SlideShare a Scribd company logo
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
In-­‐Memory	
  Compu.ng	
  and	
  the	
  Emergence	
  of	
  Tier	
  -­‐1	
  Storage	
  
John	
  Webster	
  
Senior	
  Partner	
  
John@EvaluatorGroup.com	
  
7/13/15	
   1	
  
©	
  2012	
  Evaluator	
  Group,	
  Inc.	
  
Overview of Evaluator Group
! Since	
  1997	
  delivering	
  comprehensive	
  and	
  objecHve	
  analysis	
  of	
  
storage	
  systems	
  and	
  related	
  issues	
  
—  A	
  team	
  of	
  engineers,	
  systems	
  performance	
  experts	
  and	
  go-­‐to-­‐market	
  
specialists	
  
—  Expert	
  analysis	
  with	
  detailed	
  informaHon	
  and	
  unbiased	
  opinions	
  
! PracHcal,	
  business-­‐oriented,	
  technical	
  perspecHves	
  
Slide 2
©	
  2012	
  Evaluator	
  Group,	
  Inc.	
  
Research Coverage
7/13/15	
   3	
  
InformaHon	
  
Management	
  &	
  
Data	
  Storage	
  
Virtualiza.on	
  
Data	
  
Analy.cs	
  
Performance	
  
Solid	
  State	
  
Hyper-­‐
Converged	
  
Cloud	
  
Services	
  &	
  
Object	
  
SDS	
  &	
  
Virtualiza.on	
  
Data	
  
Protec.on	
  &	
  
Archive	
  
©	
  2012	
  Evaluator	
  Group,	
  Inc.	
  
A Few of Our Customers
Slide	
  
4	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
What’s the Point?
!  A	
  new	
  scalable,	
  very	
  high	
  performance,	
  low	
  cost	
  Her	
  is	
  
persistent	
  storage	
  is	
  emerging	
  (Tier	
  -­‐1).	
  
!  AdopHon	
  will	
  be	
  accelerated	
  by	
  rapid	
  advances	
  in	
  price	
  for	
  
performance	
  and	
  capacity.	
  
!  It	
  will,	
  converged	
  with	
  other	
  technologies,	
  enable	
  broad	
  
adopHon	
  of	
  real	
  Hme	
  analyHcs.	
  
!  Market	
  potenHal	
  for	
  real	
  Hme	
  analyHcs—and	
  real	
  Hme	
  
compuHng	
  in	
  general—already	
  exists	
  (and	
  has	
  existed	
  for	
  
years).	
  
!  At	
  present,	
  the	
  possibiliHes	
  can	
  only	
  be	
  parHally	
  imaged.	
  
	
  
7/13/15	
   5	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
What is Tier -1?
7/13/15	
   6	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
SSD (Tier 0)
High-Performance
Disks (Tier 1)
High-Capacity
Disks (Tier 2)
External Tier 3
Tape
Optical
DataMovement
Storage System Tiering “Within the Box” and External
7/13/15	
   7	
  
	
  
Response	
  Time	
  
	
  
µs	
  
	
  
ms	
  	
  
	
  
	
  
	
  
	
  
ms	
  
	
  
	
  
	
  
	
  
	
  
seconds	
  
SSD	
  =	
  Solid	
  State	
  
Device	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Persistent In-
Memory SSD
High-Capacity
Disks
Tier -1 Data Persistence In-Memory
7/13/15	
   8	
  
Persistent In-
Memory SSD
Capacity SSD
.....	
  
_
SSD	
  =	
  Solid	
  State	
  
Device	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
What Does the Future for
SSD Look Like?
7/13/15	
   9	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Why Enterprise IT Likes SSD
!  Elongate	
  tech	
  refresh	
  cycles	
  	
  
!  Increased	
  efficiency/reduced	
  cost	
  of	
  virtualized	
  
servers	
  
!  DeduplicaHon	
  yields	
  gains	
  in	
  price/GB	
  vs.	
  disk	
  
!  Lower	
  Opex	
  for	
  machine	
  room	
  environmentals	
  
!  ApplicaHon	
  performance	
  (turn	
  muts	
  into	
  show	
  dogs)	
  
