Extending the Reach of Real-Time Analytics:

Hardware and Software, Engineered to Work Together

Oracle and Intel continue over 20 years of collaboration by innovating for the Oracle Exalytics In Memory
Machine, the industrys first engineered system for in-memory analytics. Deep optimizations at every
levelfrom individual components to the holistic systemdeliver extraordinary performance for analytics,
modeling, and planning that power brilliant business decisions.

Insight on Demand: The Oracle Exalytics In-Memory Machine

The worlds most successful businesses gain a competitive advantage from cutting-edge data analytics that
generate the insights behind their key decisions. On-demand intelligence is vital for driving value from massive
data sets being generated in real time, often in combination with existing data stores.
IT organizations face a variety
of challenges in delivering those
capabilities to their business users. The
technical complexity of building the
needed infrastructure can be daunting,
not to mention being incompatible
with budget and timeline pressures.
Moreover, many companies already
have fragmented silos of analytics
capability, often deployed by individual
business units, that add management
expense and act as barriers to
integration across the whole company.
Oracle Exalytics is engineered to
overcome those obstacles, enabling
the next generation of real-time data
analytics.
Extremely high performance. Time
to insight is dramatically reduced by
highly tuned hardware and software,
including a complete copy of data in
active memory to eliminate latency
effects from mechanical hard drives.
No more data silos. Consolidation
of departmental analytics into one
platform adds interoperability, so data
across the company can be correlated
to produce richer business insights.
Simplified environment. The
pre-integrated Oracle Exalytics
engineered system ensures optimized
design for IT and streamlines
implementation, helping lower total
cost of ownership.

Oracle and Intel have continued

their long history of co-engineering
and shared expertise to provide
groundbreaking capabilities in realtime data analytics. Oracle Exalytics
starts with state-of-the-art Oracle
software building blocks, deeply
optimized for the Intel Xeon

processor E7 v2 family by performance

engineers from Oracle and Intel to help
take full advantage of all the resources
available in the system. Integration,
pre-validation, and further optimization
at the engineered-system level delivers
performance, scalability, and reliability.

Driving Business Value with

Real-Time Analytics Use Cases
The Oracle Exalytics In- Memory Machine is purpose -built to generate
immediate insight from fast- changing big data. Improved business
decisions help identify revenue opportunities, avoid missteps, and create a
competitive advantage.
Strategic planning. On- demand modeling of hypothetical scenarios
can help guide decisions in areas such as prioritizing lines of business,
optimizing staff levels, and assessing potential acquisitions.
Dashboards. Correlating and visually presenting data in real time enables
executives and other business users to keep abreast of emerging trends
related to their specific needs and interests.
Demand forecasting. Predictive algorithms that model customer demand
for goods and services in advance enable companies to proactively allocate
resources, set production levels, and target marketing efforts.
Pricing optimization. Dynamically fine -tuning prices for goods and
services based on fluctuating inventory levels, per-transaction margins,
and sales volume can help increase revenue and net profit.
Supply- chain and production management. Real-time response to
shifting customer demand, resource availability, lead times, and production
capacity helps improve service levels and reduce excess inventory.

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Table of Contents
Insight on Demand: The Oracle
Exalytics In-Memory Machine . . . . . .1
Breakthrough Performance that
Drives Customer Success . . . . . . . . . . .2
Engine for Real-Time Analytics:
The Intel Xeon Processor E7
v2 Family . . . . . . . . . . . . . . . . . . . . . . . . . .3
Elastic Processor Model with
Specialized Features . . . . . . . . . . . . . .3
Targeted Optimizations for
Best-in-Class Oracle Software . . . . . .4
Oracle Essbase OLAP Server . . . . . .5
Oracle Database In-Memory . . . . . .6
Oracle Business Intelligence
Enterprise Edition (OBIEE) . . . . . . . .7
Real-Time Analytics, Taken
to the Next Level:
Engineered Systems . . . . . . . . . . . . . . . .7
Conclusion . . . . . . . . . . . . . . . . . . . . . . . . .8

Breakthrough Performance that

Drives Customer Success
Oracle Exalytics gives customers
the power to deliver fast, interactive
analytics, combining data sources from
all over the company. Visualizations
reveal patterns, trends, and insights
based on the data relationships most
important to individual job roles, all
at the command of ordinary business
users. And with Oracle Exalytics
In-Memory Machine X4-4, Oracle
and Intel have broken new ground in
performance that delivers benefits such
as the following:
Reports can be run more often, to
derive more business advantages
from data.
Interactive reporting can grow to
support new usages and more users.
More complex scenarios can be run
for what-if analysis.

