarantee Hyper-V App Performance With Hyper-V Software Defined StorageGuarantee Hyper-V App Performance With Hyper-V Software Defined Storage
Guarantee Hyper-V App Performance With Hyper-V Software Defined StorageStorage Switzerland
 
My sql competitive update
My sql competitive updateMy sql competitive update
My sql competitive updatexKinAnx
 

What's hot (20)

Live CEO Interview and Webinar Update on the State of Deduplication
 Live CEO Interview and Webinar Update on the State of Deduplication Live CEO Interview and Webinar Update on the State of Deduplication
Live CEO Interview and Webinar Update on the State of Deduplication
 
Q&A for TechWiseTV Workshop: HyperFlex
Q&A for TechWiseTV Workshop: HyperFlexQ&A for TechWiseTV Workshop: HyperFlex
Q&A for TechWiseTV Workshop: HyperFlex
 
Le soluzioni tecnologiche per il Copy Data Management
Le soluzioni tecnologiche per il Copy Data ManagementLe soluzioni tecnologiche per il Copy Data Management
Le soluzioni tecnologiche per il Copy Data Management
 
TwinStrata CloudArray - Disaster Recovery as a Service
TwinStrata CloudArray - Disaster Recovery as a ServiceTwinStrata CloudArray - Disaster Recovery as a Service
TwinStrata CloudArray - Disaster Recovery as a Service
 
Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014
Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014
Storage Considerations for VDI - Scalar presentation at Toronto VMUG 2014
 
White paper: IBM FlashSystems in VMware Environments
White paper: IBM FlashSystems in VMware EnvironmentsWhite paper: IBM FlashSystems in VMware Environments
White paper: IBM FlashSystems in VMware Environments
 
Veritas - Software Defined Storage
Veritas - Software Defined StorageVeritas - Software Defined Storage
Veritas - Software Defined Storage
 
EMC FAST VP for Unified Storage Systems
EMC FAST VP for Unified Storage Systems EMC FAST VP for Unified Storage Systems
EMC FAST VP for Unified Storage Systems
 
Webinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di Utilizzo
Webinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di UtilizzoWebinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di Utilizzo
Webinar NETGEAR - ReadyDATA lo storage di classe Enterprise - Casi di Utilizzo
 
XtremIO
XtremIOXtremIO
XtremIO
 
ravindra_resume.doc
ravindra_resume.docravindra_resume.doc
ravindra_resume.doc
 
IBM Storage Virtualization
IBM Storage VirtualizationIBM Storage Virtualization
IBM Storage Virtualization
 
All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price
All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price
All-Flash Versus Hybrid VMware Virtual SAN™: Performance vs. Price
 
Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...
Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...
Fortissimo Foundation A Clustered, Pervasive, Global Direct-remote I/O Access...
 
Evoluzione dello storage
Evoluzione dello storageEvoluzione dello storage
Evoluzione dello storage
 
NetApp FAS2200 Series Portfolio
NetApp FAS2200 Series PortfolioNetApp FAS2200 Series Portfolio
NetApp FAS2200 Series Portfolio
 
Le soluzioni tecnologiche a supporto del mondo OpenStack e Container
Le soluzioni tecnologiche a supporto del mondo OpenStack e ContainerLe soluzioni tecnologiche a supporto del mondo OpenStack e Container
Le soluzioni tecnologiche a supporto del mondo OpenStack e Container
 
Le soluzioni tecnologiche per il disaster recovery e business continuity
Le soluzioni tecnologiche per il disaster recovery e business continuityLe soluzioni tecnologiche per il disaster recovery e business continuity
Le soluzioni tecnologiche per il disaster recovery e business continuity
 
Guarantee Hyper-V App Performance With Hyper-V Software Defined Storage
Guarantee Hyper-V App Performance With Hyper-V Software Defined StorageGuarantee Hyper-V App Performance With Hyper-V Software Defined Storage
Guarantee Hyper-V App Performance With Hyper-V Software Defined Storage
 
My sql competitive update
My sql competitive updateMy sql competitive update
My sql competitive update
 

Similar to TechTarget Event - Storage Architectures for the Modern Data Centre – Chris Evans

Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to Know
Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to KnowWebinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to Know
Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to KnowStorage Switzerland
 
What’s New in Documentum 7.3
What’s New in Documentum 7.3What’s New in Documentum 7.3
What’s New in Documentum 7.3Michael Mohen
 
Migrate Existing Applications to AWS without Re-engineering
Migrate Existing Applications to AWS without Re-engineeringMigrate Existing Applications to AWS without Re-engineering
Migrate Existing Applications to AWS without Re-engineeringBuurst
 
12 Architectural Requirements for Protecting Business Data in the Cloud
12 Architectural Requirements for Protecting Business Data in the Cloud12 Architectural Requirements for Protecting Business Data in the Cloud
12 Architectural Requirements for Protecting Business Data in the CloudBuurst
 
IBM Spectrum Scale Overview november 2015
IBM Spectrum Scale Overview november 2015IBM Spectrum Scale Overview november 2015
IBM Spectrum Scale Overview november 2015Doug O'Flaherty
 
