Sparc solaris servers

SPARC Solaris Servers
Download this slide
http://ouo.io/a3Sza

The following is intended to outline our general
product direction. It is intended for information
purposes only, and may not be incorporated into any
contract. It is not a commitment to deliver any
material, code, or functionality, and should not be
relied upon in making purchasing decisions.
The development, release, and timing of any
features or functionality described for Oracle’s
products remains at the sole discretion of Oracle.

Agenda
• Why Oracle’s SPARC Servers
• SPARC Solaris
– Premier Performance for all Workloads
– Designed for Enterprise Cloud Computing
– Engineered for Enterprise Applications
• SPARC Portfolio and Roadmap

Transforming The Technology Stack
Investing in
Best of Breed
End-to-end
Engineering
Co-Engineered with
Enterprise Software
Engineered
Systems
HW/SW Engineered
to Work Together
Compute, Storage, Network
Building Blocks

• Open and standards-based
• Optimized, integrated and extensible
• Better performance, reliability, security
• Shorter deployment times
• Easier to manage and upgrade
• Lower cost of ownership
• Reduced change management risk
• Integrated support
Oracle + Sun:
Complete, Open, Integrated Systems
Hardware and Software
Engineered to Work Together

Redefining General Purpose
New Generation of SPARC
• SPARC SuperCluster
– Fastest General Purpose Platform
• SPARC T4 Servers
– Biggest Single Generation Performance Boost in History
• Solaris 11
– Breakthrough scale, virtualization and manageability

General Purpose
SPARC
SuperCluster
Purpose Built
Database Appliance Exalytics
Big DataExalogicExadata
Engineered Systems & Appliances
Cloud Built In

Purpose Built Universal Purpose
Price/PerfScaleandRAS
Exalogic
Exadata
SPARC
SuperCluster
Big Data Appliance
Engineered Systems & Appliances

SPARC SuperCluster Beats IBM
Better Performance. Better Value.

SPARC SuperCluster Beats HP
Better Performance. Better Value.
Superdome2
2x Better Price/Performance

Premier Performance

SPARC SuperCluster
Best for Oracle. Runs All Existing Workloads.
SPARC T4 Compute Pool
9 World Records over IBM and HP
across every tier
Exadata Storage Cells
1.2M IOPS, 32 GB/s query throughput
Exalogic Elastic Cloud
10x Java performance
Integrated ZFS Storage
2x faster and ½ the price of NetApp
Solaris 11
Cloud provisioning in seconds
Unmatched Scalability
Virtualization
Near zero virtualization overhead
InfiniBand
5-8x the speed of current networks
Enterprise Manager
Increase productivity, reduce
downtime
See performance substantiation slides

T4-2 T4-4T4-1
9 World Records
Introducing SPARC T4 Server
New Brain. Same Body.
T4-1B T4-2 T4-4T4-1
Unheard of generation to generation acceleration
– Up to 5x per thread performance compared to T3 servers
– Starts at $16K – Virtualization and Security included!
– Up to $160K – 1 TB of memory included!
T4 Processor
• 3.0 GHz with OOO execution
• Dedicated L2 128KB cache
• Shared L3 4MB cache
• 8 Cores with Private L2 Cache
• Dynamic Threading
• Enhanced Built-in Encryption
• Built-in Virtualization
T4 Systems
• Up to 1 TB of memory
• Built-in, no-cost virtualization
• High-bandwidth and high-capacity I/O
• Integrated 10 GbE
• Solaris binary compatibility

World Record Java and Database Performance
SPECjEnterprise2010
2.4x faster than Power7 with DB2 and WebSphere
7x better price performance for Java
IBM: One Power 780
$467,856
T4-4
Servers
$1,297,956
See performance substantiation
slides

World Record TPC-H
Beats IBM’s Claims of 4:1 Core Performance Advantage
$800K cheaper and 22% faster than Power7 & Sybase
$125K cheaper and 3.6x faster than HP Superdome & Oracle 11g
TPC-H @1000GB

Leadership in Application Performance
Delivering General Purpose Computing
PeopleSoft
2.5x faster vs HP Itanium
2.8x faster vs IBM z10
JD Edwards
50% faster vs IBM z10

