S ss0886 pendulum-swings-edge2015-v3
- 1. © Copyright IBM Corporation 2015
Technical University/Symposia materials may not be reproduced in whole or in part without the prior written permission of IBM.
sSS0886
The Pendulum Swings Back –
Understanding Converged and
Hyperconverged Environments
Tony Pearson
Master Inventor and Senior IT Specialist
IBM Corporation
- 2. © Copyright IBM Corporation 2015
Abstract
In the early days of IT,
storage was internal to its
server, over time, storage
outgrew its container, and
we started have externally
attached storage, and
benefits like RAID and
clustered servers for high
availability. Then, SANs,
LANs and WANs took the
main stage, allowing for
greater connectivity and
distance.
But now, it seems the
pendulum is swinging back
with converged and
hyperconverged systems.
This session will provide the
motivations, advantages
and disadvantages of these
new configurations.
1
- 3. © Copyright IBM Corporation 2015
This week with Tony Pearson
2
Day Time Topic
Monday 10:30am Software Defined Storage -- Why? What? How? (repeats Tuesday)
03:00pm IBM's Cloud Storage Options (repeats Wednesday)
04:30pm Data Footprint Reduction – Understanding IBM Storage Efficiency Options
Tuesday 10:30am Software Defined Storage -- Why? What? How?
12:30pm What Is Big Data? Architectures and Practical Use Cases
01:45pm IBM Smarter Storage Strategy (repeats Wednesday)
Wednesday 09:00am New Generation of Storage Tiering: Less Management Lower Investment and
Increased Performance
10:30am IBM Smarter Storage Strategy
12:30pm IBM's Cloud Storage Options
01:45pm IBM Spectrum Scale (Elastic Storage) Offerings
Thursday 12:30pm The Pendulum Swings Back -- Understanding Converged and
Hyperconverged Environments
Friday 09:00am IBM Spectrum Storage Integration with OpenStack
- 4. © Copyright IBM Corporation 2015
The Pendulum Swings on Infrastructure Design
Internal Storage
• Personal
Information
Managers (PIM)
• Mainframe
• AS/400
Advantages
Simple, self-contained
Disadvantages
• Simple, self-contained
• Scalability limited to what can fit
inside the hardware container
• Single Point of Failure (SPOF)
(unless you keep 2 or more
copies of data across
independent systems)
• Backups, Security and other
Policy enforcement is done on a
system-by-system basis
individually
3
- 5. © Copyright IBM Corporation 2015
The Pendulum Swings to External Storage
External Storage
• Mainframe
• AS/400
• Linux, UNIX,
Windows
Advantages
Two or more servers can directly
attach to external storage
High-availability clusters
RAID for data protection and
performance
Shared Cache
More room for storage growth
Centralize features, snapshots and
tape drives for backups
Disadvantages
• Scalability limited to number of
hosts attached
• Limited distance for external cables
4
- 6. © Copyright IBM Corporation 2015
IBM Entry-Level And Midrange
Storage Positioning
5
DCS3700 Performance
• Host: FC, SAS, iSCSI
• 360 SAS, NL-SAS, SSD
• DCS3700 Expansion
DCS3860
• Host: SAS
• 360 SAS, NL-SAS,
SSD
• DCS3860 Expansion
DCS3700
• Host: FC, SAS, iSCSI
• 180 SAS, NL-SAS, SSD
• DCS3700 Expansion
Storwize V5000
• Host: FC, FCoE,
SAS, iSCSI
• 960 SAS, NL-SAS,
SSD
• Storwize Expansion
Storwize V3700
• Host: FC, FCoE,
SAS, iSCSI
• 240 SAS, NL-SAS,
SSD
• Storwize
Expansion
Random workloads Sequential workloads
Deep Computing Storage (DCS)Storwize family
built with IBM Spectrum Virtualize™
12 or 24 drives
in 2U rack space
60 drives
in 4U rack space
- 7. © Copyright IBM Corporation 2015
The Pendulum Swings to Networked Storage
SAN
LAN
Advantages
Many more hosts can be attached
Greater distances enables Disaster
Recovery
Fewer, larger systems like Tape
Libraries easier to manage
Disadvantages
• SANs and LANs requires different
skill sets
• OS-specific and device-specific
management tools
Networked Storage
• SAN and NAS attached
storage systems
• IBM Spectrum Scale
6
- 8. © Copyright IBM Corporation 2015
The Problem: Islands of SAN, File and Object level data
SAN-level Storage
OS-specific file systems on
direct -attach or SAN-based
devices
Sharing requires file transfers
Provides “Context” for
Analytics of Social and Mobile
transactions
File-level Storage
NAS encourages sharing
across social networks
Desire for file sync-and-
share across desktops and
mobile
HDFS requires transfer
(ingest) from other sourcesJFS2
EXT4
NTFS
CIFS
HDFS
NFS
Object-level Storage
New Web and Mobile apps
prefer Object-level access
Amazon S3
OpenStack
Swift
7
- 9. © Copyright IBM Corporation 2015
Spectrum Scale – Flexible File and Object Storage
FS1
FS256. . .Exabyte-Scale,
Global
Namespace
One big file system or
divide into as many as
256 smaller file systems
Each file system
can be further
divided into fileset
containers
Flash and Disk LUNs
are called Network
Shared Disks (NSD) Metadata can be separated
to its own Pool or
intermixed with data
Files can be
migrated to Tape
8
- 10. © Copyright IBM Corporation 2015
IBM Spectrum Scale™ – Supported Topologies
Twin-tailed
SAN
Internal, Direct-Attach
Shared PoolsFPO Pools
NSD Servers
Access files on direct,
twin-tailed or SAN
attached disk
Can export files to
application nodes
File Placement
Optimization (FPO)
Servers
Access files on direct
attached disk
Exports files to other
FPO servers
External Clients
Access files via file and object
protocols over IP network
TCP/IP
9
NSD Clients
Access files via
SAN, TCP/IP or
RDMA
TCP/IP or RDMA network
- 11. © Copyright IBM Corporation 2015
IBM Spectrum Scale™ replaces
other POSIX file systems
SAN
Direct-Attach
SAN-level Storage
JFS2
EXT4
NTFS
• Works like OS-specific file systems
• No file transfers required between OS
• Linux on x86, POWER and z Systems
TCP/IP or RDMA Network
Twin-tailed
10
- 12. © Copyright IBM Corporation 2015
IBM Spectrum Scale™ – More than just a file system!
11
ROBO
Other NFS
Other
Datacenters
Scale
Active File
Management
(AFM) caches
data to where it is
needed, can be
used to migrate
from other NFS
Information Lifecycle
Management (ILM)
moves data across tiers of
flash and disk
Hierarchical Storage
Management (HSM)
migrates infrequently
accessed files to tape,
automatically recalls back
when accessed
Local Read-Only Cache
(LROC) caches the busiest
blocks of files on local flash
Disaster Recovery
(DR) asynchronously
mirrors data to remote
locations
Migrate/Recall Tape
NSD Client
- 13. © Copyright IBM Corporation 2015
SAN
Twin-tailed
Protocol Servers
NFS, CIFS
AIX, Linux,
Mac OS X,
Windows,
VMware,
z/OS
Feature of Spectrum Scale
on Linux
Share files with clients
using NFS, CIFS and Object
protocols
All nodes can share the
same data
If Protocol Server Node
fails client connections are
moved to another server
Protocol Server Node(s)
need “NSD Server” License
External Clients need no
Spectrum Scale License
Clustered Protocol Servers
12
TCP/IP
OpenStack