IBM general parallel file system - introduction

Presentation from the HPC event at IBM Denmark - September 2013, Copenhagen

1. IBM General Parallel File System (GPFS™) 3.5
and GSS Introduction
IBM Confidential
Karl Hansen, Nordic HPC/Technical Computing Sales Manager
IBM Systems & Technology Group
A New Era in Technical Computing: Powerful. Comprehensive. Intuitive.

2. Technical Computing: Powerful. Comprehensive. Intuitiv e
Extreme Scalability
Proven Reliability
Manageability
The IBM General Parallel File System
TM
(GPFS
TM
)
Shipping since 1998
© 2012 IBM Corporation
File system
2
63
files per file system

Maximum file system
size: 2
99
bytes

Production 19PB file
system
Number of nodes

1 to 8192
Extreme Scalability
No special nodes
Add/remove nodes
and storage on the fly
Rolling upgrades
Administer from any
node
Data replication
Snapshots
File system journaling
Proven Reliability
Integrated tiered storage
Storage pools
Quotas
Policy-Driven automation
Clustered NFS
SNMP monitoring
TSM / HPSS (DMAPI)
Manageability

3. Technical Computing: Powerful. Comprehensive. Intuitiv e
IBM General Parallel File System (GPFS)
IBM General Parallel File
System (GPFS) is
a scalable high-
performance file
A highly available cluster architecture
Concurrent shared disk access to
© 2012 IBM Corporation
3
performance file
management
infrastructure for
AIX
®
, Linux
®
and
Windows™ systems.
Concurrent shared disk access to
a global namespace
Capabilities for high performance
parallel workloads

4. Technical Computing: Powerful. Comprehensive. Intuitiv e
File data infrastructure optimization GPFS enables:
A global namespace
across platforms
High performance
common storage
Connections
SAN TCP/IP
InfiniBand
Management
Databases File servers
© 2012 IBM Corporation
4
common storage
Eliminating copies
of data
Improved storage
utilization
Simplified file
management
Availability
Data Migration
Replication
Backup
Management
Centralized
Monitoring
Automated
File Mgmt
Application
servers
Backup
and
archive

5. Technical Computing: Powerful. Comprehensive. Intuitiv e
How is GPFS different?
GPFS
Centrally deployed, managed,
Massive namespace support
© 2012 IBM Corporation
5
All features are included. All software features: s napshots, replication and multi-site
connectivity are included in the GPFS license. With no license keys except for client and
server to add on, you get all of the features up fr ont.
SAN
Centrally deployed, managed,
backed up and grown.
Seamless capacity and
performance scaling

6. Technical Computing: Powerful. Comprehensive. Intuitiv e
Network-based block input output
LAN
NSD clients NSD
GPFS
Application data access on network
attached nodes is exactly the same as
a SAN attached node. GPFS
transparently sends the block level IO
request over a TCP/IP network.
© 2012 IBM Corporation
6
Why?
SEnable virtually seamless multi-site operations
SReduce costs for data administration
SProvide flexibility of file system access
SEstablish highly scalable and reliable data storage
SFuture protection by supporting mixed technologies
SAN
NSD servers
SAN
GPFS

7. Technical Computing: Powerful. Comprehensive. Intuitiv e
IBM General Parallel File System (GPFS™) –History & Evolution
HPC
GPFS
General File
Serving
Virtual
Tape Server
(VTS)
Linux®
Clusters
(Multiple
architectures)
GPFS 2.1-2.3
HPC
Research
Visualization
Digital media
Seismic
Weather
Life sciences
32 bit /64 bit
GPFS 3.1-3.2
First
called
GPFS
GPFS 3.4
Enhanced
Windows cluster
support
- Homogenous
Windows server
Performance and
scaling
improvements
GPFS 3.3
Restricted
admin functions
Improved
installation
New license
model
Improved
Ease of
administration
Multiple
-
Information
lifecycle
management (ILM)
Storage pools
File sets
Policy engine
GPFS 3.5
Caching via
Active File
Management
(AFM)
GSS - GPFS
Storage Server
GPFS File
Placement
© 2012 IBM Corporation
2006 2005 2002 1998
Serving
Standards
Portable
operating
system
interface
(POSIX)
semantics
-Large block
Directory and
small file perf
Data
management
architectures) IBM AIX®
Loose Clusters
Inter-op (IBM AIX
& Linux)
GPFS Multicluster
GPFS over wide
area networks
(WAN)
Large scale
clusters
thousands of
nodes
2009
Enhanced
migration and
diagnostics
support
2010
snapshot and
backup
Improved ILM
policy engine
2012
Multiple
-
networks/ RDMA
Distributed token
management
Windows 2008
Multiple NSD
servers
NFS v4 support
Small file
performance
Placement Optimizer (FPO)

