IBM Cloud Object Storage System, presented Oct 16, 2017 at IBM Systems Technical University in New Orleans, LA. This covers the object storage from IBM from the acquisition of Cleversafe, formerly known as DSnet product.
1. IBM Power Systems
and IBM Storage
Technical University
Introduction to IBMÂŽ
Cloud Object Storage
Systemâ˘
(powered by Cleversafe)
Tony Pearson
Master Inventor and Senior IT Architect,
IBM Corporation
2. Abstract
This session explains why object storage is becoming so
popular and how it differs from block or file-level storage.
The speaker will provide an overview of IBM Cloud Object
Storage System and its clever, highly scalable, highly reliable
use of Erasure Coding to reduce storage costs by up to 70
percent from traditional disk arrays
2
3. This week with Tony Pearson
Day Time Topic
Monday 10:15am
Business Continuity â The seven tiers of business
continuity and disaster recovery
1:45pm IBMâs Cloud Storage Options
4:30pm
Introduction to IBM Cloud Object Storage System
and its Applications (powered by Cleversafe)
Tuesday 10:15am
The Pendulum Swings Back â Understanding
Converged and Hyperconverged Environments
11:30am
New generation of storage tiering: Simpler
management, Lower costs and Increased
performance
3:15pm
Introduction to IBM Cloud Object Storage System
and its Applications (powered by Cleversafe)
Wednesday
9:00am IBM Spectrum Scale for File and Object Storage
3
5. Clients are facing explosive growth in Unstructured Data,
which is exactly why Object Storage is so critical
*Exabytes
0
20
40
60
80
100
120
2009 2010 2011 2012 2013 2014 2015 2016 2017
Unstructured Data
Structured Data
Source: IDC
Unstructured
data growth of
60â80%
per year
creates
Web-scale
storage needs
*1 exabyte = 1,000 petabytes =1 million terabytes = 1 billion
gigabytes
Problem - Legacy Storage Designed for
Transactional Databases, Not Unstructured Data
5
6. How is Object Storage Different?
Block and File Storage
â Decide where to put it
⢠For block, which array/volume/LUN
⢠For file, which filer/subdirectory
â Remember where it is to get it back
â Donât let anyone else move it
â Read/Write records, append data
â Limits on LUN size, number of files
Object Storage
â Provide data over to the Object
storage
⢠Get âclaim stubâ reference locator
â Use or share âclaim stubâ to access
data HTTP, Openstack Swift, S3
â Get/Put/Delete object in its entirety
â Effectively âunlimitedâ scalability
6
7. Object Store for Unstructured data
Hot Data
High-IOPS and Low-Latency
All-Flash and Hybrid Flash/Disk
Information Lifecycle Management
(ILM)
Structured data / Random-Access
Active logs and traces
Virtual Machines and VDI
Single-Tenancy
Static and Stable data
Backups, Files, Archives
Seismic, Research, Telemetry, HPC
Video, Animation, Body Cams
Photos, Images, CAD/CAM, GIS
Music, Audio
Genomic, Medical Images
Multi-tenancy
Object Store provides a
⢠Secure
⢠Reliable
⢠Scalable
⢠Cost Effective
Platform For Unstructured data
Object Store
is not designed for
⢠High IOPS workflows
⢠Transaction
Processing
⢠Inherent ILM
7
8. Object Storage is Simpler for Application Development
POSIX â over 60 commands NFS â over 30 commands
HEAD
⢠Read metadata
GET
⢠Read content and
metadata
PUT
⢠Write content and
metadata
DELETE
⢠Remove object or
empty container
POST
⢠Update metadata
Object â 5 commands
SMB â over 80 commands
8
9. Object storage is dominant in cloud
0
50
100
150
200
250
2012 2013 2014 2015 2016 2017 2018
File-based software, appliances, and gateways
Object-based software and appliances (includes self-built/SaaS/cloud NAS)
IDC: Worldwide file- and object-based storage
2014â2018 Forecast (EB)
9
10. Why Object Storage?
It enables my
business and reduces
costs!
⢠Handles 80% of my
data
⢠Scales easily, growth
does not impact my
business plan
⢠Less expensive than
traditional flash and
disk arrays
It is easy to deploy and
manage!