7/13/15	
   10	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Why Business Likes SSD
7/13/15	
   11	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Multiple SSD Technologies Advancing in Parallel
Non-­‐volaHle	
  memory	
  (NVM)	
  technologies	
  under	
  current	
  
development	
  
!  NAND	
  flash	
  
!  MRAM	
  
!  Memristor	
  
!  ResisHve	
  RAM	
  
!  PCM	
  
!  Nano	
  
12	
  7/13/15	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Cost of Disk vs Flash
7/13/15	
   13	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Moore’s Law for DRAM
7/13/15	
   14	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
NVMe Performance
7/13/15	
   15	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Bottom Line
Increasing	
  demand	
  +	
  increasing	
  compe..on	
  +	
  increasing	
  produc.on	
  
for	
  SSD	
  
=	
  
Con.nual	
  improvements	
  in	
  SSD	
  price	
  for	
  performance	
  and	
  scale	
  
=	
  
Con.nued	
  cost	
  reduc.ons	
  and	
  performance	
  enhancements	
  for	
  In-­‐
Memory	
  Compu.ng	
  
=	
  
Accelera.ng	
  advancements	
  in	
  real	
  .me	
  
compu.ng	
  
7/13/15	
   16	
  
©	
  2015	
  Evaluator	
  Group,	
  Inc.	
  
Where is the Demand for
Real Time Computing?
7/13/15	
   17	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
Enabling Converged, Real-time Analytics
7/13/15	
   18	
  
The	
  confluence	
  of	
  two	
  
major	
  compuHng	
  
trajectories:	
  	
  
!  ConHnued	
  improvements	
  in	
  
memory	
  technology	
  that	
  
began	
  with	
  the	
  first	
  digital	
  
computers	
  	
  
!  Advancement	
  of	
  MPP	
  
compuHng	
  that	
  began	
  in	
  the	
  
early	
  1980's	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
Visions of Real Time Analytics in 2004
7/13/15	
   19	
  
!  AnalyHc-­‐based	
  applicaHons	
  
based	
  on	
  convergence	
  of	
  
mulHple	
  data	
  sources-­‐
potenHally	
  in	
  real	
  Hme	
  
!  Interviewed	
  50	
  CEOs,	
  CIOs	
  
and	
  Technologists	
  from	
  
verHcal	
  industry	
  segments	
  
!  Visions	
  were	
  of	
  leveraging	
  
disparate	
  and	
  pervasive	
  
data	
  sources	
  (mobile,	
  
sensor,	
  database)	
  in	
  real	
  
Hme.	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
CEO Real Time Visions 2004
!  RFID	
  converged	
  with	
  paHent	
  data	
  in	
  to	
  prevent	
  accidentally	
  
inducing	
  adverse	
  reacHons	
  to	
  drugs	
  in	
  a	
  hospital	
  sejng	
  
!  MulHple	
  sensory	
  data	
  inputs	
  to	
  monitor	
  and	
  control	
  chemical	
  
leaks	
  in	
  chemical	
  producHon	
  
!  In-­‐store	
  shoe	
  manufacturing	
  
	
  
!  ….	
  and	
  more	
  
7/13/15	
   20	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
Great Visions in 2004, but…
!  Data	
  Warehousing	
  systems	
  built	
  around	
  transacHonal,	
  
structured	
  data	
  
!  Reports	
  made	
  available	
  to	
  business	
  decision	
  makers	
  days	
  to	
  
weeks	
  aler	
  data	
  creaHon	
  
!  No	
  ability	
  to	
  converge	
  mulHple	
  data	
  sources	
  
!  No	
  ability	
  to	
  leverage	
  unstructured	
  data	
  	
  
!  Real	
  Hme?	
  Are	
  you	
  kidding?	
  
CEOs	
  were	
  incapable	
  of	
  turning	
  their	
  visions	
  
into	
  reality	
  
7/13/15	
   21	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
The Point?
For	
  years	
  there	
  has	
  been	
  demand	
  for	
  analyHcs	
  systems	
  
that	
  can	
  converge	
  mulHple	
  data	
  sources	
  and	
  deliver	
  
results	
  in	
  real	
  Hme	
  to:	
  
	
  
!  Business	
  decision	
  makers	
  
!  Healthcare	
  and	
  Public	
  sector	
  
!  …	
  and	
  Consumers	
  
	
  
Now	
  we	
  have	
  the	
  systems	
  	
  
7/13/15	
   22	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
What are the Possibilities?
7/13/15	
   23	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
What if You Could Combine…?
!  RoboHcs	
  
!  Electronic	
  sensors	
  and	
  sensory	
  data	
  (like	
  IoT)	
  