Red Hat Storage Day Seattle: Why Software-Defined Storage Matters
Red Hat Storage Day Seattle: Why Software-Defined Storage MattersRed Hat Storage Day Seattle: Why Software-Defined Storage Matters
Red Hat Storage Day Seattle: Why Software-Defined Storage MattersRed_Hat_Storage
 
Cloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCO
Cloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCOCloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCO
Cloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCOStorage Switzerland
 
Updates to Apache CloudStack and LINBIT SDS
Updates to Apache CloudStack and LINBIT SDSUpdates to Apache CloudStack and LINBIT SDS
Updates to Apache CloudStack and LINBIT SDSShapeBlue
 
Data core overview - haluk-final
Data core overview - haluk-finalData core overview - haluk-final
Data core overview - haluk-finalHaluk Ulubay
 
Scale up is history! is scale out the future for storage
Scale up is history!  is scale out the future for storageScale up is history!  is scale out the future for storage
Scale up is history! is scale out the future for storageStarWind Software
 
QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014
QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014
QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014Gene Leyzarovich
 
Nimble Storage - The Predicitive Multicloud Flash Fabric
Nimble Storage - The Predicitive Multicloud Flash FabricNimble Storage - The Predicitive Multicloud Flash Fabric
Nimble Storage - The Predicitive Multicloud Flash FabricVITO - Securitas
 
Make a Move to the Azure Cloud with SoftNAS
Make a Move to the Azure Cloud with SoftNASMake a Move to the Azure Cloud with SoftNAS
Make a Move to the Azure Cloud with SoftNASBuurst
 
Deploying All-Flash Cloud Infrastructure without Breaking the Bank
Deploying All-Flash Cloud Infrastructure without Breaking the BankDeploying All-Flash Cloud Infrastructure without Breaking the Bank
Deploying All-Flash Cloud Infrastructure without Breaking the BankWestern Digital
 
DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...
DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...
DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...inside-BigData.com
 
OpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack Features
OpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack FeaturesOpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack Features
OpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack FeaturesEd Balduf
 
Speed up Digital Transformation with Openstack Cloud & Software Defined Storage
Speed up Digital Transformation with Openstack Cloud & Software Defined StorageSpeed up Digital Transformation with Openstack Cloud & Software Defined Storage
Speed up Digital Transformation with Openstack Cloud & Software Defined StorageMatthew Sheppard
 
Webinar: End NAS Sprawl - Gain Control Over Unstructured Data
Webinar: End NAS Sprawl - Gain Control Over Unstructured DataWebinar: End NAS Sprawl - Gain Control Over Unstructured Data
Webinar: End NAS Sprawl - Gain Control Over Unstructured DataStorage Switzerland
 

Similar to TechTarget Event - Storage Architectures for the Modern Data Centre – Chris Evans (20)

Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to Know
Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to KnowWebinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to Know
Webinar: NVMe, NVMe over Fabrics and Beyond - Everything You Need to Know
 
What’s New in Documentum 7.3
What’s New in Documentum 7.3What’s New in Documentum 7.3
What’s New in Documentum 7.3
 
Migrate Existing Applications to AWS without Re-engineering
Migrate Existing Applications to AWS without Re-engineeringMigrate Existing Applications to AWS without Re-engineering
Migrate Existing Applications to AWS without Re-engineering
 
12 Architectural Requirements for Protecting Business Data in the Cloud
12 Architectural Requirements for Protecting Business Data in the Cloud12 Architectural Requirements for Protecting Business Data in the Cloud
12 Architectural Requirements for Protecting Business Data in the Cloud
 
IBM Spectrum Scale Overview november 2015
IBM Spectrum Scale Overview november 2015IBM Spectrum Scale Overview november 2015
IBM Spectrum Scale Overview november 2015
 
Red Hat Storage Day Seattle: Why Software-Defined Storage Matters
Red Hat Storage Day Seattle: Why Software-Defined Storage MattersRed Hat Storage Day Seattle: Why Software-Defined Storage Matters
Red Hat Storage Day Seattle: Why Software-Defined Storage Matters
 
Cloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCO
Cloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCOCloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCO
Cloudian Webinar - 7 Key Reasons why Object Storage lowers Storage TCO
 
Updates to Apache CloudStack and LINBIT SDS
Updates to Apache CloudStack and LINBIT SDSUpdates to Apache CloudStack and LINBIT SDS
Updates to Apache CloudStack and LINBIT SDS
 
Data core overview - haluk-final
Data core overview - haluk-finalData core overview - haluk-final
Data core overview - haluk-final
 
Scale up is history! is scale out the future for storage
Scale up is history!  is scale out the future for storageScale up is history!  is scale out the future for storage
Scale up is history! is scale out the future for storage
 
Mellanox's Sales Strategy
Mellanox's Sales StrategyMellanox's Sales Strategy
Mellanox's Sales Strategy
 
QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014
QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014
QLogic - CrossIT - ACNC/JetStor FibricCache VMUG 2014
 