⢠Easily provision to users
and teams
⢠Location-independent,
Flat namespace
⢠Software-Defined or
Pre-built systems
⢠Ethernet-based and
Cloud friendly
⢠Online backup/archive
alternative to tape and
virtual tape libraries
It is great for
developing software!
⢠GET/PUT interface
reduces coding effort
⢠Robust metadata for
search of documents,
images, and videos
⢠Can be used for
Analytics, IoT,
Mobile, Web and
Cloud Applications
IT
Director
Software
Developer
Storage
Administrator
10
12. Storage Positioning â Filling a Gap
Unified file and
object storage.
Optimized for high
performance, across
flash and disk
Flash 15K
Unified file and object
storage on tape
Information Lifecycle
Management (ILM) across tiers
HighestPerformance
Lowest cost
Tape
IBM was looking to offer easy to
manage, scalable disk-based object
storage for unstructured data
⢠Moderate performance
⢠Moderate cost
10K 7200 rpm
12
13. Data Protection Schemes
Tolerate 1 drive failure Tolerate 2 drive failures Tolerate âMâ failures
RAID-1 / RAID-10
K pieces 2 x K slices
RAID-5
K pieces K + 1 slices
2.0X
1.2X
3.0X
1.5X
1.7XTriplication
K pieces 3 x K slices
RAID-6
K pieces K + 2 slices
Erasure Coding
K pieces K+M = N
slices
13
14. Data Growth at Petabyte (PB) Scale
PB of data
3 to 5x
Data Protection
RAID, Mirrors,
Replication, Tape
Data Protection
High Availability & Disaster Recovery
Geo-Distribution & Erasure Coding
Infrastructure
Proprietary, specialized
hardware, multiple systems
Operations
More than 1 FTE per PB
Maintenance outages
Infrastructure
Software Defined,
Commodity Hardware,
Single System
Operations
Less than 1 FTE per 6 PB
Single system, Secure
Self-healing
1.7 x
60% Less
Hardware &
Rack space
Traditional
Approach
IBM Cloud
Object Storage
Approach
70%
Lower
TCO
14
15. IBM Cloud Object Storage System economics beat
legacy NAS storage and Amazon Web Services (AWS)
$8,400
$4,210
$1,613
$1,053
Legacy NAS DR protected
Legacy NAS single copy
IBM Cloud Object + NAS
gateway
dsNet object protectedIBM Cloud Object
IBM Cloud Object vs NAS
Cost: 80% lower
$0
$2,000,000
$4,000,000
$6,000,000
$8,000,000
$10,000,000
480 TB 960 TB 1920 TB 3840 TB
S3
dsNet
Cost: 10 to 60% lower
IBM Cloud Object vs AWS S3
IBM Cloud Object
Amazon S3
15
16. IBM Cloud Object Storage System â Terminology
HTTP
Application
Server
IBM COS
Manger
Accesser
Slicestor
Device
Set
Accesser PoolVault
Load
Balancer
Site A Site B Site C
End user
Global Namespace
IBM Cloud Object Storage
System organizes objects into
âVaultsâ and âContainersâ
Storage Pools can hold multiple
vaults
Vaults can have millions of
containers
OpenStack Swift
S3 Protocol
16
17. The Math Behind Reed-Solomon Erasure Coding
âKâ variables of input data: a, b, c, d, e
Example: K=5 M=3
K+M= 8
Here we create âK+Mâ
equations, adding and
subtracting by different
co-factors
Results in âK+Mâ slices
that can be geographically
dispersed
We can tolerate losing up
to âMâ slices of data, and
still solve for the original
âKâ pieces of data.