!  Bluetooth	
  communicaHons	
  (mobility)	
  
!  Centralized,	
  real	
  Hme	
  analyHcs	
  	
  
!  Automated	
  programming	
  
!  ConsumerizaHon	
  
!  Open	
  source	
  
7/13/15	
   24	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
Mobotics
7/13/15	
   25	
  
	
  	
  	
  	
  	
  	
  
•  RoboHcs	
  
•  Electronic	
  sensors	
  and	
  sensory	
  
data	
  (like	
  IoT)	
  
•  Bluetooth	
  communicaHons	
  
(mobility)	
  
•  Centralized,	
  real	
  Hme	
  analyHcs	
  	
  
•  Automated	
  programming	
  
•  ConsumerizaHon	
  
•  Open	
  source	
  
MeaowMeaow
The first open-source web-programmable
device and robotics platform.
Designed to be used by common web-site programmers,
allowing device logic to be updated constantly, by simple
programs/web-pages that connect to all other internet data
sources and technologies.
Chris	
  Stakutis	
  
chris.stakutis@	
  
ConcordSoftwareAndExecutiveConsulting.com	
  
978	
  764	
  3488	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
The Point
!  A	
  new	
  scalable,	
  very	
  high	
  performance,	
  low	
  cost	
  Her	
  is	
  
persistent	
  storage	
  is	
  emerging	
  (Tier	
  -­‐1).	
  
!  AdopHon	
  will	
  be	
  accelerated	
  by	
  rapid	
  advances	
  in	
  price	
  for	
  
performance	
  and	
  capacity.	
  
!  It	
  will,	
  converged	
  with	
  other	
  technologies,	
  enable	
  broad	
  
adopHon	
  of	
  real	
  Hme	
  analyHcs.	
  
!  Market	
  potenHal	
  for	
  real	
  Hme	
  analyHcs—and	
  real	
  Hme	
  
compuHng	
  in	
  general—already	
  exists	
  (and	
  has	
  existed	
  for	
  
years).	
  
!  At	
  present,	
  the	
  possibiliHes	
  can	
  only	
  be	
  parHally	
  imaged.	
  
	
  
7/13/15	
   27	
  
©	
  2013	
  Evaluator	
  Group,	
  Inc.	
  
7/13/15	
   28	
  
Does	
  In-­‐Memory	
  CompuHng	
  change	
  everything?	
  
	
  
If	
  so,	
  then	
  this	
  summit	
  marks	
  the	
  beginning	
  of	
  a	
  new	
  era	
  in	
  
compuHng	
  and	
  I	
  hope	
  to	
  see	
  you	
  at	
  the	
  next	
  one.	
  

More Related Content

PDF
IMCSummit 2015 - Day 2 IT Business Track - Drive IMC Efficiency with Flash E...
PDF
IMCSummit 2015 - Day 2 Developer Track - Implementing a Highly Scalable In-Me...
PDF
9/ IBM POWER @ OPEN'16
PDF
IBM POWER8 as an HPC platform
PPT
Fighting the Hidden Costs of Data Storage
PPT
The Need for Speed: Parallel I/O and the New Tick-Tock in Computing
PPTX
Oracle big data appliance and solutions
PPTX
Understanding the IBM Power Systems Advantage
IMCSummit 2015 - Day 2 IT Business Track - Drive IMC Efficiency with Flash E...
IMCSummit 2015 - Day 2 Developer Track - Implementing a Highly Scalable In-Me...
9/ IBM POWER @ OPEN'16
IBM POWER8 as an HPC platform
Fighting the Hidden Costs of Data Storage
The Need for Speed: Parallel I/O and the New Tick-Tock in Computing
Oracle big data appliance and solutions
Understanding the IBM Power Systems Advantage

What's hot (20)

PDF
HPC DAY 2017 | HPE Storage and Data Management for Big Data
PPTX
DataCore At VMworld 2016
PDF
HPC DAY 2017 | The network part in accelerating Machine-Learning and Big-Data
PDF
Modernize Your Oracle Environment with an Agile Data Infrastructure
PPTX
Tendencias Storage
PPTX
Can $0.08 Change your View of Storage?
PPTX
Proactive Threat Detection and Safeguarding of Data for Enhanced Cyber resili...
PPTX
Webinar: The Bifurcation of the Flash Market
PPTX
Webinar: Performance vs. Cost - Solving The HPC Storage Tug-of-War
PPTX
Integrating Hyper-converged Systems with Existing SANs
PDF
Co-Design Architecture for Exascale
PDF
Denver Broncos Case Study
PDF
HPC DAY 2017 | Accelerating tomorrow's HPC and AI workflows with Intel Archit...
PDF
Demartek lenovo s3200_sql_server_evaluation_2016-01
PDF
Scale Out Your Big Data Apps: The Latest on Pivotal GemFire and GemFire XD
PPTX
Oracle Big data at work
PPTX
PostgreSQL as a Strategic Tool
 