“Our benchmark testing of Oracle’s SPARC T4 system on our ERP System led to
impressive results. We saw performance enhancements of up to 4x compared to our
legacy server. Deployment was easy; just a drop in. Our planned worldwide rollout will take
advantage of the SPARC T4’s very modest space, energy and heat requirements. In our
production deployment we plan to use Oracle Solaris virtualization to implement containers
and consolidate several workloads onto a single server.”
Thomas Kleber, Department Lead IT, Kromberg & Schubert GmbH & Co
“At Qualcomm, our testing on the SPARC T4 systems showed huge performance
improvements and impressive results across the operating system and SunRay login
server. With the T4's new CPU performance, RAM capacity and stability of SPARC
Solaris, our ‘per system user limit’ will increase from 100 users on an M3000 to 200
users, while providing 4x CPU headroom. With increased experience and proficiency at
managing the quantity of ‘liability’ on one box, we hope to see numbers as high as 400
users per T4 Server.”
Rob Mallory - IT Architect
SPARC T4 Servers
More than 50 impressed Beta customers

• 3.0 GHz
• 8 Cores, 64 Threads
• Dynamic Threading
• Out of Order Execution
• 2 On Chip Dual-Channel DDR3 Memory
Controllers
• 2 On Chip 10 GbE Networking
• 2 On Chip x8 PCIe gen2 I/O Interfaces
• 16 On Chip Crypto algorithms
• Balanced high-bandwidth interfaces & internals
• Co-engineered with Oracle software
SPARC T4

“Oracle's SPARC T4 Servers are designed to deliver
performance at every tier of the data center, across a wide range
of workloads, while providing a robust building block for Oracle’s
next-generation of engineered systems.”
Jean S. Bozman, Research VP, Enterprise Servers, IDC
“The 5X improvement in Oracle’s SPARC T4 single thread
performance shows that a well balanced CPU design can deliver
both high throughput and high single thread performance.”
Nathan Brookwood, Research Fellow, Insight 64
Oracle’s Next Generation SPARC Servers

Designed for Enterprise
Cloud Computing

NEW
New Generation of SPARC Solaris
Foundation For the Cloud
• SPARC T4 Servers
– Up to 5x Faster than T3
• SPARC SuperCluster
– As Fast as Exadata for Database
– As Fast is Exalogic for Java
• Solaris 11 and Solaris 10
• SPARC Solaris Upgrade Path

Existing Data
Just Add To Current Datacenters
Fibre Channel and Ethernet Ready.

+
SPARC
SuperCluster
Half Rack
Existing Data Existing
Applications
Ease of Deployment
Run existing Applications immediately.

+
SPARC
SuperCluster
Full Rack
+
SPARC T5-4
Future Upgrade
Existing Data
Seamless Scalability: Future Upgrades
Start small. No limits.

Exadata Storage
Expansion Rack
Sun ZFS Storage 7420
+ ++
SPARC
SuperCluster
Full Rack
Existing Data
Highly Scalable Storage
Easily expand DB and regular storage.

SPARC SuperClustersExisting Data Exadata Storage
Expansion Rack
Highly Scalable Compute Options
Easily add in SPARC SuperClusters

Oracle Solaris 11
Built for Enterprise Clouds
Seamless Scaling with Hardware: 9 NEW World Records
Dynamic Threading + Next Gen I/O Terrabyte of Memory + 10,000s of threads
Simplified Administration: 4x faster upgrades, 2.5x faster reboots
Provision in seconds + Fool-proof updates + Horizontal Scaling
Virtualization: Cloud provisioning in seconds, Near zero overhead
Virtualized network with QoS + Solaris 10 Zones + Built in
Efficient Data Management: 5x compression
Integrated deduplication, compression + Infinite snapshots and clones + No cost replication
Advanced Protection: 3x faster encryption
Hardware Accelerated Encryption + Integrated auditing + Advanced user access controls

Leading Security
Built-in Encryption. No Overhead. No Software Changes.
43% faster secure queries
Encryption: In-memory
1.8x better throughput
(reduced time)
ZFS Encryption: Encrypted Filesystem
3x faster encryption

Comprehensive Management Capabilities
Across the lifecycle for the entire hardware infrastructure
*Live migration not currently supported on SuperCluster
Discovery and
Provisioning
Discover servers
Provision Bare Metal
servers
Clone VM templates
Maintainance
Health checks
Remote Management
 Telemetry
Phone Home
 Proactive Support
SR Management
Resource Monitoring and
Management
Monitor Servers, Storage, Network
Exception driven Alarms
Integrate with Incident
Management systems
Configuration
Management
Manage configuration drifts
Patch Operating System
Update Firmware
Track Patch Compliance
Virtualization Management
Manage Oracle VM, Containers
Template Provisioning
Live migrate workloads