Nimble Storage - The Predicitive Multicloud Flash Fabric
Nimble Storage - The Predicitive Multicloud Flash FabricNimble Storage - The Predicitive Multicloud Flash Fabric
Nimble Storage - The Predicitive Multicloud Flash Fabric
 
Make a Move to the Azure Cloud with SoftNAS
Make a Move to the Azure Cloud with SoftNASMake a Move to the Azure Cloud with SoftNAS
Make a Move to the Azure Cloud with SoftNAS
 
Deploying All-Flash Cloud Infrastructure without Breaking the Bank
Deploying All-Flash Cloud Infrastructure without Breaking the BankDeploying All-Flash Cloud Infrastructure without Breaking the Bank
Deploying All-Flash Cloud Infrastructure without Breaking the Bank
 
Oracle Storage a ochrana dat
Oracle Storage a ochrana datOracle Storage a ochrana dat
Oracle Storage a ochrana dat
 
DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...
DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...
DDN GS7K - Easy-to-deploy, High Performance Scale-Out Parallel File System Ap...
 
OpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack Features
OpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack FeaturesOpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack Features
OpenStack Silicon Valley - Enterprise Storage Trends Driving OpenStack Features
 
Speed up Digital Transformation with Openstack Cloud & Software Defined Storage
Speed up Digital Transformation with Openstack Cloud & Software Defined StorageSpeed up Digital Transformation with Openstack Cloud & Software Defined Storage
Speed up Digital Transformation with Openstack Cloud & Software Defined Storage
 
Webinar: End NAS Sprawl - Gain Control Over Unstructured Data
Webinar: End NAS Sprawl - Gain Control Over Unstructured DataWebinar: End NAS Sprawl - Gain Control Over Unstructured Data
Webinar: End NAS Sprawl - Gain Control Over Unstructured Data
 

More from NetApp

DevOps the NetApp Way: 10 Rules for Forming a DevOps Team
DevOps the NetApp Way: 10 Rules for Forming a DevOps TeamDevOps the NetApp Way: 10 Rules for Forming a DevOps Team
DevOps the NetApp Way: 10 Rules for Forming a DevOps TeamNetApp
 
10 Reasons to Choose NetApp for EUC/VDI
10 Reasons to Choose NetApp for EUC/VDI10 Reasons to Choose NetApp for EUC/VDI
10 Reasons to Choose NetApp for EUC/VDINetApp
 
Spot Lets NetApp Get the Most Out of the Cloud
Spot Lets NetApp Get the Most Out of the CloudSpot Lets NetApp Get the Most Out of the Cloud
Spot Lets NetApp Get the Most Out of the CloudNetApp
 
NetApp #WFH: COVID-19 Impact Report
NetApp #WFH: COVID-19 Impact ReportNetApp #WFH: COVID-19 Impact Report
NetApp #WFH: COVID-19 Impact ReportNetApp
 
4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success
4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success
4 Ways FlexPod Forms the Foundation for Cisco and NetApp SuccessNetApp
 
NetApp 2020 Predictions
NetApp 2020 Predictions NetApp 2020 Predictions
NetApp 2020 Predictions NetApp
 
NetApp 2020 Predictions
NetApp 2020 Predictions NetApp 2020 Predictions
NetApp 2020 Predictions NetApp
 
NetApp 2020 Predictions in Tech
NetApp 2020 Predictions in TechNetApp 2020 Predictions in Tech
NetApp 2020 Predictions in TechNetApp
 
Corporate IT at NetApp
Corporate IT at NetAppCorporate IT at NetApp
Corporate IT at NetAppNetApp
 
Modernize small and mid-sized enterprise data management with the AFF C190
Modernize small and mid-sized enterprise data management with the AFF C190Modernize small and mid-sized enterprise data management with the AFF C190
Modernize small and mid-sized enterprise data management with the AFF C190NetApp
 
Achieving Target State Architecture in NetApp IT
Achieving Target State Architecture in NetApp ITAchieving Target State Architecture in NetApp IT
Achieving Target State Architecture in NetApp ITNetApp
 
10 Reasons Why Your SAP Applications Belong on NetApp
10 Reasons Why Your SAP Applications Belong on NetApp10 Reasons Why Your SAP Applications Belong on NetApp
10 Reasons Why Your SAP Applications Belong on NetAppNetApp
 
Turbocharge Your Data with Intel Optane Technology and MAX Data
Turbocharge Your Data with Intel Optane Technology and MAX DataTurbocharge Your Data with Intel Optane Technology and MAX Data
Turbocharge Your Data with Intel Optane Technology and MAX DataNetApp
 
Redefining HCI: How to Go from Hyper Converged to Hybrid Cloud Infrastructure
Redefining HCI: How to Go from Hyper Converged to Hybrid Cloud InfrastructureRedefining HCI: How to Go from Hyper Converged to Hybrid Cloud Infrastructure
Redefining HCI: How to Go from Hyper Converged to Hybrid Cloud InfrastructureNetApp
 