Reporting can be done directly on

planning (read-write) data.
Batch jobs can be run on a daily basis
instead of on weekends, keeping
fresher results available.
Organizations across industry verticals
are finding that Oracle Exalytics inmemory database technology is a key
solution for addressing the challenges
in managing and taking advantage of
very large-scale data stores. Processes
that used to be performed in batches
are now handled in real time. Canned
reports that used to be created by IT
and run on a scheduled basis are being
replaced by immediate answers, even
to complex, ad hoc questions. And
business decisions are being improved
through the use of more sophisticated
insights, based on larger, more
complete sets of data.

Additional Resources . . . . . . . . . . . . . . .8

We needed a BI solution for

With Oracle Exalytics, we
the future, as data sets are
increased the cube history
growing fast and our users are
from two years to four years,
mobile. In order to compete
giving us access to additional
effectively in the global retail
historical data. And we can
industry, our business needs support more than 10,000 users
excellent performance for
with hundreds of concurrent
more advanced analysis
queries, without concern of
and fast decisions. Oracle
crashing the system.
Exalytics is a future-oriented
Kishore Agasthi, vice president
solution that we can grow
of information management,
Sodexo
with and handle analysis
on big data and forecasts
within seconds, delivered to
executives anywhere.
Torsten Balslev, CIO, Nilson Group AB
2

Upgrading to Oracle
Essbase on Oracle Exalytics
In-Memory Machine has at
least tripled the speed at
which we can generate and
employ business intelligence,
enhancing our management
reporting capabilities. And
our collaboration with Oracle
Advanced Customer Support
Services has ensured access
to continuous, easy-to-budget
technical expertise to support
our growth.
Martin Jacobs, Project Manager,
Application Development,
Deutsche Telekom AG

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Engine for Real-Time Analytics: The Intel Xeon Processor E7 v2 Family

The scale-up computing capabilities of the Intel Xeon processor E7 v2 family, illustrated in Figure 1, are an excellent fit for
the demands of real-time data analytics. Compared to its predecessors, the platform has 1.5x as many cores, for a total of 60
(120 threads) in a four-socket Oracle Exalytics system. Together with 1.25x as much cache (up to 37.5 MB per socket), these
increased execution resources allow for fast processing of large-scale calculations, an essential requirement for data analytics,
as well as the ability to generate advanced visualizations to make insights more actionable.

Execution Resources

Enhanced Memory

Advanced Reliability

50% more cores

(up to 15/socket)

3x memory capacity
(up to 1.5TB/socket)

Intel Run Sure Technology

to help reduce downtime2

25% more cache

(up to 37.5 MB/socket)

Up to 4x increased bandwidth
with integrated PCI Express* 3.01

Resilient Memory Technologies

to help ensure data integrity

Supports holding massive

data sets in memory

Enhances suitability
for mission-critical solutions

Provides headroom for large-scale,

real-time computation

Figure 1. Advances in the Intel Xeon processor E7 v2 family for real-time, in-memory analytics.

The processor also introduces

enhancements to the memory
subsystem that improve its capability
to hold huge sets of business data in
memory, near the execution pipeline, a
vital factor in ensuring extremely high
performance. Capacity for up to 1.5 TB
of RAM per socket is an increase of 3x
over its predecessor. Complementing
the platforms large memory capacity,
integrated PCI Express* 3.0 increases
I/O bandwidth up to 4x,1 facilitating fast
movement of data within the system.
Because making the best decisions
possible is vital to the health of any
company, mission-critical dataanalytics workloads are protected by
the advanced reliability, availability,
and serviceability (RAS) features
of the Intel Xeon processor E7 v2
family. Intel Run Sure Technology, 2

Elastic Processor Model with

Specialized Features

needs of different workloads. This

processor combines the capabilities
of the three highest-performing
members of the Intel Xeon processor
E7 v2 family, providing the innovative
capability to run cores at higher
frequency when fewer of them are
active, as shown in Figure 2. This
elasticity provides an optimized
balance between thread count and
frequency for specific workloads.