"We have tested the new SPARC T4-4 system against our current Siebel infrastructure on
the SPARC M5000 and the performance of the T4 system was very positive.
The moving of the applications from the current infrastructure to the new T4-4 was flawless
with no need for a porting plan or specific tuning.
Using virtualization with Solaris Zones in the T4-4 system against a non virtualized Solaris
environment in M5000—this was most impressive. Performance has doubled from our
current servers. We think the SPARC T4-4 is the ideal platform for virtualizing and
consolidating Siebel and other applications, from the web server down to the database server
tier.”
Umberto Angelucci, CTO, SKY Italia
“Our beta testing of Oracle’s SPARC T4 server demonstrated impressive performance and an
architecture that reduced complexity and operating costs. This excellent performance on our mixed
throughput needs, combined with the SPARC T4’s attractive software licensing terms, led to our
decision to use the SPARC T4 system for our next generation archiving system.”
Mr. Hans-Juergen Wolf, Archive Manager, DLR
“Because we develop content management systems for many German federal and local governments, it is critical
that Materna base its solutions on the most scalable and high performance technology. We are beta testing
Oracle’s SPARC T4-1B blades with our "Government Site Builder" application and seeing throughput and single
thread performance improvements of 5-7 times in comparison to the previous generation SPARC blades.”
Dr. Georg Kösters, Head of Business Process Management - Business Division Applications, MATERNA GmbH
Virtualize Applications with SPARC T4 Servers

Engineered for
Enterprise Applications

VS.
The Unique Oracle Advantage
Hardware and Software Engineered to Work Together

Oracle Optimized Solutions
Applying SPARC SuperCluster and SPARC T4 to Applications
• Designed with SPARC SuperCluster
– Oracle Optimized Solution for PeopleSoft HCM
– Oracle Optimized Solution for WebCenter Content
• Designed with SPARC T4
– Oracle Optimized Solution for Oracle Database
– Oracle Optimized Solution for WebLogic Server
– Oracle Optimized Solution for JD Edwards EnterpriseOne

Oracle
Outperforms all competitors
Every Tier
Every Application
Security
3x x86
Database
3x IBM Power7
Java
2.2M JMS ops/sec
Communications
Billing
4x x86 consolidation
Comms Service
Broker
2.7x x86
PeopleSoft
3x HP Itanium
PeopleSoft
2.8x IBM z10
Data Warehousing
3.6x HP Superdome 2
Web
1M HTTP
ops/sec
JD Edwards
2.5x IBM Power7
Security
5x IBM Power7
Java Enterprise
2.4x IBM Power7
Database Refresh
4x IBM Power7
Database
2.3M IOPS
WebCenter
Content
11x x86

“As a leading global financial institution, ING needs servers that provide the highest levels
of reliability and performance to support our mission-critical business applications. Our
testing of the ING application stack on Oracle's SPARC T4 servers running Oracle
Solaris proves Oracle’s Sun servers deliver quality, predictability and complete backwards
compatibility while greatly enhancing system performance.”
Frank Schots, Senior UNIX Systems Engineer
“We chose to test Oracle SPARC SuperCluster for performance improvement of our
current datawarehouse system and consolidation of sparse Oracle databases through
implementation of RAC features. Our heaviest operation is datawarehouse loading
and with SPARC SuperCluster we are expecting a huge improvement.”
Aldo Chiaradia, CIO, Benetton Group S.p.A.
“SAS® Business Analytics enables faster, more accurate data-driven decisions. Whether it
is risk analysis, fraud detection or complex optimization mathematics, implementing SAS
Business Analytics on SPARC T4 solves critical business issues in transformational ways.
SPARC has long been a proven platform for SAS applications; T4 delivers SPARC
performance at its best.”
Paul Kent, SAS Vice President of Platform Research and Development
SPARC T4 Servers and SuperCluster

SPARC Portfolio and Roadmap

Mission Critical
Consolidation, large DB,
complex workload
Engineered Systems
Enterprise Clusters,
Optimized Solutions
Enterprise Servers
Designed for optimal
performance on all workloads
M5000
M4000
M3000
M9000
M8000
T4-4
T4-2
T4-1
T4-1B
SPARC Server Portfolio
Foundation for Mission Critical Computing

Extreme Performance
Breakthrough Innovation
Secure by Design
SPARC/Solaris Built-in Encryption &
Compression
Oracle VM, Solaris Zones, low latency
virtualized environments
All fully redundant hardware,
shorter failover times, Solaris
Cluster, physically partitioned
compute nodes, ASM
Streamlined
Management
Fast & foolproof upgrades.
High Availability
Built in
Fully tested, configured, optimized,
sized, and certified
Exadata Storage Cells, Intelligent Storage Grid, SPARC
T4-4 compute nodes, and the Solaris 11 Operating
System.
Telemetry across the stack, Integration
across the stack, Solaris Ease of Patching,
New image packaging system provides
fast application installs.
Virtualization
Low Overhead
Ease of Deployment
Integrated Hardware and Software
Integrated crypto hardware automatically
secures Database, Middleware, Applications,
and Solaris File System, new admin controls,
and secure OS startup.
Dynamic Threads
Resource Management
Performance Workload Management
SPARC SuperCluster T4-4