Webinar: NetApp SaaS Backup
Webinar: NetApp SaaS BackupWebinar: NetApp SaaS Backup
Webinar: NetApp SaaS BackupNetApp
 
NetApp 2019 Perspectives
NetApp 2019 PerspectivesNetApp 2019 Perspectives
NetApp 2019 PerspectivesNetApp
 
Künstliche Intelligenz ist in deutschen Unter- nehmen Chefsache
Künstliche Intelligenz ist in deutschen Unter- nehmen ChefsacheKünstliche Intelligenz ist in deutschen Unter- nehmen Chefsache
Künstliche Intelligenz ist in deutschen Unter- nehmen ChefsacheNetApp
 
Iperconvergenza come migliora gli economics del tuo IT
Iperconvergenza come migliora gli economics del tuo ITIperconvergenza come migliora gli economics del tuo IT
Iperconvergenza come migliora gli economics del tuo ITNetApp
 
10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning
10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning
10 Good Reasons: NetApp for Artificial Intelligence / Deep LearningNetApp
 
NetApp IT’s Tiered Archive Approach for Active IQ
NetApp IT’s Tiered Archive Approach for Active IQNetApp IT’s Tiered Archive Approach for Active IQ
NetApp IT’s Tiered Archive Approach for Active IQNetApp
 

More from NetApp (20)

DevOps the NetApp Way: 10 Rules for Forming a DevOps Team
DevOps the NetApp Way: 10 Rules for Forming a DevOps TeamDevOps the NetApp Way: 10 Rules for Forming a DevOps Team
DevOps the NetApp Way: 10 Rules for Forming a DevOps Team
 
10 Reasons to Choose NetApp for EUC/VDI
10 Reasons to Choose NetApp for EUC/VDI10 Reasons to Choose NetApp for EUC/VDI
10 Reasons to Choose NetApp for EUC/VDI
 
Spot Lets NetApp Get the Most Out of the Cloud
Spot Lets NetApp Get the Most Out of the CloudSpot Lets NetApp Get the Most Out of the Cloud
Spot Lets NetApp Get the Most Out of the Cloud
 
NetApp #WFH: COVID-19 Impact Report
NetApp #WFH: COVID-19 Impact ReportNetApp #WFH: COVID-19 Impact Report
NetApp #WFH: COVID-19 Impact Report
 
4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success
4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success
4 Ways FlexPod Forms the Foundation for Cisco and NetApp Success
 
NetApp 2020 Predictions
NetApp 2020 Predictions NetApp 2020 Predictions
NetApp 2020 Predictions
 
NetApp 2020 Predictions
NetApp 2020 Predictions NetApp 2020 Predictions
NetApp 2020 Predictions
 
NetApp 2020 Predictions in Tech
NetApp 2020 Predictions in TechNetApp 2020 Predictions in Tech
NetApp 2020 Predictions in Tech
 
Corporate IT at NetApp
Corporate IT at NetAppCorporate IT at NetApp
Corporate IT at NetApp
 
Modernize small and mid-sized enterprise data management with the AFF C190
Modernize small and mid-sized enterprise data management with the AFF C190Modernize small and mid-sized enterprise data management with the AFF C190
Modernize small and mid-sized enterprise data management with the AFF C190
 
Achieving Target State Architecture in NetApp IT
Achieving Target State Architecture in NetApp ITAchieving Target State Architecture in NetApp IT
Achieving Target State Architecture in NetApp IT
 
10 Reasons Why Your SAP Applications Belong on NetApp
10 Reasons Why Your SAP Applications Belong on NetApp10 Reasons Why Your SAP Applications Belong on NetApp
10 Reasons Why Your SAP Applications Belong on NetApp
 
Turbocharge Your Data with Intel Optane Technology and MAX Data
Turbocharge Your Data with Intel Optane Technology and MAX DataTurbocharge Your Data with Intel Optane Technology and MAX Data
Turbocharge Your Data with Intel Optane Technology and MAX Data
 
Redefining HCI: How to Go from Hyper Converged to Hybrid Cloud Infrastructure
Redefining HCI: How to Go from Hyper Converged to Hybrid Cloud InfrastructureRedefining HCI: How to Go from Hyper Converged to Hybrid Cloud Infrastructure
Redefining HCI: How to Go from Hyper Converged to Hybrid Cloud Infrastructure
 
Webinar: NetApp SaaS Backup
Webinar: NetApp SaaS BackupWebinar: NetApp SaaS Backup
Webinar: NetApp SaaS Backup
 
NetApp 2019 Perspectives
NetApp 2019 PerspectivesNetApp 2019 Perspectives
NetApp 2019 Perspectives
 
Künstliche Intelligenz ist in deutschen Unter- nehmen Chefsache
Künstliche Intelligenz ist in deutschen Unter- nehmen ChefsacheKünstliche Intelligenz ist in deutschen Unter- nehmen Chefsache
Künstliche Intelligenz ist in deutschen Unter- nehmen Chefsache
 