PPTX
Big data oracle_introduccion
PDF
Superior Cloud Economics with Power Systems
PDF
Containers and Big Data
HPC DAY 2017 | HPE Storage and Data Management for Big Data
DataCore At VMworld 2016
HPC DAY 2017 | The network part in accelerating Machine-Learning and Big-Data
Modernize Your Oracle Environment with an Agile Data Infrastructure
Tendencias Storage
Can $0.08 Change your View of Storage?
Proactive Threat Detection and Safeguarding of Data for Enhanced Cyber resili...
Webinar: The Bifurcation of the Flash Market
Webinar: Performance vs. Cost - Solving The HPC Storage Tug-of-War
Integrating Hyper-converged Systems with Existing SANs
Co-Design Architecture for Exascale
Denver Broncos Case Study
HPC DAY 2017 | Accelerating tomorrow's HPC and AI workflows with Intel Archit...
Demartek lenovo s3200_sql_server_evaluation_2016-01
Scale Out Your Big Data Apps: The Latest on Pivotal GemFire and GemFire XD
Oracle Big data at work
PostgreSQL as a Strategic Tool
 
Big data oracle_introduccion
Superior Cloud Economics with Power Systems
Containers and Big Data
Ad

Viewers also liked (20)

PDF
IMCSummit 2015 - Day 2 Developer Track - The NVM Revolution
PPSX
3 d xpointtm technology
PDF
Database & Technology 1 _ Martin Power _ Delivering Oracles hight availabilit...
PPTX
PCIe Gen 3.0 Presentation @ 4th FPGA Camp
PDF
Using Storage Class Memory
PDF
Oracle Systems _ Kevin McIsaac _The IT landscape has changed.pdf
PDF
10 Things you need to know abut designing enterprise SSD
PDF
Writing a fast HTTP parser
PDF
PCI Express* based Storage: Data Center NVM Express* Platform Topologies
PPTX
Ceph on rdma
PDF
Mellanox High Performance Networks for Ceph
PPTX
in-memory database system and low latency
PDF
Storage Class Memory: Technology Overview & System Impacts
PDF
Session 8,9 PCI Express
PPTX
Storage class memory
PPTX
5 Things You Need to Know About Enterprise Fl
PPTX
11 lec 11 storage class
PDF
Intel and DataStax: 3D XPoint and NVME Technology Cassandra Storage Comparison
PPTX
In-Memory DataBase
PPTX
Slideshare - PCIe
IMCSummit 2015 - Day 2 Developer Track - The NVM Revolution
3 d xpointtm technology
Database & Technology 1 _ Martin Power _ Delivering Oracles hight availabilit...
PCIe Gen 3.0 Presentation @ 4th FPGA Camp
Using Storage Class Memory
Oracle Systems _ Kevin McIsaac _The IT landscape has changed.pdf
10 Things you need to know abut designing enterprise SSD
Writing a fast HTTP parser
PCI Express* based Storage: Data Center NVM Express* Platform Topologies
Ceph on rdma
Mellanox High Performance Networks for Ceph
in-memory database system and low latency
Storage Class Memory: Technology Overview & System Impacts
Session 8,9 PCI Express
Storage class memory
5 Things You Need to Know About Enterprise Fl
11 lec 11 storage class
Intel and DataStax: 3D XPoint and NVME Technology Cassandra Storage Comparison
In-Memory DataBase
Slideshare - PCIe
Ad

Similar to IMCSummit 2015 - Day 2 Keynote - In-Memory Computing and the Emergence of Tier -1 Storage (20)