A true breakthrough in co-engineering,

the Intel and Oracle teams worked
together to create an elastic enterprise
workload SKU: the Intel Xeon
processor E7-8895 v2. This elastic
SKU, delivered in conjunction with
specialized Oracle software, enables
the Exalytics to dynamically vary core
counts and frequencies to meet the

High-concurrency workloads, such

as a large number of concurrent
users executing planning and analysis
queries, benefit from execution on
all available cores. The needs of such
highly parallel loads are met with
all 60 of the systems cores (15 per
socket) running simultaneously, at the
processors base frequency of 2.8 GHz.

newly introduced with this processor

generation, integrates processor,
firmware, and software capabilities to
help diagnose, contain, and recover
from otherwise-fatal errors. Resilient
Memory Technologies help ensure data
integrity within the memory subsystem,
for more reliable operation.

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Variable Active Core Count and Frequency to Meet the Needs of Changing Workloads
15 cores (30 threads)

10 cores (20 threads)

6 cores (12 threads)

2.8 GHz

3.2 GHz

3.6 GHz

Intel Xeon Processor E7-8895 v2

Fixed Core Count

and Frequency

Fixed Core Count

and Frequency

Fixed Core Count

and Frequency

15 cores (30 threads)

10 cores (20 threads)

6 cores (12 threads)

2.8 GHz

3.2 GHz

3.4 GHz

Intel Xeon Processor

E7-8890 v2

Intel Xeon Processor

E7-8891 v2

Intel Xeon Processor

E7-8893 v2

Figure 2. Flexibility comparable to choice among the capabilities of three high-end processors.

Intensive per-thread workloads,

such as complex Oracle Essbase
Online Analytical Processing (OLAP)
operations, run relatively fewer
threads, leaving some cores inactive.
In this scenario, 20 to 24 active cores
(5 to 6 per socket) increase their
frequency to 3.6 GHz, providing
increased resources for each
individual operation and reducing
the overall execution time.
Mixed workloads are supported by
dynamically switching the system to
an intermediate state, activating 40
cores (10 per socket) and setting the
processor frequency to 3.2 GHz.
The balance between concurrency
and frequency can be handled
automatically as needs dictate, or it
can be set explicitly by the application.
This innovation lets Oracle Exalytics
customers improve on the otherwisetypical commitment to a set core

count and frequency when choosing

a processor model. Together, this
combination of hardware and software
simplifies the process of determining
system configurations and purchasing,
while allowing dynamic repurposing of
assets.
Co-engineering work by Intel and
Oracle in the development of Oracle
Linux* provides enhanced support for
the dynamic active-core count and
operating frequency of the Intel Xeon
Processor E7-8895 v2. The full range
of optimizations helps ensure that the
Oracle Exalytics In-Memory Machine
benefits businesses with powerful, realtime decision making.
For full configuration details of the
Oracle Exalytics In-Memory Machine
X4-4, see the Oracle Data Sheet at
http://www.oracle.com/us/products/
applications/exalytics-x24-1500259.pdf.

Targeted Optimizations for

Best-in-Class Oracle Software
Oracle Exalytics is based on a set
of complementary Oracle software
elements, integrated to work together
as a unified whole that delivers even
greater value than the sum of its
considerable parts.
Oracle Essbase OLAP Server.
Powerful analytical processing engine
for business users to model fastrunning, complex what if scenarios.
Oracle Database In-Memory. Inmemory operation reduces latency
and increases throughput compared
to traditional relational databases on
conventional hard drives. It supports
column format, which is ideal for
analytics, as it allows faster retrieval of
specific columns of large data sets.

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Oracle Exalytics In-Memory

Software. A robust collection of
in-memory features, optimizations,
and configurations enables the BI
suite to take full advantage of the
memory architecture of the Intel Xeon
processor E7 v2 family. The Exalytics
In-Memory Software includes OBIEE
Accelerator, Essbase Accelerator,
In-Memory Data Caching, and BI
Publisher Accelerator.
Oracle and Intel engineers have been
laying the optimization groundwork
for Oracle Exalytics to take full
advantage of this processor platform
since long before it became available
to the general public. Building on the
processors large-scale execution
resources, enhanced memory
subsystem, and advanced reliability,
the team rigorously optimized each
part of the software stack to scale well
across the platforms execution cores
and large amounts of RAM and non
volatile memory.
Oracle Essbase OLAP Server
Oracle Essbase, the OLAP server for
Oracle Exalytics, provides a ready
environment for deploying existing
shrink-wrapped applications and
developing custom ones. Oracle
Essbase provides the basis for
organizations to enable business users
with robust, autonomous capabilities
to model complex business scenarios
using straightforward end-user tools,
including integration with familiar
productivity tools such as Microsoft
Office*.