17
18. Information Dispersal Algorithm (IDA)
01 02 03 04 05 06 07 08 09 10 11 12
A B C D
E F G
IDA is represented as Width/Read/Write
This example is 12/7/9
Original data is received by
IBM COS Accesser and
chopped up into 7 pieces
(Read threshold)
Erasure Coding:
7 variables in,
12 equation results out
(Width)
12 slices written in parallel
to IBM COS Slicestor devices
A B C D
E F G
IBM SmartWriteâ˘
At least 9 slices (Write Threshold) must be written
for the I/O to be considered complete, rest will be
written later. IBM COS periodically scans and
rebuilds slices as needed
IBM SmartReadâ˘
To read the data back,
only 7 slices are used
to solve 7 equations
(Read threshold)
Original data is
re-assembled
18
19. Dedicated
(Private)
IBM Cloud Object Storage System â
Single-site, Two-site and Multi-Site
Single Site
Better performance,
when site-tolerance
not a factor, better
than traditional
RAID-5 / RAID-6
Two Site Vault Mirroring
Allows customers to leverage existing two-site
infrastructure. Provides concurrent reads and
writes despite communication disruption
between locations. Local data better than
traditional RAID-5 / RAID-6
Geographically Dispersed
Three to Nine Sites
Consider adding IBM Cloud or leverage existing
datacenter locations to provide a broader distribution
of data for higher availability, site-tolerance and
scalability
19
20. Two Dispersal Modes
Standard Dispersal
⢠One or more Accessers
⢠Each slice on its own
Slicestor
Concentrated Dispersal (CD Mode)
⢠Entry Level System (1-3 sites)
⢠Accessers can be embedded into
Slicestor (Optional)
⢠Each slice stored on separate
HDD within Slicestor for availability
01 02 03
04 05 06
07 08 09
10 11 12
01 02
03 04
05 06
07 08
09 10
11 12
20
21. Scalability
Site 1 Site 2 Site 3
S3 Compatible API Openstack Swift
Compatible API
Simple Object API
Slicestor
Software
Accesser
Software
Scalability â Scale performance and/or
capacity at any time with no downtime to
operations
Need more Performance?
Add more Accesser nodes
Need more Capacity?
Add more disks to existing Slicestor nodes,
add more device sets to existing storage
pools, or add more storage pools
â All nodes in single device set must have
same amount of storage
⢠New device sets can have different size
drives than existing drives sets
â All device sets in a storage pool must
have the same IDA
⢠Standard and Concentrated IDA can be
intermixed in same storage pool
â Different storage pools can have
different IDA settings
Device
Set 1
Device
Set 2
Device
Set 3
21
22. Encryption Options for IBM Cloud Object Storage
A
A B C D
E F G
Data
A
A#
AONT
Data
SecureSliceâ˘
System-generated key
Used to encrypt data
Data
A
#
Hash taken of
encrypted data
Key wrapped with hash makes
All-or-Nothing Transform (AONT)
Information
Dispersal
Algorithm (IDA)
User sends data
Data
B
Data
A
A
B
Server-Side Encryption with Customer key
System-generated key used to encrypt data
Key wrapped with customer-supplied key
User sends data and key
SSE-C
Data
CData
C UMK
Application-managed key
User sends encrypted data
Encryption disabled in IBM COS
22
23. Compliance-Enabled Vault
Standard
Vault
Compliance
Enabled
Vault
Government Regulations specify that
âelectronic storage must preserve the
records exclusively in a non-erasable,
non-rewriteable, formatâ (SEC 17a-4)
Various Terms have been used:
⢠Write Once Read Many (WORM)
for tape and optical media
⢠Fixed-Content or Content-
Addressable for disk-based storage
⢠Immutable for file systems
⢠Non-erasable, Non-rewriteable
(NENR) to cover all of the above
Compliance-Enabled Vaults prevent
objects from being deleted or modified
for a specified retention period
Vaults can co-exist
in same
storage pool
23
24. IBM Cloud Object Storage System â
Software, Pre-Built Systems or Cloud Services
Software
⢠ClevOS â IBM Cloud
Object Storage System
software packaged with
Debian Linux OS
⢠Software-defined,
hardware-aware model for
flexibility of x86 platform
choice
⢠Can be deployed on
qualified vendor hardware
Pre-built Systems
⢠Fully integrated appliance
models for easy deployment
and support
⢠On-premises object storage
solution
Cloud Services
⢠IBM Cloud Object
Storage System as a
service for customers
that want security and
controlled performance
24
25. IBM Spectrum Storage Suite
Bundle license for
flash, disk, tape
and storage-rich
servers
Per TiB
IBM Spectrum
Storage Suite
Control Protect
Archive
Virtualize Accelerate
Scale
IBM Cloud Object Storage
Iâve deployed Spectrum Storage Suite. Now
what?
1. Well, since you already own entitlement to
all capabilities in the Spectrum Storage
suite, you are free to download any of the
software you like, including IBM Cloud
Object Storage!