SPARC SuperCluster T4-4
Engineered to Work Together Extreme Performance
Breakthrough Innovation
10x compression for Data Warehouse queries
10x OLTP Response time
2x better performance against IBM and HP
Secure by Design
SPARC/Solaris Built-in Encryption &
Compression
5x faster security than IBM
Power7, 3x faster than x86
High Availability
Built in
99.999% Uptime
Streamlined Management
Fast & foolproof upgrades
2.2 minutes vs 24 minutes
Virtualization
Low Overhead
Ease of Deployment
Integrated Hardware and Software
Deploy in days not months
On the fly performance adaptation
to your business needs
Resource Management
Performance Workload Management

SPARC T4-4 Compute Nodes
Exadata Storage Servers
• 1,200 CPU threads
• 4 TB DRAM
• 97 to 198 TB Hard Disk
• 8.66 TB Flash
• 1.2M IOPS
• 42 GB/sec Storage Bandwidth
• 896 Gb/sec InfiniBand Interconnect
InfiniBand Switches
ZFS Storage Appliance
SPARC SuperCluster Architecture
Best infrastructure solution for enterprise applications

Virtualization
Oracle Solaris Zones and Oracle
VM Server for SPARC
Clustering
Oracle Solaris Cluster
Oracle Clusterware
Database
11g R2, 10g and other DB
Applications
E-Business Suite, PeopleSoft, SAP,
Siebel, and much more
Middleware
Oracle Fusion Middleware
Applications
Optional with Exalogic Elastic Cloud
Management
Oracle Ops Center and Enterprise
Manager Grid Control
+
Operating System
Oracle Solaris 11
for Exadata and Exalogic nodes
Solaris 10/11 nodes for applications
Software Stack

• Intelligent storage
– Smart Scan query offload
– Scale-out storage
+ ++
• Hybrid Columnar Compression
– 10x compression for warehouses
– 15x compression for archives
Compressed
primary
standby
test
dev
backup
Uncompressed
• Smart Flash Cache
– Accelerates random I/O up to 30x
– Doubles data scan rate
Data remains
compressed
for scans and
in Flash
Benefits
Multiply
Exadata Storage Cells

SPARC SuperCluster Management
Converged Hardware Management
• Unified management of Servers,
Storage, and Network Fabric
• Simplified management of virtual
infrastructure for easy application
consolidation
• Instant Network and Storage
provisioning
• Automated update of all firmware
and software components
• Direct connection to Oracle
knowledge-base speeds problem
resolution

Engineered Systems
Significantly Accelerate Time to Value
Build From Scratch
with Components
Take delivery of an
Engineered System
Pre-implementation
System sizing
Acquisition of
components
Installation and
configuration
Testing and
Validation
Weeks to Months
Hardware.
Software.
Complete.
Faster deployment
Lower Risk
Pre-configured
Days
Deploy up to 4x faster
than IBM or HP

SPARC T4 Servers
• SPARC T4 Systems designed for:
– Broad end-to-end Enterprise deployments
• Database, Middleware and Web applications
– Deployments requiring high security with performance
– Virtualized environments and consolidation
• SPARC T4 delivers:
– Up to 5x single-thread performance vs. T3 systems
• Database optimized – OLTP, DW and Batch
• Faster application boot times for Oracle WLS and SOA
– Similar throughput performance as T3 systems
– Cost-effective consolidation platform for legacy workloads
– Low cost secure data encryption transactions

SPARC T4-1B SPARC T4-1 SPARC T4-2 SPARC T4-4
Processor
SPARC T4
2.85GHz
SPARC T4
2.85GHz
SPARC T4
2.85GHz SPARC T4 3.0GHz
Max Processor Chips 1 1 2 4
Max Cores/Threads 8, 64 8, 64 16, 128 32, 256
DIMM Slots 16 16 32 64
Max Memory 256 GB 256 GB 512 GB 1 TB
Drive Bays 2 8 6 8
I/O Slots
2 x PCIe 2.0 EM,
2 NEM, 1 REM,
1 FEM slots
6 LP x 8 PCIe 2.0,
4 x 1 GbE ports,
2 x 10 GBE
XAUI ports
10 x PCIe 2.0,
4 x 1 GbE ports,
4 x 10 GbE
XAUI ports
16 x PCIe 2.0 EM,
4 x 1 GbE ports,
8 x 10 GbE
XAUI ports
Form Factor/RU Blade Rack 2U Rack 3U Rack 5 U
Key Differentiators of SPARC T4
• 5x single thread performance increase over SPARC T3 processor while retaining throughput
performance of SPARC T3
• Expanded application workload fit to meet requirements for both multi thread and single thread
applications
SPARC T4 Servers