Oracle and Intel engineers made

a series of optimizations to Oracle
Essbase specifically for the Intel Xeon
processor E7 v2 family that is the
foundation of the Oracle Exalytics
engineered system.

revealed excellent scalability for up to

30 users, after which the number of
transactions per second leveled off and
then began to fall off at 50 users and
above, as the Before Optimization line
shows.

Accelerated performance hotspots .

Using the Intel VTune Amplifier, the
team identified areas in the Oracle
Essbase code that consumed the most
processor resources. Focusing on
those sections, the team hand-tuned
code and applied new techniques to
accelerate performance.

Hotspot analysis with the Intel VTune

environment identified two spinlocks
EsSdSpinlockAcquireNonReentrant
and EsSdSpinlockAcquirethat
were using increasing numbers of
CPU clock ticks per transaction as the
number of users grew. Both these locks
were highly contended because they
protected large, frequently accessed
critical sections of code. As a result,
execution threads on the processor
spent significant amounts of time
spinning to wait for locks, instead of
performing useful work, as they needed
the access to locked code before they
could proceed.

Increased parallelism. Stress testing

revealed potential instances where
Oracle Essbase didnt take full
advantage of all processor cores
available. Tuning with the aid of Intel
VTune Amplifier, the team ensured
that Oracle Essbase maximized the
number of concurrent users and
throughput to support scaling to 60
cores and beyond while maintaining
fast response times.
To test scalability as the number of
users increases, the engineering teams
used a load generator to send requests
(a mixture of queries and updates)
directly to Oracle Essbase. The results
are shown in Figure 3. 3 Initial results

ForEsSdSpinlockAcquireNonReentrant,
the critical section was split into smaller
sections (reducing the size of the critical
section) and a condition was added on
lock acquisition (reducing the frequency
of locking). For EsSdSpinlockAcquire,
some lock usages were identified that
could be eliminated.

50
45
Transactions per Second

Oracle Business Intelligence

Enterprise Edition (OBIEE).
Comprehensive, easily integrated BI
platform for dashboards, alerting, ad
hoc queries, reporting, and more.

40
35
30
25
20

Before Optimization
After Optimization

15
10
5
0
10

20

30

40

50

60

Number of Users
Figure 3. Scalability optimization of Oracle Essbase OLAP Server. 3

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Optimization of the two spinlocks resulted in dramatic improvements to scaling as the number of users increases, as the After
Optimization line in Figure 3 shows. The system delivers faster response times and higher performance, at the same CPUutilization levels, when scaling to higher number of users after optimization. 3 Another aspect of the performance improvement
attained through the optimization is the effect on throughput stability, shown in Figure 4, using 40 users as an example. The
Before Optimization chart shows the level of transactions per second varying significantly over time as the test scenario
proceeded. In contrast, in the After Optimization chart, the level of transactions per second is far more stable.

Before Optimization

Total Number of Transactions

45

Total Transactions per Second

40
35
30
25
20
15
10
5
0

0:00 0:10 0:20 0:30 0:40 0:50 1:00 1:10 1:20 1:30 1:40 1:50 2:00 2:10 2:20 2:30
Elapsed Scenario Time (hh:mm)

After Optimization

Total Number of Transactions

45

Total Transactions per Second

40
35
30
25
20
15
10
5
0

0:00 0:10 0:20 0:30 0:40 0:50 1:00 1:10 1:20 1:30 1:40 1:50 2:00 2:10 2:20 2:30
Elapsed Scenario Time (hh:mm)