2. To help facilitate quick adoption of
additional use cases, Spectrum Storage
Suite licensing offers the ability to
perform extended tests in an evaluation
sandbox proving ground without
additional charge. Tebibytes (TiB) in this
sandbox are not counted.
So go ahead, experiment with your next use
case. Prove it, become familiar with it, pay for
it only when itâs deployed for productive use.
25
26. IBM Cloud Object Storage System â Deployment
Options
Infrastructure
Dedicated
(Private)
Public
Local
(Private)
On-premises
Off-premises
Cloud
Infrastructure
Off-premises
(IBM Cloud)
Public
⢠Standard regional
⢠Nearline regional
⢠Nearline geo-
dispersed
Consumed
pricing
Dedicated
⢠IBM Managed
⢠Client managed
⢠Hybrid / Mixed
Allocated
pricing
Client
infrastructure
On-premises
Locally managed
⢠Software
⢠Pre-built systems
Entry Level
System as
small as 72 TB
26
27. Pre-Built Systems for IBM Cloud Object Storage
System
IBM COS Manager 2105 / 3105
Accesser 2100 / 3105 / 4105
Slicestor 2212
Slicestor 2448
Health and performance monitoring
GUI and API access
Configuration and security
Zero downtime upgrades
12 drives @ 4, 6, 8 or 10 TB
48 to 120 TB Nearline HDD
in 2U rack space
16, 32 or 48 drives @ 4, 6, 8 or 10 TB
64 to 480 TB Nearline HDD
in 4U rack space
Slices, Disperses and Retrieves data
27
28. VersaStack for IBM Cloud Object Storage System
Cisco UCS S3260 Storage Server
Dual nodes in 4U space
56 hot-swappable 3.5â LFF HDD
4, 6, 8, or 10 TB 7200-rpm NL-SAS
(28 drives per COS Slicestor)
Cisco UCS 6300 Fabric Interconnect
Low-latency, lossless 10 and 40 GbE
Cisco UCS 220 M4 servers
1U with 36 cores, 24 DDR4 memory DIMMs
For COS Manager and Accessers
Cisco Validated Design (CVD)
784 to 1960 TB
Usable Capacity
28
29. IBM Cloud Object Storage public cloud services
IBM Cloud Object Storage Flex
DataAccessPattern
IBM Cloud Object Storage
Standard
IBM Cloud Object
Storage
Vault
Variable data access
pattern / Mixed
workloads
Predictable
data access
pattern /
workload
Frequent
data access
Less frequent
data access
Minimal
data access
IBM Cloud Object
Storage
Cold Vault
⢠High Availability â Cross Regional USA
⢠Low Latency â Regional (Dallas)
29
31. A broad spectrum of strategies to build, buy, or assemble an end to end solution
Integrations, Connectors,
Gateways
Ready to RunCustom Applications
Off-PremHybridOn-Prem
and/or and/or
How to Consume IBM Cloud Object Storage
31
32. Large
multimedia files
Video, Audio,
Images
Store
telemetric,
instrumentation
and sensor
data
Web generated
and social
media data
Backup, Active
archive, DR,
inactive and
compliance
data
Security audits,
Log files and
Analytics
Collaboration
and content
distribution
Global file
services,
mobile access,
and StaaS
Genomics,
HPC, Geo-
spatiial
8
7
6
5
4
1
2
3
Object Storage Typical Use Cases Storage
Services
32
33. Market Industry Verticals
Secure
customer
trust and
business
compliance.
Financial
Services &
Insurance
Incorporate
into IoT
processes,
analytics
Industrial,
Distribution
Manage the
data essential
to serving the
good of the
public.
Public Sector,
Government
Scale your
market
offering
without
worry.
Telco &
Service
Providers
Create and
Deliver
Content on
demand.
Media &
Entertainment
Heath Care &
Life Sciences
Put medical
progress
before
everything
else.