Key SPARC T4 System Advantages
Optimizing the Datacenter
Feature Attribute Benefit
2.85 & 3.0 GHz clock
Faster single threaded
processing
Shorter application boot times, rapid
batch processing for quicker results
Integrated encryption
engines
Up to 3x faster security
for data encryption
Reduced cost for secure datacenter
operation without a performance
penalty
Multithreaded
architecture
Preserve T3 system
levels of throughput
Time savings due to no application
changes for T4 systems
Binary compatibility
Existing SPARC/Solaris
applications run
unmodified
Rapid time to adoption of new
systems and new service deployment
Built-in Virtualization with
OVM Server for SPARC
Flexible logical
partitioning and live
application mobility
Improved uptimes for critical services
and higher system utilization rates
Solaris 11
Reliable, secure and
streamlined operation
Faster system updates and reboots to
improve datacenter operations

New SPARC T4 Systems
Optimized for Oracle Enterprise Software
Scales to meet all datacenter
workloads from web tier to
database, with up to 5x better
single threaded performance than
prior generation
Scalable
With Oracle Solaris, optimized for
Oracle applications including
built-in virtualization and powerful
systems management
Integrated
Faster performance for Oracle
Database and WebLogic Suite,
building block for fault tolerant
SPARC SuperCluster systems
Optimized
Wire speed encryption
capabilities without extra cost or
performance penalties
Secure
End-to-End Datacenter Performance

SPARC M-Series
Delivers Longevity, Security and Scalability
• Industry leading reliability, availability and serviceability
• Extreme single threaded performance
• 4th in-box upgrade
• Seamless scalability: 1 to 64 CPUs
• Mix/Match CPUs for improved TCO
• Highest level of consolidation
• Breakthrough virtualization
• SPARC64 VII+
– 2x L2 Cache for faster database performance
– Up to 3.0 GHz – 20% performance improvement

ResultsBusiness Challenges
• Quickly add 150% capacity
to single instance ERP
• Additional 30,000 employees
from Sun acquisition
• Rapid in-place deployment
with minimal business impact
• Non-stop database operations
• Mixed workload in a single
environment, not just OLTP
• Future Acquisitions & Agility
• Migrated to 3 x M9000’s
• 2.5x in Performance
• 50% less:
- space, power, operational
requirements, complexity
• $400K savings
• 95% Improvement
in RTO
• Over 5x faster App deployment
• Built for Zero unplanned downtime
Oracle Runs on Oracle
For 15+ Years Oracle has run its global ERP on SPARC Solaris

“8 out of 10 CIOs would like to implement
a more proactive approach to supporting
their IT environment”
- IDG Research
My Oracle Support
Powerful Proactive Support Tools
• Advanced knowledge sharing to tap into
the best practices and expertise of the
My Oracle Support Community
• Guided lifecycle advice tools to provide
step-by-step assistance in planning
updates and upgrades
• Problem prevention health checks that
provide advice based on your unique
Oracle configuration
• Problem detection and fast resolution
with tools such as Oracle Auto Service
Request
• Integrated stack delivery with Oracle
Enterprise Manager

Hardware & Software Designed with Support in Mind
• Standardized components and
extensive configuration testing
• Fault tolerance through built-in
redundancy and replication, clustering,
and predictive self-healing
• Embedded diagnostics enabling deep
observability and real-time
troubleshooting
• Unified management from the hardware
to the application

Integrated Support
Single Point of Accountability
• Consistent service across the Oracle
solution stack, backed by a single level
of 24/7 support covering all hardware
and software elements
• Integrated support and updates
through one service organization and
one online support interface: My Oracle
Support
• Accountability for the complete Oracle
solution, combining the benefits of
vendor-direct service with a single
point of contact

2010 SPARC Server Roadmap
Maximizing Results
5 Year Trajectory
Cores 4x
Threads 32x
Memory Capacity 16x
Database TPM 40x
Java Ops Per Second 10x