Figure 4. Optimized run stability of Oracle Essbase OLAP Server with 40 users. 3

Oracle Database In-Memory

Oracle Database In-Memory (Oracle
DBIM) delivers leading-edge inmemory performance without the
need to restrict functionality or
accept compromises, complexity,
and risk. With a data structure that
resides entirely within active memory,
Oracle DBIM accelerates database
performance by orders of magnitude
for analytics, data warehousing, and
reporting, without requiring application
changes. Certified to run on the Oracle
Exalytics In-Memory Machine, Oracle
DBIM brings the robustness and
maturity of Oracle Database 12c to the
environment. Enhancements that are
specific to Oracle Exalytics In-Memory
Software help take full advantage of
the capabilities of Oracle DBIM.
Innovative algorithms for in-memory
scans, joins, and aggregation enable
Oracle DBIM to run queries at the rate
of billions of rows per second for each
CPU core. Columnar compression helps
reduce the memory footprint of data.
The database is capable of operating
on data in its compressed form, which
reduces the size of the data set that
scanning and filtering operations
execute against. The database is also
capable of allowing fast queries without
decompression overhead. Data is
decompressed only when it is required
for the result set.
To help Oracle DBIM take full advantage
of the memory enhancements in the
Intel Xeon processor E7 v2 family, the
team used the Intel VTune Amplifier,
identifying a number of performancetuning opportunities. Using the Intel

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Compiler and the Intel Advanced

Vector Extensions (Intel AVX4)
instruction set, the team optimized
memory-utilization efficiency,
helping Oracle DBIM deliver further
performance advantages to the Oracle
Exalytics platform as a whole.
Oracle Business Intelligence
Enterprise Edition (OBIEE)
A full spectrum of business intelligence
capabilities, such as reporting,
dashboards, ad hoc analysis, data
visualization, and forecasting,
are provided by Oracle Business
Intelligence Enterprise Edition (OBIEE)
for Oracle Exalytics. The platform
enables business users to more easily
access and create value from all types
of data, while its standards-based
design offers a service-oriented
architecture that integrates readily with
existing IT infrastructure.
Profiling the OBIEE code using the Intel
VTune Amplifier, the engineering team
found hotspots that helped them focus
their optimization efforts where they
would have the most benefit. Those
tuning efforts paid off in increased
performance that helps accelerate
Oracle Exalytics further. Specifically,
the OBIEE team used the statistics
gathered from the Intel VTune analyzer
to improve a wide range of software
capabilities, including the following:5
A 20-percent reduction in the time it
takes for the server to start
A 3x reduction in time requirements
for consistency checker (which
validates code written by application
developers)
A 5x reduction in the time it takes
developers to check their code into
OBIEE
These improvements led to
dramatically improved productivity for
Oracles 150+ application developers.

Real-Time Analytics, Taken to the

Next Level: Engineered Systems
Engineering at the system level adds
dramatically to the performance
and capabilities of Oracle Exalytics,
beyond the optimizations to the
building blocks. The engineered
system is the product of extensive
expert development, purpose-built
and highly tuned for real-time data
analytics use cases. It is based on a
carefully selected, tested, and validated
combination of components, drawing
on the expertise of the engineering
teams from Oracle and Intel.
That added level of engineering helps
ensure that customers get the best
insights possible from real-time data
analytics. Specific business benefits
from the system-level engineering of
Oracle Exalytics include the following:
System-level optimizations
(including the use of specialized
components described below) build
on optimizations to the software
building blocks, for performance
that enhances value from capital
expenditure.
Consolidation of analytics solutions
into a single platform simplifies
management of the environment
(lowering operational expenditures)
and breaks down silos to improve
interoperability.
Comprehensive, unified solution
stack without gapsmade of pieces
engineered to work together
enhances flexibility and reliability for
business-critical analytics workloads.

Oracle Engineered
Systems for Big Data
The Oracle Exalytics InMemory Machine is part of
Oracles portfolio of engineered
systems for big data, all of which
are purpose -built with pre integrated stacks of hardware
and software, engineered to
work together.
Optimizations throughout the
stack at the component and
system levels enable Oracle
engineered systems to drive up
performance and productivity,
while reducing complexity
and cost.
Oracle Exadata Database
Machine delivers very
high performance for data
warehousing, database
consolidation, and online
transaction processing (OLTP)
applications.
Oracle Exalogic Elastic
Cloud is designed and highly
optimized for mission - critical
middleware and applications
from both Oracle and thirdparty suppliers.
Oracle Exalytics In- Memory
Machine features a pre optimized combination of
in-memory software, hardware,
and a business intelligence
platform with advanced
visualization.
Oracle Big Data Appliance
is a highly tuned, purpose built engineered system for
acquiring, organizing, and
loading unstructured data into
Oracle Database.