8. Technical Computing: Powerful. Comprehensive. Intuitiv e
New Storage Solution fulfilled exclusively through IBM Intelligent
Cluster
High capacity, High performance, High Value offerin g
Product importance

Single, integrated, fully supported IBM solution
A Disruptive HPC Play - GPFS Storage Server (GSS) At a Glance
The New High Capacity, High PerformanceStorage Solution
© 2012 IBM Corporation 8

Single, integrated, fully supported IBM solution
Built to leverage a strong GPFS software market
High capacity, scalable building-block approach - pe rformance and
capacity increases as you add multiple building blo cks
Cost competitive
Extreme data integrity and reduced latency with fas ter rebuild times

9. Technical Computing: Powerful. Comprehensive. Intuitiv e
GPFS Storage Server - Product Description
GSS is a new storage solution fulfilled exclusively through the IBM Intelligent Cluster Two x3650 servers combined with either four or six JBODs
Two models: GSS 24 and GSS 26
– GSS 24 (Entry): 4 JBODs – starts at nearly 500TB of stor age
space
– GSS 26 (Main): 6 JBODs – starts at over 700TB of storage
space
Data
striped
across
all disks
© 2012 IBM Corporation
2 and 3 TB options
10GbE or FDR Infiniband interconnects, or both!
Scalable Building Block approach to HPC Storage - perfor mance
and capacity increase as you add multiple building blocks
Complete Storage Solution with noStorage controllers
De-Clustered RAID Techniques
Built on GPFS software
Industry standard components
–Leverages standard components including x3650s, NetApp
JBODs, LSI SAS cards and lots of HDDs, and Intelligent
Cluster fulfillment as a single, integrated, fully supp orted IBM
GSS 24 Model

10. Technical Computing: Powerful. Comprehensive. Intuitiv e
x3650 M4 Server
Storage solution includes Data Servers,
Disk (2TB or 3TB NL-SAS, SSD), Software,
InfiniBand / Ethernet with no Storage Controllers
IBM System x GPFS Storage Server provides a comprehensive
storage solution with a scalable building block approach
© 2012 IBM Corporation
JBOD
Disk Enclosure
GSS 24: Light and Fast
2 3650 servers +
4 JBOD 20U rack
10 GB/Sec
GSS 26: HPC Workhorse
2 3650 servers +
6 JBOD Enclosures, 28U
12 GB/sec
High-Density HPC Option
6 3650 servers + 18 JBOD
2 - 42U Standard Racks
36 GB/sec

11. Technical Computing: Powerful. Comprehensive. Intuitiv e
Why you should care… Cost competitive
Fewer parts means lower cost
Leverages System x servers and Commercial JBODs
Affordably Scalable Building Block
approach to HPC Storage
Performance and capacity increases as you add
multiple buidling blocks
Start Small and Scale via incremental additions
Add capacity AND bandwidth
•
Fast rebuild times and industry
-
leading performance
© 2012 IBM Corporation 11
Extreme data integrity and reduced
latency
•
Fast rebuild times and industry
-
leading performance
• Better sustained performance
• Industry-leading throughput using efficient De-
Clustered RAID Techniques
Built on GPFS
• The Infrastructure for Global Technical Computing
Data Management
Fully integrated, fully supported
Single, integrated, fully supported IBM solution
Complete Storage Solution with no Storage
controllers
Easy to order through Intelligent Cluster

12. Technical Computing: Powerful. Comprehensive. Intuitiv e
A data management portfolio for Technical Computing
Focus:
Ease of Use
Reliability
IBM Data
Management
Leadership
Focus:
Managed Building
Block
Focus:
Raw Raw Performance
I/O Bandwidth
Government
High End Research
Petroleum
Financial
Media/Ent.
GPFS Storage Server
© 2012 IBM Corporation
Financial Services
Smaller
Installations
Higher End
University
Media/Ent.
CAE
Bio/Life Science IBM Tape, Tivoli Storage Manager, and HPSS
Direct Attached
(DS3500 + V3700)
SONAS
DCS3700
DCS3700+

13. Technical Computing: Powerful. Comprehensive. Intuitiv e
NSD File Server 1
Clients
FDR IB
10 GbE
NSD File Server 1
x3650
Clients
What makes this different
File/Data Servers
© 2012 IBM Corporation
JBOD Disk Enclosures
NSD File Server 2
Servers!
Migrate RAID
and Disk
Management to
Standard File
Servers!
Custom Dedicated
Disk Controllers
JBOD Disk Enclosures
x3650
NSD File Server 2
GPFS Native RAID GPFS Native RAID
File/Data Servers

14. Technical Computing: Powerful. Comprehensive. Intuitiv e
System x
Smarter Systems for a Smarter Planet
© 2012 IBM Corporation
14

15. Technical Computing: Powerful. Comprehensive. Intuitiv e
For more information
ibm.com/systems/software/gpfs
© 2012 IBM Corporation
Email [email protected] or contact your IBM Representative