Every industry does backup, all have some
amount of static-and-stable unstructured data,
most need file or content archiveâŚ
33
34. Client Referenceâ Web Based Photo Sharing
Users upload photo
and video content via
web based application
Photo and video objects are sent to
Cleversafe via REST based protocols
Metadata is
captured and
stored
⢠Scale â 150 petabytes and growing: more than 50 Billion images stored
⢠Manageability â 3 Administrators manage entire environment
⢠Security â 50,000+ uploads per minute with zero touch security
⢠Always-on availability â SLA of 100% download on demand â even during
California to Nevada datacenter move
⢠Economics â Operating costs reduced by more than 70%
⢠Key decision makers â Technical team backed by financial cost cutting
mandates
34
35. IBM Spectrum Protect â
Cloud and Object Store Capabilities
35
Client nodes
Off-premises:
⢠IBM Bluemix Cloud
⢠Amazon Web Services
⢠Microsoft Azure
On-premises
IBM Spectrum
Protect Server
⢠IBM Cloud Object Storage System
(using S3-compatible API)
36. Cost and Performance Comparison
36
0
5
10
15
20
25
IBM Tape $5/TB IBM COS $76/TB IBM ESS $110/TB
Throughput Rates
Backup TB/hr Restore TB/hr
8 drives
1 drive 30 sessions
Backup and Restore rates vary depending on
network connection, workload, file size and
other environmental factors
Costs may vary, based on discounted street prices of LTO-8, IBM COS, and
ESS GLxS models as configured for âLargeâ IBM Spectrum Protect blueprint on AIX
8 file systems
37. IBM Spectrum Storage and IBM Cloud Object
Unified file and object
storage. Optimized for
high performance, across
flash, disk and object
store
Flash
Object
Store
15K
Object storage on disk
( File, backup and archive interfaces
available through variety of options )
IBM Cloud
Amazon Web Services
Microsoft Azure
Swift S3 emulation
OpenStack Swift
Unified file and object
storage on tape
Transparent Cloud Tiering
Information Lifecycle
Management (ILM) across tiers
HighestPerformance
Lowest cost
Tape10K 7200
37
38. Summary
Object Storage scales easily to
handle your unstructured data, easy to
manage, ideal for software
development
IBM is the #1 leading vendor of
Object Storage, with over 400 patents,
and several clients >100PB
IBM Cloud Object Storage can be
used on-premises, hybrid, or in the
Cloud
Integrates with the rest of the IBM
Spectrum Storage suite
38
42. IBM Tucson Executive Briefing Center
⢠Tucson, Arizona is
home for storage
hardware and
software design and
development
⢠IBM Tucson
Executive Briefing
Center offers:
â Technology
briefings
â Product
demonstrations
â Solution workshops
⢠Take a video tour!
â http://youtu.be/CXr
poCZAazg
https://www.ibm.com/it-infrastructure/services/client-centers
ccenter@us.ibm.com
42
43. About the Speaker
43
Tony Pearson is a Master Inventor and Senior IT Architect for the IBM Storage product line. Tony joined IBM Corporation in
1986 in Tucson, Arizona, USA, and has lived there ever since. In his current role, Tony presents briefings on storage topics
covering the entire IBM Storage product line, IBM Spectrum Storage software products, and topics related to Cloud Computing,
Analytics and Cognitive Solutions. He interacts with clients, speaks at conferences and events, and leads client workshops to
help clients with strategic planning for IBMâs integrated set of storage management software, hardware, and virtualization
solutions.
Tony writes the âInside System Storageâ blog, which is read by thousands of clients, IBM sales reps and IBM Business Partners
every week. This blog was rated one of the top 10 blogs for the IT storage industry by âNetworking Worldâ magazine, and #1
most read IBM blog on IBMâs developerWorks. The blog has been published in series of books, Inside System Storage: Volume
I through V.
Over the past years, Tony has worked in development, marketing and consulting for various storage hardware and software
products. Tony has a Bachelor of Science degree in Software Engineering, and a Master of Science degree in Electrical
Engineering, both from the University of Arizona. . Tony is an inventor or co-inventor of 19 patents in the field of electronic data
storage.
9000 S. Rita Road
Bldg 9032 Floor 1
Tucson, AZ 85744
+1 520-799-4309 (Office)
tpearson@us.ibm.com
Tony Pearson
Master Inventor
Senior IT Architect
IBM Storage
45. Notice and disclaimers
⢠Copyright Š 2017 by International Business Machines Corporation (IBM). No part of this document may be reproduced or transmitted
in any form without written permission from IBM.
⢠U.S. Government Users Restricted Rights â use, duplication or disclosure restricted by GSA ADP Schedule Contract with
IBM.