T-Series
1-4 Socket
+ 2x Throughput
M-Series
1-64 Socket
+ 20%
M-Series
8-64 Sockets
+6x Throughput
+1.5x Single Strand
T-Series
1-4 Sockets
+1x Throughput
+3x Single Strand
M-Series
8-64 Sockets
+2x Throughput
>1x Single Strand
T-Series
1-8 Sockets
+2.5x Throughput
>1x Single Strand
SPARC
1-64 Sockets
+2x Throughput
+1.5x Single Strand
Solaris 11 Update
High-Availability
Memory Scalability
Virtualization
Solaris 11
Software Lifecycle
Scalability, Networking
Security
Solaris 11 Update
System Management
IO Scalability
Solaris 11 Update
Core Scalability
Solaris 11 Express
Software Lifecycle
Scalability
Networking
2010 2011 2012 2013 2014 2015

2011 SPARC Server Roadmap
Maximizing Results
5 Year Trajectory
Cores 4x
Threads 32x
Memory Capacity 16x
Database TPM 40x
Java Ops Per Second 10x



T-Series
1-4 Socket
+ 2x Throughput
M-Series
1-64 Socket
+ 20%
T-Series
1-8 Sockets
+2.5x Throughput
>1x Single Strand
T-Series
1-8 Sockets
+20%
Solaris 11 Update
High-Availability
Memory Scalability
Virtualization
Solaris 11
Software Lifecycle
Scalability, Networking
Security
Solaris 11 Update
System Management
IO Scalability
Solaris 11 Update
Core Scalability
Solaris 11 Express
Software Lifecycle
Scalability
Networking
2010 2011 2012 2013 2014 2015

Available
Now
On Track
In Test
In Test
On Track
T-Series
1-4 Sockets
+1x Throughput
+5x Single Strand
M-Series
16-64 Sockets
+6x Throughput
+1.5x Single Strand
M-Series
16-64 Sockets
+2x Throughput
>1x Single Strand
SPARC
1-64 Sockets
+2x Throughput
+1.5x Single Strand
NEW
Faster!
Early!
NEW
Software in Silicon
Feature Set

Software in Silicon
•Security: Enhanced
cryptography
•Oracle numbers Arithmetic
Acceleration
•Hardware Decompression
•In Memory Columnar
Database Acceleration
•Memory Versioning
•Low Latency Clustering
Increased Performance
•Higher core frequency
•Multiple pipelines per core
•Increased core counts per chip
•Larger caches
•More memory bandwidth
Performance • Reliability • Security • In-memory Database • Big Data
SPARC Future Work
2x application performance improvement every two years

Why SPARC Customers Choose SPARC
Top Reasons for Investing in Oracle SPARC Solaris Systems
1. Reliability: SPARC Systems have the best uptime
2. Performance: New Benchmarks demonstrate performance
3. Roadmap: Exciting path to future performance gains
4. Solaris: Industry’s best Operating System for Oracle and Partner
applications
5. Compatibility: Preserves investment in software to avoid costly
and complex migrations
6. Security: No-cost integrated software and hardware features
7. Single Vendor: Easier to deal with one company
8. Solutions: Better value from Oracle integration

Oracle
Engineered
Systems
Compute
Building
Blocks
One Engineering Team.
Best Performance. Best Value.

Appendix

Dynamic Threading
• Core resources are shared between the active threads
– Load-buffers, store-buffers, pick-queue, working-register-file, reorder-buffer, etc.
• Resources can be statically or dynamically allocated
• Dynamic allocation enables higher throughput by seamlessly adjusting
resourcing based upon thread behavior
– Applications scale better
– Increased throughput especially on heterogeneous workloads
• Thread hog:
If a thread occupies a resource, it must release the resource in a timely
fashion
– If not, thread is considered a “hog” and hardware limits the resources available to
it
– High and low watermarks for various core resources
– High watermark reached by allocation, low watermark by de-allocation
– Upon reaching high watermark, thread resource allocation stalls until low
watermark reached

#1 in Throughput response
Industry Applications
• 7.7x faster than Itanium
• 2.7x faster than Intel Nehalem
• 2.2x faster than legacy x86
• 4x reduction in server count versus x86
ASAP Activation
Service Broker
BRM Prepaid Billing
BRM Prepaid Billing

Consolidation on SPARC SuperCluster T4-4
T4-4 Node 1
Solaris 10 OVM
Solaris 11 OVM
T4-4 Node 2
Solaris 10 OVM
Solaris 11 OVM
T4-4 Node 3
Solaris 10 OVM
Solaris 11 OVM
T4-4 Node 4
Solaris 10 OVM
Solaris 11 OVM
ZFS STORAGE APPLIANCE
InfiniBand Network
EXADATA
STORAGE
EXADATA
STORAGE
EXADATA
STORAGE
EXADATA
STORAGE
EXADATA
STORAGE
EXADATA
STORAGE
1.2M IOPS
•1,200 CPU threads
•4 TB DRAM
•97 to 198 TB Hard Disk
•8.66 TB Flash
•42 GB/sec Storage
Bandwidth
•896 Gb/sec InfiniBand
Interconnect