Iperconvergenza come migliora gli economics del tuo IT
Iperconvergenza come migliora gli economics del tuo ITIperconvergenza come migliora gli economics del tuo IT
Iperconvergenza come migliora gli economics del tuo IT
 
10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning
10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning
10 Good Reasons: NetApp for Artificial Intelligence / Deep Learning
 
NetApp IT’s Tiered Archive Approach for Active IQ
NetApp IT’s Tiered Archive Approach for Active IQNetApp IT’s Tiered Archive Approach for Active IQ
NetApp IT’s Tiered Archive Approach for Active IQ
 

Recently uploaded

Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherRemote DBA Services
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024The Digital Insurer
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyKhushali Kathiriya
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfOverkill Security
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024The Digital Insurer
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024The Digital Insurer
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesrafiqahmad00786416
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdfSandro Moreira
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistandanishmna97
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...DianaGray10
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FMESafe Software
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...apidays
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodJuan lago vázquez
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...apidays
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxRustici Software
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Jeffrey Haguewood
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processorsdebabhi2
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProduct Anonymous
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...Martijn de Jong
 

Recently uploaded (20)

Strategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a FresherStrategies for Landing an Oracle DBA Job as a Fresher
Strategies for Landing an Oracle DBA Job as a Fresher
 
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data DiscoveryTrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
TrustArc Webinar - Unlock the Power of AI-Driven Data Discovery
 
Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024Axa Assurance Maroc - Insurer Innovation Award 2024
Axa Assurance Maroc - Insurer Innovation Award 2024
 
Artificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : UncertaintyArtificial Intelligence Chap.5 : Uncertainty
Artificial Intelligence Chap.5 : Uncertainty
 
Ransomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdfRansomware_Q4_2023. The report. [EN].pdf
Ransomware_Q4_2023. The report. [EN].pdf
 
Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024Manulife - Insurer Transformation Award 2024
Manulife - Insurer Transformation Award 2024
 
FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024FWD Group - Insurer Innovation Award 2024
FWD Group - Insurer Innovation Award 2024
 
ICT role in 21st century education and its challenges
ICT role in 21st century education and its challengesICT role in 21st century education and its challenges
ICT role in 21st century education and its challenges
 
[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf[BuildWithAI] Introduction to Gemini.pdf
[BuildWithAI] Introduction to Gemini.pdf
 
CNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In PakistanCNIC Information System with Pakdata Cf In Pakistan
CNIC Information System with Pakdata Cf In Pakistan
 
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
Connector Corner: Accelerate revenue generation using UiPath API-centric busi...
 
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers:  A Deep Dive into Serverless Spatial Data and FMECloud Frontiers:  A Deep Dive into Serverless Spatial Data and FME
Cloud Frontiers: A Deep Dive into Serverless Spatial Data and FME
 
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
Apidays New York 2024 - Passkeys: Developing APIs to enable passwordless auth...
 
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin WoodPolkadot JAM Slides - Token2049 - By Dr. Gavin Wood
Polkadot JAM Slides - Token2049 - By Dr. Gavin Wood
 
Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...Apidays New York 2024 - The value of a flexible API Management solution for O...
Apidays New York 2024 - The value of a flexible API Management solution for O...
 
Corporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptxCorporate and higher education May webinar.pptx
Corporate and higher education May webinar.pptx
 
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
Web Form Automation for Bonterra Impact Management (fka Social Solutions Apri...
 
Exploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone ProcessorsExploring the Future Potential of AI-Enabled Smartphone Processors
Exploring the Future Potential of AI-Enabled Smartphone Processors
 
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemkeProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
ProductAnonymous-April2024-WinProductDiscovery-MelissaKlemke
 
2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...2024: Domino Containers - The Next Step. News from the Domino Container commu...
2024: Domino Containers - The Next Step. News from the Domino Container commu...
 

TechTarget Event - Storage Architectures for the Modern Data Centre – Chris Evans