PPTX
In-Memory Computing Webcast. Market Predictions 2017
PPTX
Are your ready for in memory applications?
PPTX
5 Things that Make Hadoop a Game Changer
PDF
ManMachine&Mathematics_Arup_Ray_Ext
PPTX
Is the world ready for Exascale Flash
PDF
Data Culture Series - Keynote & Panel - Birmingham - 8th April 2015
PDF
HKG15-The Machine: A new kind of computer- Keynote by Dejan Milojicic
PDF
IDC Technology Spotlight: Big Memory Computing Emerges to Better Enable Dat...
PPT
Solving Big Data Problems
PDF
Accenture hana-in-memory-pov
PDF
Flash memory summit 2015 gary lyng session 301-a
PDF
Hadoop and SQL: Delivery Analytics Across the Organization
PDF
Meta scale kognitio hadoop webinar
PDF
OSDC 2018 | The operational brain: how new Paradigms like Machine Learning ar...
PDF
Future of Data Strategy (ASEAN)
PDF
Memory Database Technology is Driving a New Cycle of Business Innovation
PPT
Big Data = Big Decisions
PDF
Demystifying In-Memory Technologies: Best Uses and Competitive Advantages for...
PDF
50 Shades of SQL
PDF
The New Database Frontier: Harnessing the Cloud
In-Memory Computing Webcast. Market Predictions 2017
Are your ready for in memory applications?
5 Things that Make Hadoop a Game Changer
ManMachine&Mathematics_Arup_Ray_Ext
Is the world ready for Exascale Flash
Data Culture Series - Keynote & Panel - Birmingham - 8th April 2015
HKG15-The Machine: A new kind of computer- Keynote by Dejan Milojicic
IDC Technology Spotlight: Big Memory Computing Emerges to Better Enable Dat...
Solving Big Data Problems
Accenture hana-in-memory-pov
Flash memory summit 2015 gary lyng session 301-a
Hadoop and SQL: Delivery Analytics Across the Organization
Meta scale kognitio hadoop webinar
OSDC 2018 | The operational brain: how new Paradigms like Machine Learning ar...
Future of Data Strategy (ASEAN)
Memory Database Technology is Driving a New Cycle of Business Innovation
Big Data = Big Decisions
Demystifying In-Memory Technologies: Best Uses and Competitive Advantages for...
50 Shades of SQL
The New Database Frontier: Harnessing the Cloud

More from In-Memory Computing Summit (20)

PPTX
IMC Summit 2016 Breakout - Per Minoborg - Work with Multiple Hot Terabytes in...
PPTX
IMC Summit 2016 Breakout - Henning Andersen - Using Lock-free and Wait-free I...
PPTX
IMC Summit 2016 Breakout - Roman Shtykh - Apache Ignite as a Data Processing Hub
PDF
IMC Summit 2016 Breakout - Nikita Shamgunov - Propelling IoT Innovation with ...
PPTX
IMC Summit 2016 Breakout - Matt Coventon - Test Driving Streaming and CEP on ...
PDF
IMC Summit 2016 Innovation - Derek Nelson - PipelineDB: The Streaming-SQL Dat...
PPTX
IMC Summit 2016 Innovation - Dennis Duckworth - Lambda-B-Gone: The In-memory ...
PPTX
IMC Summit 2016 Innovation - Steve Wilkes - Tap Into Your Enterprise – Why Da...
PPTX
IMC Summit 2016 Innovation - Girish Mutreja - Unveiling the X Platform
PPTX
IMC Summit 2016 Breakout - Ken Gibson - The In-Place Working Storage Tier
PPTX
IMC Summit 2016 Breakout - Brian Bulkowski - NVMe, Storage Class Memory and O...
PDF
IMC Summit 2016 Breakout - Yanping Wang - Non-volatile Generic Object Program...
PPTX
IMC Summit 2016 Breakout - Andy Pavlo - What Non-Volatile Memory Means for th...
PPTX
IMC Summit 2016 Breakout - Gordon Patrick - Developments in Persistent Memory
PPTX
IMC Summit 2016 Breakout - Girish Kathalagiri - Decision Making with MLLIB, S...
PPTX
IMC Summit 2016 Breakout - Steve Wikes - Making IMC Enterprise Grade
PPTX
IMC Summit 2016 Breakout - Noah Arliss - The Truth: How to Test Your Distribu...
PPTX
IMC Summit 2016 Breakout - Aleksandar Seovic - The Illusion of Statelessness
PPTX
IMC Summit 2016 Breakout - Girish Mutreja - Extreme Transaction Processing in...
PPTX
IMC Summit 2016 Breakout - Greg Luck - How to Speed Up Your Application Using...
IMC Summit 2016 Breakout - Per Minoborg - Work with Multiple Hot Terabytes in...
IMC Summit 2016 Breakout - Henning Andersen - Using Lock-free and Wait-free I...
IMC Summit 2016 Breakout - Roman Shtykh - Apache Ignite as a Data Processing Hub
IMC Summit 2016 Breakout - Nikita Shamgunov - Propelling IoT Innovation with ...
IMC Summit 2016 Breakout - Matt Coventon - Test Driving Streaming and CEP on ...
IMC Summit 2016 Innovation - Derek Nelson - PipelineDB: The Streaming-SQL Dat...
IMC Summit 2016 Innovation - Dennis Duckworth - Lambda-B-Gone: The In-memory ...
IMC Summit 2016 Innovation - Steve Wilkes - Tap Into Your Enterprise – Why Da...
IMC Summit 2016 Innovation - Girish Mutreja - Unveiling the X Platform
IMC Summit 2016 Breakout - Ken Gibson - The In-Place Working Storage Tier
IMC Summit 2016 Breakout - Brian Bulkowski - NVMe, Storage Class Memory and O...
IMC Summit 2016 Breakout - Yanping Wang - Non-volatile Generic Object Program...
IMC Summit 2016 Breakout - Andy Pavlo - What Non-Volatile Memory Means for th...
IMC Summit 2016 Breakout - Gordon Patrick - Developments in Persistent Memory
IMC Summit 2016 Breakout - Girish Kathalagiri - Decision Making with MLLIB, S...
IMC Summit 2016 Breakout - Steve Wikes - Making IMC Enterprise Grade
IMC Summit 2016 Breakout - Noah Arliss - The Truth: How to Test Your Distribu...
IMC Summit 2016 Breakout - Aleksandar Seovic - The Illusion of Statelessness
IMC Summit 2016 Breakout - Girish Mutreja - Extreme Transaction Processing in...
IMC Summit 2016 Breakout - Greg Luck - How to Speed Up Your Application Using...