Benchmark Disclosure
SPEC and the benchmark name SPECjEnterprise are registered trademarks of the Standard Performance Evaluation Corporation.
Other names may be trademarks of their respective owners. Competitive data obtained from http://www.spec.org as of the date
located next to the respective claim and this report. See the Website for latest results. SPARC T4-4 cluster: 40,104.86
SPECjEnterprise2010 EjOPS; 1,671 SPECjEnterprise2010 EjOPS performance per processor across the configuration. IBM
Power 780 and IBM Power 750 Express: 16,646.34 SPECjEnterprise2010 EjOPS; 1,387 SPECjEnterprise2010 EjOPS performance
per processor across the configuration. SPECjEnterprise2010 models contemporary Java-based applications that run on large Java EE
(Java Enterprise Edition) servers, backed by network infrastructure and database servers. The Application tier cost of acquisition for
four SPARC T4-4 servers with Solaris 10 is $ $467,856 or $11.67/SPECjEnterprise2010 EjOPS. Oracle pricing from
https://shop.oracle.com/ on 9/26/1011. The Application tier cost of acquisition for IBM Power 780
(3.86GHz Power7, 512GB RAM, AIX 7.1) is $1,297,956 or $77.97/SPECjEnterprise2010 EjOPS. IBM system pricing is from
http://tpc.org/results/FDR/TPCH/TPC-H_1TB_IBM780_Sybase-FDR.pdf, adjusted to license 64 cores (w/o TurboCore). AIX 7.1
pricing is from http://www-304.ibm.com/easyaccess3/fileserve?contentid=214347. $77.97/$11.67=6.7x. Oracle app. tier
configuration occupies 20RU of space, 40,104.86/20=2005 SPECjEnterprise2010 EjOPS/RU. IBM app. tier configuration
occupies 16RU of space, 16,646.34/16=1040 SPECjEnterprise2010 EjOPS/RU. 2007/1040=1.92x round nearest 2x.
Source: Transaction Processing Performance Council (TPC) www.tpc.org as of September 24, 2011. SPARC T4-4 server
(4 sockets/32 cores/256 threads) 201,487 QphH@1000GB, $4.60/QphH@1000GB, 50,371 QphH@1000GB/per socket,
available 10/30/11. IBM Power 780 Model 9179-MHB server (8 sockets/32 cores/128 threads) 164,747.2 QphH@1000GB,
$6.85 /QphH@1000GB, 20,593 QphH@1000GB per socket, available 3/31/11.
HP Integrity Superdome 2 server (16 sockets/64 cores/64 threads) 140,181 QphH@1000GB, $12.15/QphH@1000GB,
8,761 QphH@1000GB per socket, available 10/20/10. 50,371 QphH@1000GB per socket / 20,593 QphH@1000GB
per socket = 2.44
http://www.tpc.org/results/individual_results/Oracle/Oracle_T4-4_1TB_TPCH_ES_092611.pdf

1M IOPS: Based on internal measurement of Exadata Storage cells
10x Java performance: Based on internal measurement of Exalogic
ZFS 2x faster, ½ the price of NetApp: Demonstrates the performance of ZFS Storage via the Oracle Sun ZFS Storage 7420
appliance which delivered outstanding performance and price/performance on the SPC Benchmark 1, beating results published
on the NetApp FAS3270A.The Sun ZFS Storage 7420 appliance delivered 137,066.20 SPC-1 IOPS at $2.99 $/SPC-1 IOPS on
the SPC-1 benchmark.The Sun ZFS Storage 7420 appliance outperformed the NetApp FAS3270A by 2x on the SPC-1
benchmark.The Sun ZFS Storage 7420 appliance outperformed the NetApp FAS3270A by 2.5x on price/performance on the
SPC-1 benchmark.SPC-1, SPC-1 IOPS, $/SPC-1 IOPS reg tm of Storage Performance Council (SPC). More info
www.storageperformance.org. Oracle Sun ZFS Storage 7420 Appliance
http://www.storageperformance.org/results/benchmark_results_spc1#a00108 As of October 3, 2011
Cloud provisioning, unmatched scalability: Scalability of Solaris 11, leader in scaling to 512 threads
Near zero virtualization overhead: Solaris Zones, based on internal tests
InfiniBand: 5-8x speed of current networks
Enterprise Manager reduction of downtime: white paper including description of reduced downtime:
http://www.oracle.com/oms/enterprisemanager11g/application-to-disk-067846.html
Leading security:
--Comparison is based on internal testing of data warehousing queries that accessed tablespaces encrypted with Oracle
transparent data encryption(AES-256-CFB).
--Based on internal testing of ZFS on Oracle Solaris 11 Express 2010.11 using AES with key lengths of 256, 192, and 128 in the
CCM and GCM operation modes.
Performance Substantiation