  • 1. STORAGE ARCHITECTURES FOR THE MODERN DATA CENTRE Chris  M  Evans Langton  Blue  Ltd Copyright  ©  2016  Langton  Blue  Ltd 1
  • 2. A  WORD ABOUT ME Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 2 § 28  years  experience  in  the  IT  industry  across  all   platforms,  including  IBM  mainframe,  Windows   &  Open  Systems.     § Co-­‐founder  and  independent  consultant  at   Langton  Blue  Ltd,  a  specialist  consulting   company  in  the  UK. § Blogger  and  part-­‐time  analyst  at  Architecting  IT § Twitter:  @chrismevans,  @architectingIT,   @langtonblue § Web:  www.architecting.it,   www.langtonblue.com
  • 3. AGENDA § Evolution  of  The  Data  Centre § Understanding  Customer  Needs § Storage  Architectures § Product  Marketplace § Q&A Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 3
  • 4. EVOLUTION  OF  THE  DATA  CENTRE A  quick  primer  on  storage  heritage Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 4
  • 5. 20TH CENTURY DESIGN § Monolithic § Bespoke  hardware § Expensive  to  acquire,  maintain  &   manage § Static  designs  and  configuration § (Think  time  taken  to  run  binfile change) § Limited  or  no  data  services § No  RAID,  replication,  snapshots § All  features  that  developed  over  time Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 5
  • 6. 20TH CENTURY ARCHITECTURE § Custom  hardware  designs  &  components  still  reign § Hosts  own  LUNs  mapped  from  physical  RAID  groups § Performance  defined  by  RAID  group  size  and  disk   speed § Wide  striping  for  performance § Throughput  &  I/O  latency  influenced  (and  restricted)   by  controller  capabilities § Lots  of  manual  effort  to  rebalance  systems,  especially   at  scale § LUN  migration,  rebuilds,  reliance  on  host-­‐level  LVMs  for  a   lot  of  the  work § Lots  of  risk  fragmentation  and  orphan  storage § Wasted  space  in  RAID  groups Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 6 CONTROLLER  A CONTROLLER  B BED BED BED BED BED BED BED BED
  • 7. 21ST CENTURY TECHNOLOGY § Vendors  have  moved  away  from  bespoke  hardware  designs § x86  Standardisation § Commodity,  reliable  components § High  performance  processors  &  memory  – Moore’s  Law § Interoperability § Components  generally  work  together  without  problems § Storage  protocols  are  robust  and  mature  (still  also  developing  – NVMe) § New  Technology § Flash § NVDIMM § 3D  Xpoint § Migration  of  “smarts”  to  software    -­‐ Software  Defined  Storage § Features  implemented   in  code,  not  microcode Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 7
  • 8. FLASH &  NEW MEDIA § Designs § SLC  (Single  Level  Cell),  MLC  (Multi-­‐level  Cell),  TLC  (Triple  Level  Cell) § 3D-­‐NAND  – 3-­‐D  rather  than  planar  (2D)  technology  for  higher  density § 3D-­‐Xpoint  – faster  and  more  scalable  persistent  storage  than  Flash,   slower  than  DRAM  and  in  between  on  price § Intel/Micron  joint  development § NVRAM/NVDIMM  – DIMM  form  factor  products  delivering   persistent  flash  memory  on  the  motherboard § NVMe – new  connectivity  standard  for  writing  to  persistent  storage § PCIe and  SDD  type  form  factors  (not  yet  widely  adopted) Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 8
  • 9. PROCESSOR &  MEMORY PERFORMANCE Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 9 § Processor,  memory  and  bus  speeds  are   increasing § Processor  performance  and  bus  speeds   follow  Moore’s  Law § Bus  speeds  double  with  each  release § PCIe  v1.x  =  250MB/s  (per  lane)  -­‐ 2003 § PCIe  v2.x  =  500MB/s  (per  lane)  -­‐ 2007 § PCIe  v3.0  =  985MB/s  (per  lane)  -­‐ 2010 § PCIe  v4.0  =  1969MB/s  (per  lane)  – 2014 § Features  much  easier  to  deliver  from   software  on  x86 § New  instruction  sets  like  SSSE3  and  AVX2   allow  storage  computations  like  erasure   coding/Reed  Solomon  to  be  done  in  software § Fewer  reasons  to  develop  bespoke  ASICs  and   FGPAs
  • 10. SOFTWARE DEFINED STORAGE(KEY FEATURES) § Abstraction – I/O  services  should  be  delivered   independent  of  the  underlying  hardware,  through  logical   constructs  like  LUNs,  volumes,  file  shares  and   repositories. § Automation – resources  should  be  consumed  using  CLIs   and  APIs  rather  than  manually  allocated  through  a  GUI. § Policy/Service  Driven  – the  service  received  (IOPS,   latency)  should  be  established  by  policies  that   implement  Quality  of  Service,  availability  and  resiliency. § Scalable – solutions  should  enable  performance  &   capacity  scaling  independent  of  I/O  delivery. Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 10
  • 11. MAJOR STORAGE THEMES § More  choice  in  products  (both  hardware  and  software   than  ever  before) § Divergence  – storage  platforms  for  specific  purposes   (no  single  platform  for  all  requirements) § Object  storage,  cloud,  backup,  primary § Convergence  – collapsing  the  storage/server  hardware   into  a  single  unit  integrating  new  technology   (converged  &  hyper-­‐converged) § Integration  – VVOLs,  APIs,  platform  drivers  (e.g.   Cinder),  hypervisor  extensions  (VASA,  VAAI) § New  Media § Flash  and  its  successors Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 11