Recently uploaded (20)

PDF
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
PDF
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
PDF
Getting Started with Data Integration: FME Form 101
PDF
Heart disease approach using modified random forest and particle swarm optimi...
PPTX
OMC Textile Division Presentation 2021.pptx
PPTX
Tartificialntelligence_presentation.pptx
PDF
Assigned Numbers - 2025 - Bluetooth® Document
PPTX
A Presentation on Artificial Intelligence
PDF
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
PDF
Accuracy of neural networks in brain wave diagnosis of schizophrenia
PPTX
Chapter 5: Probability Theory and Statistics
PDF
WOOl fibre morphology and structure.pdf for textiles
PPTX
Programs and apps: productivity, graphics, security and other tools
PDF
project resource management chapter-09.pdf
PDF
Zenith AI: Advanced Artificial Intelligence
PDF
August Patch Tuesday
PDF
Web App vs Mobile App What Should You Build First.pdf
PDF
Encapsulation theory and applications.pdf
PDF
A comparative study of natural language inference in Swahili using monolingua...
PDF
gpt5_lecture_notes_comprehensive_20250812015547.pdf
From MVP to Full-Scale Product A Startup’s Software Journey.pdf
Transform Your ITIL® 4 & ITSM Strategy with AI in 2025.pdf
Getting Started with Data Integration: FME Form 101
Heart disease approach using modified random forest and particle swarm optimi...
OMC Textile Division Presentation 2021.pptx
Tartificialntelligence_presentation.pptx
Assigned Numbers - 2025 - Bluetooth® Document
A Presentation on Artificial Intelligence
ENT215_Completing-a-large-scale-migration-and-modernization-with-AWS.pdf
Accuracy of neural networks in brain wave diagnosis of schizophrenia
Chapter 5: Probability Theory and Statistics
WOOl fibre morphology and structure.pdf for textiles
Programs and apps: productivity, graphics, security and other tools
project resource management chapter-09.pdf
Zenith AI: Advanced Artificial Intelligence
August Patch Tuesday
Web App vs Mobile App What Should You Build First.pdf
Encapsulation theory and applications.pdf
A comparative study of natural language inference in Swahili using monolingua...
gpt5_lecture_notes_comprehensive_20250812015547.pdf

IMCSummit 2015 - Day 2 Keynote - In-Memory Computing and the Emergence of Tier -1 Storage