⢠Information in these presentations (including information relating to products that have not yet been announced by IBM) has been
reviewed for accuracy as of the date of initial publication and could include unintentional technical or typographical errors. IBM shall
have no responsibility to update this information. This document is distributed âas isâ without any warranty, either express or
implied. In no event shall IBM be liable for any damage arising from the use of this information, including but not limited to,
loss of data, business interruption, loss of profit or loss of opportunity. IBM products and services are warranted according to
the terms and conditions of the agreements under which they are provided.
⢠IBM products are manufactured from new parts or new and used parts. In some cases, a product may not be new and may have
been previously installed. Regardless, our warranty terms apply.â
⢠Any statements regarding IBM's future direction, intent or product plans are subject to change or withdrawal without notice.
⢠Performance data contained herein was generally obtained in a controlled, isolated environments. Customer examples are presented
as illustrations of how those customers have used IBM products and the results they may have achieved. Actual performance, cost,
savings or other results in other operating environments may vary.
⢠References in this document to IBM products, programs, or services does not imply that IBM intends to make such products,
programs or services available in all countries in which IBM operates or does business.
⢠Workshops, sessions and associated materials may have been prepared by independent session speakers, and do not necessarily
reflect the views of IBM. All materials and discussions are provided for informational purposes only, and are neither intended to, nor
shall constitute legal or other guidance or advice to any individual participant or their specific situation.
⢠It is the customerâs responsibility to insure its own compliance with legal requirements and to obtain advice of competent legal
counsel as to the identification and interpretation of any relevant laws and regulatory requirements that may affect the customerâs
business and any actions the customer may need to take to comply with such laws. IBM does not provide legal advice or represent or
warrant that its services or products will ensure that the customer is in compliance with any law.
45
46. Notice and disclaimers continued
Information concerning non-IBM products was obtained from the
suppliers of those products, their published announcements or
other publicly available sources. IBM has not tested those
products in connection with this publication and cannot confirm
the accuracy of performance, compatibility or any other claims
related to non-IBM products. Questions on the capabilities of
non-IBM products should be addressed to the suppliers of those
products. IBM does not warrant the quality of any third-party
products, or the ability of any such third-party products to
interoperate with IBMâs products. IBM expressly disclaims all
warranties, expressed or implied, including but not limited
to, the implied warranties of merchantability and fitness for
a particular, purpose.
The provision of the information contained herein is not intended
to, and does not, grant any right or license under any IBM
patents, copyrights, trademarks or other intellectual
property right.
IBM, the IBM logo, ibm.com, AIX, BigInsights, Bluemix, CICS,
Easy Tier, FlashCopy, FlashSystem, GDPS, GPFS,
Guardium, HyperSwap, IBM Cloud Managed Services, IBM
Elastic Storage, IBM FlashCore, IBM FlashSystem, IBM
MobileFirst, IBM Power Systems, IBM PureSystems, IBM
Spectrum, IBM Spectrum Accelerate, IBM Spectrum Archive,
IBM Spectrum Control, IBM Spectrum Protect, IBM Spectrum
Scale, IBM Spectrum Storage, IBM Spectrum Virtualize, IBM
Watson, IBM z Systems, IBM z13, IMS, InfoSphere, Linear
Tape File System, OMEGAMON, OpenPower, Parallel
Sysplex, Power, POWER, POWER4, POWER7, POWER8,
Power Series, Power Systems, Power Systems Software,
PowerHA, PowerLinux, PowerVM, PureApplica- tion, RACF,
Real-time Compression, Redbooks, RMF, SPSS, Storwize,
Symphony, SystemMirror, System Storage, Tivoli,
WebSphere, XIV, z Systems, z/OS, z/VM, z/VSE, zEnterprise
and zSecure are trademarks of International Business
Machines Corporation, registered in many jurisdictions
worldwide. Other product and service names might
be trademarks of IBM or other companies. A current list of
IBM trademarks is available on the Web at "Copyright and
trademark information" at:
www.ibm.com/legal/copytrade.shtml.
Linux is a registered trademark of Linus Torvalds in the United
States, other countries, or both. Java and all Java-based
trademarks and logos are trademarks or registered
trademarks of Oracle and/or its affiliates.
46