  • 12. UNDERSTANDING  CUSTOMER  NEEDS The  changing  face  of  application  deployment Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 12
  • 13. WHAT ARE CUSTOMERS’  PAIN POINTS § Cost   § Performance   § Operational  Complexity   § Reliability § Floor  Space § Power  Consumption Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 13 Source:  Tintri State  of  Storage  Report,  March  2015 https://www.tintri.com/sites/default/files/fie ld/pdf/whitepapers/state_of_storage_ infographic_150513t10216.pdf
  • 14. WHAT DO CUSTOMERS CARE ABOUT § Efficiency § Cost  ($/GB,  effective  or  raw) § Physical  media § Performance § Latency,  IOPS,  throughput,  Quality  of  Service § Management § Multi-­‐tenancy,  APIs  (Cinder),  Automation,  Abstraction  (VVOLs) § Reliability § Resiliency,  Availability § Features § Data  Services,  Data  Protection § Breaking  the  buying   cycle § No  forklift  upgrades  or  3-­‐4  year  refresh  projects Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 14
  • 15. PREDICTABLYUNPREDICTABLE § Workloads  becoming  more  random  in  nature § I/O  blender § Peaks  and  troughs  in  demand § VDI  (Bootstorms,  logoff  storms) § Increase  in  I/O  Density § Driven  by  virtualisation  and  soon  containers § Agility § Desire  to  move  data  around,  spin  up  copies,  create  test  dev environments,  destroy   and  recreate § Persistence § Applications  expecting  to  be  transient  in  nature,  but  more  need  for  storage  to  be   100%  available  across  the  entire  infrastructure Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 15
  • 16. VIRTUAL FIRST STRATEGY § Provided  the  ability  to  consolidate  over-­‐provisioned   server  resources § Systems  typically  25%  allocated § Reduced  delivery  time  for  (virtual)  servers  from  days/months   to  hours § Assuming  sufficient  capacity   available § Enabled  high  availability  through  infrastructure § No  need  to  deploy  clustered  systems § Centralised  data  protection  (e.g.   VADP) § Centralised  storage  and  virtualisation  don’t  play  well  together § Storage  doesn’t  understand  VMs § Storage  sees  LUNs  &  volumes § Virtualisation  sees  VMs § LUNs  encapsulate   many  volumes  – no  visibility  of  VMs § Problems  being  worked  around  with  VVOLs  and  other  solutions  that  integrate  directly   into  hyper-­‐converged  platforms. Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 16
  • 17. TECHNICAL REQUIREMENTS– QUALITY OF SERVICE § Eliminate  the  “noisy  neighbour”  problem § Provide  storage  resources  based  on  policy § Ensure  storage  isn’t  under-­‐delivering § Guarantee  IOPS,  performance  minima § Ensure  storage  isn’t  over-­‐delivering § 1st on  the  array  gets  best  performance  (until   there  are  more) § Set  limits  and  prioritisation § Apply  to  VM/VMDK  where  possible Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 17
  • 18. TECHNICAL REQUIREMENTS-­‐ API  DRIVEN § Admin-­‐based  functionality  driven  by  API § Ability  to  create,  manage,  destroy  LUNs   from  code § Integrate  storage  functions  into   workflow § Cloud,  Automation,  Self  Service § Reduce  dependence  on  manual  process § Agility Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 18
  • 19. TECHNICAL REQUIREMENT – SUPPORT VIRTUAL § Reduce  waste  from  over-­‐provisioning § Support  – object  (e.g.  VM-­‐based)   management § Apply  policies  at  the  VM  level § Data  location,  performance,  latency,  backup § VM  friendly  for  data  protection   (snapshots/replication) § Offload  heavy  lifting  tasks  (e.g.  VAAI) Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 19
  • 20. STORAGE  ARCHITECTURES A  few  words  about  design Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 20
  • 21. DIVERSITY OF STORAGE PLATFORMS Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 21 HYBRID COMMODITY OPEN-­‐ SOURCE APP/DATA   AWARE SMART   STORAGE OBJECT   STORES SCALE-­‐OUT   SAN SCALE-­‐OUT   NAS ALL-­‐FLASH SOFTWARE   DEFINED SCALE-­‐OUT   BACKUP ULTRA-­‐FAST   FLASH HIGH-­‐CAPACITY   FLASH HPC
  • 22. INFRASTRUCTURE CONSUMPTION CONTINUUM Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 22 As-­‐a-­‐Service Hyper-­‐ converged Converged Proprietary   Appliance Software  on   Commodity SMB/Start-­‐ups Large  Enterprise Hyperscale &  SP Ease  of  Implementation Implementation  Flexibility Degree  of  Vendor  Lock-­‐in Cost  Efficiency Harder More  Flexible Less  Lock-­‐in At  Large  Scale Easier Less  Flexible More  Lock-­‐in At  Small  Scale
  • 23. SCALE UP § Typically  Dual  controller  implementation § Active/Active  &  Active/Passive § Throughput   limited  by  controller  performance § More  throughput  means  better  controller   capability  (CPU/memory) § I/O  performance  limited  by  media § Flash  based  implementations   failed  to  use   flash  to  fullest  capability § Older  designs  don’t  cater  for  write  endurance   issues § Automated  tiering capabilities   to  place  data  on   most  appropriate  technology Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 23 CONTROLLER  A CONTROLLER  B