Java Enterprise: TPC-H, QphH, $/QphH are trademarks of Transaction Processing Performance Council (TPC). For more
information, see www.tpc.org. SPARC T4-4 201,487 QphH@1000GB, $4.60/QphH@1000GB, avail 10/30/2011, 4 processors,
32 cores, 256 threads; SPARC Enterprise M8000 209,533.6 QphH@1000GB, $9.53/QphH@1000GB, avail 09/22/11, 16 processors,
64 cores, 128 threads; IBM Power 780 QphH@1000GB, 164,747.2 QphH@1000GB, $6.85/QphH@1000GB, avail 03/31/11,
8 processors, 32 cores, 128 threads; HP Integrity Superdome 2 140,181.1 QphH@1000GB, $12.15/QphH@1000GB avail 10/20/10,
16 processors, 64, cores, 64 threads.
WebCenter Content: (SuperCluster) 11x faster claim based on internal testing showing 8x cores of x86 ingesting approx
150 docs/sec compared to 8x cores of SuperCluster at 1700 docs/sec.
PeopleSoft 3x faster than Itanium: (T4) Compared to the best published PeopleSoft Enterprise Payroll 9.0 (non-UNICODE version)
result by HP, the SPARC T4-4 server result with PeopleSoft Enterprise Payroll 9.1 (UNICODE version) is 3.1 times faster that the
Itanium-based HP Integrity rx7640 server result of 96.17 minutes. HP has not published results with Unicode version of this benchmark.
Oracle Database 3x IBM P7: (SuperCluster) Based on internal testing of full rack SPARC SuperCluster T4-4 vs full rack
16 CPU IBM POWER 7 system.
Web 1M http: (SuperCluster) Based on extensive internal testing of Exalogic that was done during product development.
Security 5x v P7: (T4) Comparison is based on internal testing of AES-256-CBC encryption at 8K using OpenSSL against
published test results for IBM: http://xmlisnotaprotocol.blogspot.com/2010/10/openssl-098-benchmark-on-power7-35ghz.html.
Database Refresh: Based on TPC-H@1000GB benchmark result of 201,487 QphH@1000GB, the SPARC T4-4 server is up to
3.8 times faster than the IBM server for the Refresh Function.
More details at:http://www.oracle.com/us/solutions/performance-scalability/default-495351.html
Communications Billing: The SPARC T4-4 servers running the Oracle Communications Billing and Revenue Management
benchmark and Oracle Solaris Containers delivered 2.2x the performance and a 4x reduction in the number of servers, for
customers migrating from eight quad-core Intel Xeon E5335 servers and twelve dual-core AMD Opteron servers.
Java 2.2M JMS: (SuperCluster) Based on extensive internal testing of Exalogic that was done during product development.
PeopleSoft 2.8x faster than z10: (T4-4) Compared to the best published PeopleSoft Enterprise Payroll 9.0 (UNICODE version)
result by IBM, the SPARC T4-4 server result with PeopleSoft Enterprise Payroll 9.1 (UNICODE version) is 2.8 times faster that
the IBM z10 EC 2097 mainframe result of 87.4 minutes.
JD Edwards: (T4-2) JD Edwards - “Day in Life” online+batch - T4-2 (9.0.2 DIL) 2.5x faster than IBM P750 Power7
(9.0.1 DIL) - 10,000 users with sub-second response time
Database 2.3M IOPs: (SuperCluster) Based on extensive internal testing of Exadata that was done during product development.
Comms Service Broker: (T4-1) 2.7x more performance (400 CAPS) than an Intel Nehalem-based system (150 CAPS).
Security 3x v x86: (T4) Based on internal testing of ZFS on Oracle Solaris 11 Express 2010.11 using AES with key lengths of 256, 192,
and 128 in the CCM and GCM operation modes.
Data Warehousing: (SuperCluster) Based on internal testing of full rack SPARC SuperCluster T4-4 vs full rack 16 CPU HP Itanium
Superdome 2 system.
Performance Substantiation

Sparc solaris servers

More Related Content

What's hot

Similar to Sparc solaris servers

More from solarisyougood

Recently uploaded

Sparc solaris servers

Editor's Notes