  • 1. ©  2013  Evaluator  Group,  Inc.   In-­‐Memory  Compu.ng  and  the  Emergence  of  Tier  -­‐1  Storage   John  Webster   Senior  Partner   [email protected]   7/13/15   1  
  • 2. ©  2012  Evaluator  Group,  Inc.   Overview of Evaluator Group ! Since  1997  delivering  comprehensive  and  objecHve  analysis  of   storage  systems  and  related  issues   —  A  team  of  engineers,  systems  performance  experts  and  go-­‐to-­‐market   specialists   —  Expert  analysis  with  detailed  informaHon  and  unbiased  opinions   ! PracHcal,  business-­‐oriented,  technical  perspecHves   Slide 2
  • 3. ©  2012  Evaluator  Group,  Inc.   Research Coverage 7/13/15   3   InformaHon   Management  &   Data  Storage   Virtualiza.on   Data   Analy.cs   Performance   Solid  State   Hyper-­‐ Converged   Cloud   Services  &   Object   SDS  &   Virtualiza.on   Data   Protec.on  &   Archive  
  • 4. ©  2012  Evaluator  Group,  Inc.   A Few of Our Customers Slide   4  
  • 5. ©  2013  Evaluator  Group,  Inc.   What’s the Point? !  A  new  scalable,  very  high  performance,  low  cost  Her  is   persistent  storage  is  emerging  (Tier  -­‐1).   !  AdopHon  will  be  accelerated  by  rapid  advances  in  price  for   performance  and  capacity.   !  It  will,  converged  with  other  technologies,  enable  broad   adopHon  of  real  Hme  analyHcs.   !  Market  potenHal  for  real  Hme  analyHcs—and  real  Hme   compuHng  in  general—already  exists  (and  has  existed  for   years).   !  At  present,  the  possibiliHes  can  only  be  parHally  imaged.     7/13/15   5  
  • 6. ©  2013  Evaluator  Group,  Inc.   What is Tier -1? 7/13/15   6  
  • 7. ©  2015  Evaluator  Group,  Inc.   SSD (Tier 0) High-Performance Disks (Tier 1) High-Capacity Disks (Tier 2) External Tier 3 Tape Optical DataMovement Storage System Tiering “Within the Box” and External 7/13/15   7     Response  Time     µs     ms             ms             seconds   SSD  =  Solid  State   Device  
  • 8. ©  2015  Evaluator  Group,  Inc.   Persistent In- Memory SSD High-Capacity Disks Tier -1 Data Persistence In-Memory 7/13/15   8   Persistent In- Memory SSD Capacity SSD .....   _ SSD  =  Solid  State   Device  
  • 9. ©  2015  Evaluator  Group,  Inc.   What Does the Future for SSD Look Like? 7/13/15   9  
  • 10. ©  2015  Evaluator  Group,  Inc.   Why Enterprise IT Likes SSD !  Elongate  tech  refresh  cycles     !  Increased  efficiency/reduced  cost  of  virtualized   servers   !  DeduplicaHon  yields  gains  in  price/GB  vs.  disk   !  Lower  Opex  for  machine  room  environmentals   !  ApplicaHon  performance  (turn  muts  into  show  dogs)   7/13/15   10  
  • 11. ©  2015  Evaluator  Group,  Inc.   Why Business Likes SSD 7/13/15   11  
  • 12. ©  2015  Evaluator  Group,  Inc.   Multiple SSD Technologies Advancing in Parallel Non-­‐volaHle  memory  (NVM)  technologies  under  current   development   !  NAND  flash   !  MRAM   !  Memristor   !  ResisHve  RAM   !  PCM   !  Nano   12  7/13/15  
  • 13. ©  2015  Evaluator  Group,  Inc.   Cost of Disk vs Flash 7/13/15   13  
  • 14. ©  2015  Evaluator  Group,  Inc.   Moore’s Law for DRAM 7/13/15   14  
  • 15. ©  2015  Evaluator  Group,  Inc.   NVMe Performance 7/13/15   15  
  • 16. ©  2015  Evaluator  Group,  Inc.   Bottom Line Increasing  demand  +  increasing  compe..on  +  increasing  produc.on   for  SSD   =   Con.nual  improvements  in  SSD  price  for  performance  and  scale   =   Con.nued  cost  reduc.ons  and  performance  enhancements  for  In-­‐ Memory  Compu.ng   =   Accelera.ng  advancements  in  real  .me   compu.ng   7/13/15   16  
  • 17. ©  2015  Evaluator  Group,  Inc.   Where is the Demand for Real Time Computing? 7/13/15   17  