  • 24. SCALE OUT (CLOSELY COUPLED) § Scale  out  through  adding  nodes   and/or  capacity § Tight  node  integration  either  paired   or  with  bespoke  backplane  (e.g.   PCIe/Infiniband) § Symmetric  design  – scale  with  nodes   of  similar  size/capacity § LUNs/Volumes  delivered  from   specific  nodes  – not  fully  load   balanced § Scale  out  limited  by  backplane  traffic § Add/remove  nodes  dynamically Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 24 CTRL  A CTRL  B CTRL  C CTRL  D CTRL  E CTRL  F
  • 25. SCALE OUT (LOOSELY COUPLED) § Fully  distributed  “Shared  Nothing”  architecture § Loose  node  coupling   (usually  IP-­‐based) § All  nodes  contribute  storage  (no  disk  shelves) § At  least  one  node  of  “spare”  capacity  required  across  the  cluster § More  nodes  means  more  efficiency  in  redundancy § Data  resiliency  based  on  replicating  data  between  nodes § Standard  hardware  (typically  1U  server)  design § Node  power  efficiency  &  size  become  important § High  scalability § Smaller  increments  in  adding  capacity/performance § More  complex  design   Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 25 NODE  A NODE  B NODE  C NODE  D NODE  E NODE  F
  • 26. HYPER-­‐CONVERGED § Single  hardware  device  for  storage  and  server § Node-­‐based  scale-­‐out § Integrated  hypervisor  &  storage  services § Storage  implemented  in  VM  or  hypervisor  kernel § Distributed  file  system  (scale-­‐out  shared  nothing) § Efficient  use  of  resources § Simplified  management § At  scale  probably  more  expensive  than  separate   components § Reduced  VM  density § Issues  managing  performance  vs  capacity  scaling Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 26 NODE  A NODE  B NODE  C NODE  D NODE  E NODE  F
  • 27. HYPER-­‐CONVERGENCE -­‐ TCO § Simplified § Less  “plan”  time  -­‐ no  requirement  to  certify  and  test  components § Less  “build”  time  – deploy  a  new  node,  power  up  and  add  to  the   configuration § Less  “operate”  time  – single  management  console,  typically   integrated  with  the  hypervisor,  tight  hypervisor  awareness § Efficient § Scale  per  node;  less  wastage,  no  up-­‐front  hardware  acquisition § “Decommission”  time  almost  eliminated;   simply  add  new  nodes   and  evacuate/remove  old  nodes  over  time § Provides  overall  TCO  savings  (not  storage  specific) § Not  typically  suited  to  high-­‐scale  environments Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 27
  • 28. STORAGE VIRTUALISATION-­‐ EVOLUTION § Monolithic   – single  central  controller § Inline  in  the  data  path § Controller  maps  logical  to  “physical”  storage § data  management  features  built  into  the  controller § Not  highly  scalable § Centralised Metadata  – Distributed  Data § Central  metadata  functions § Data  distributed/replicated  across  many  devices § System  not  in  the  data  path § Separation  of  control  and  data  planes § Totally  Distributed § No  central  metadata  or  data § Data  distributed  across  many  devices § Fully  scalable  architecture Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 28
  • 29. ADVANCED SERVICES § Space  Efficiency § Compression § De-­‐duplication § Thin  Provisioning § Quality  of  Service § Object  abstraction  (VVOLS,  per  VM  services) § Application  consistent  snapshots/backup Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 29
  • 30. SPACE OPTIMISATION/EFFICIENCY Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 30 Feature Compression De-­‐duplication Thin  Provisioning Physical  space  reduction ✔ ✔ ❌ Architecture/metadata  dependent ❌ ✔ ✔ CPU Friendly ❌ ✔ ✔ Traditional  OLTP  Database  friendly ✔ ❌ ❌ VDI/VSI Friendly ❌ ✔ ✔ HDD  Friendly ✔ ❌ ✔ Flash  Friendly ✔ ✔ ✔
  • 31. QUALITY OF SERVICE (AGAIN) § Assign  service-­‐based  metrics  to  application  workloads § Reduce/eliminate  the  “noisy  neighbour”  effect § Set  application  owner  “expectations”  on  performance § Avoid  first/last  application  syndrome  from  legacy  arrays § Metrics § Latency § IOPS § Throughput § Establish  maximum/minimum/burst  values  &  priority Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 31
  • 32. OBJECT ABSTRACTION § Implement  per-­‐VM  services § Abstract  away  from  LUNs/Volumes § Implement  object-­‐specific  services § Snapshots § Replication § Cloning § Application-­‐consistent  data  management § Examples § VVOLs § Object  Stores Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 32
  • 33. VENDOR  LANDSCAPE A  look  at  the  product  marketplace Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 33
  • 34. STORAGE NASCAR  SLIDE! Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 34
  • 35. PUTTING IT ALL TOGETHER § Start  with  Requirements § Don’t  buy  what  you  don’t  need § Focus  on  your  sensitivities  (cost,  performance,  scalability) § See  the  “bigger  picture” § Don’t  just  storage  in  isolation § Don’t  buy  a  technology § Look  past  the  glamour  of  shiny  boxes! § Evaluate  your  options § Use  tools  to  create  an  equitable  test  environment § Use  online  resources  and  other  views § Wisdom  of  the  crowd  applies Copyright  ©  2009-­‐2016 Langton  Blue  Ltd 35