  • 18. ©  2013  Evaluator  Group,  Inc.   Enabling Converged, Real-time Analytics 7/13/15   18   The  confluence  of  two   major  compuHng   trajectories:     !  ConHnued  improvements  in   memory  technology  that   began  with  the  first  digital   computers     !  Advancement  of  MPP   compuHng  that  began  in  the   early  1980's  
  • 19. ©  2013  Evaluator  Group,  Inc.   Visions of Real Time Analytics in 2004 7/13/15   19   !  AnalyHc-­‐based  applicaHons   based  on  convergence  of   mulHple  data  sources-­‐ potenHally  in  real  Hme   !  Interviewed  50  CEOs,  CIOs   and  Technologists  from   verHcal  industry  segments   !  Visions  were  of  leveraging   disparate  and  pervasive   data  sources  (mobile,   sensor,  database)  in  real   Hme.  
  • 20. ©  2013  Evaluator  Group,  Inc.   CEO Real Time Visions 2004 !  RFID  converged  with  paHent  data  in  to  prevent  accidentally   inducing  adverse  reacHons  to  drugs  in  a  hospital  sejng   !  MulHple  sensory  data  inputs  to  monitor  and  control  chemical   leaks  in  chemical  producHon   !  In-­‐store  shoe  manufacturing     !  ….  and  more   7/13/15   20  
  • 21. ©  2013  Evaluator  Group,  Inc.   Great Visions in 2004, but… !  Data  Warehousing  systems  built  around  transacHonal,   structured  data   !  Reports  made  available  to  business  decision  makers  days  to   weeks  aler  data  creaHon   !  No  ability  to  converge  mulHple  data  sources   !  No  ability  to  leverage  unstructured  data     !  Real  Hme?  Are  you  kidding?   CEOs  were  incapable  of  turning  their  visions   into  reality   7/13/15   21  
  • 22. ©  2013  Evaluator  Group,  Inc.   The Point? For  years  there  has  been  demand  for  analyHcs  systems   that  can  converge  mulHple  data  sources  and  deliver   results  in  real  Hme  to:     !  Business  decision  makers   !  Healthcare  and  Public  sector   !  …  and  Consumers     Now  we  have  the  systems     7/13/15   22  
  • 23. ©  2013  Evaluator  Group,  Inc.   What are the Possibilities? 7/13/15   23  
  • 24. ©  2013  Evaluator  Group,  Inc.   What if You Could Combine…? !  RoboHcs   !  Electronic  sensors  and  sensory  data  (like  IoT)   !  Bluetooth  communicaHons  (mobility)   !  Centralized,  real  Hme  analyHcs     !  Automated  programming   !  ConsumerizaHon   !  Open  source   7/13/15   24  
  • 25. ©  2013  Evaluator  Group,  Inc.   Mobotics 7/13/15   25               •  RoboHcs   •  Electronic  sensors  and  sensory   data  (like  IoT)   •  Bluetooth  communicaHons   (mobility)   •  Centralized,  real  Hme  analyHcs     •  Automated  programming   •  ConsumerizaHon   •  Open  source  
  • 26. MeaowMeaow The first open-source web-programmable device and robotics platform. Designed to be used by common web-site programmers, allowing device logic to be updated constantly, by simple programs/web-pages that connect to all other internet data sources and technologies. Chris  Stakutis   chris.stakutis@   ConcordSoftwareAndExecutiveConsulting.com   978  764  3488  
  • 27. ©  2013  Evaluator  Group,  Inc.   The Point !  A  new  scalable,  very  high  performance,  low  cost  Her  is   persistent  storage  is  emerging  (Tier  -­‐1).   !  AdopHon  will  be  accelerated  by  rapid  advances  in  price  for   performance  and  capacity.   !  It  will,  converged  with  other  technologies,  enable  broad   adopHon  of  real  Hme  analyHcs.   !  Market  potenHal  for  real  Hme  analyHcs—and  real  Hme   compuHng  in  general—already  exists  (and  has  existed  for   years).   !  At  present,  the  possibiliHes  can  only  be  parHally  imaged.     7/13/15   27  
  • 28. ©  2013  Evaluator  Group,  Inc.   7/13/15   28   Does  In-­‐Memory  CompuHng  change  everything?     If  so,  then  this  summit  marks  the  beginning  of  a  new  era  in   compuHng  and  I  hope  to  see  you  at  the  next  one.