Extending the Reach of Real-Time Analytics: Hardware and Software, Engineered to Work Together

Conclusion

Additional Resources

As an engineered system, Oracle Exalytics delivers cost

benefits in addition to improving business operationspreintegrated hardware and software reduces IT complexity
and accelerates time to production, while the single-supplier
solution helps reduce implementation and support costs.

Oracle Engineered Systems: Products and Resources

www.oracle.com/us/products/engineered-systems

With their joint work on the Oracle Exalytics In-Memory

Machine X4-4, Oracle and Intel continue their enduring
tradition of co-engineering. As a result, the solution delivers
extraordinary performance through deep optimizations
throughout the stack and at the engineered-systems level, as
well as tight integration between hardware and software.
That co-engineering lies at the core of improving the quality
of business decisions with real-time analytics, positioning
customers for success.

Oracle and Big Data: The Power to Transform Your Business

www.oracle.com/us/technologies/big-data
Intel IT Center: Big Data Analytics Begins with Intel
www.intel.com/bigdata
Oracle Exalytics In-Memory Machine X4-4 Data Sheet
www.oracle.com/us/products/applications/exalytics-x24-1500259.pdf
Intel Big Data: Turning Big Data into Intelligence www.intel.com/
content/www/us/en/big-data/big-data-analytics-turning-big-datainto-intelligence-cmpg.html?cid=sem116p29477g-c&gclid=CLysps
rhrr4CFc9AMgodr3UAng
Intel and Oracle Webcast Series: Big Data at Work http://event.
on24.com/r.htm?e=654398&s=1&k=330C7590A2893D5DC4B19C1
F79407973&partnerref=OnDemandWebcasts

Learn more about the Intel Xeon processor E7 v2 family at www .intel .com/xeon
Learn more about the Oracle Exalytics In-Memory Machine at www .oracle .com/exalytics

Up to 4x I/O bandwidth claim based on Intel internal estimates of the Intel Xeon processor E7-4890 v2 performance normalized against the improvements over dual-IOH Intel Xeon processor
E7-4870 based on internal bandwidth tool running the 1R1W test.
No computer system can provide absolute reliability, availability or serviceability. Requires an Intel Xeon processor E7-8800/4800/2800 v2 product families or Intel Itanium 9500 seriesbased system (or follow-on generations of either.) Built-in reliability features available on select Intel processors may require additional software, hardware, services and/or an internet
connection. Results may vary depending upon configuration. Check with your system manufacturer.
3
Tests document performance of components on a particular test, in specific systems. Differences in hardware, software, or configuration will affect actual performance. Consult other sources of
information to evaluate performance as you consider your purchase.
Testing conducted using Oracle Hyperion Public Sector Planning and Budgeting Essbase application (PSB); 12 dimensions with 100k members on the largest dimension; Size: 25 GB compressed
on disk or 120 GB when uncompressed.
Server: Intel Xeon processor E7-4890 v2 @ 2.80 GHz, LLC 38400 KB, 15 cores per sockets with four sockets, 60 cores in total, 512 GB RAM, Red Hat Enterprise Linux* 6.4 64-bit, Essbase
11.2.3.500 GA build 64-bit.
Client: Intel Xeon processor E5440 @ 2.83 GHz, L2 cache 12 MB, 4 cores, Windows Server* 2003, HP LoadRunner* 8.1.2.
4
Intel Advanced Vector Extensions (Intel AVX)* are designed to achieve higher throughput to certain integer and floating point operations. Due to varying processor power characteristics,
utilizing AVX instructions may cause a) some parts to operate at less than the rated frequency and b) some parts with Intel Turbo Boost Technology 2.0 to not achieve any or maximum turbo
frequencies. Performance varies depending on hardware, software, and system configuration and you should consult your system manufacturer for more information. *Intel Advanced Vector
Extensions refers to Intel AVX, Intel AVX2 or Intel AVX-512. For more information on Intel Turbo Boost Technology 2.0, visit http://www.intel.com/go/turbo.
5
Testing was conducted using Oracle Business Intelligence Enterprise Edition and Admintool running on metadata from Oracle Business Intelligence Applications metadata (41Mb compressed,
300Mb uncompressed). The hardware used was a server with a Intel Xeon CPU X5675 and 8GB of RAM running 64-bit Window 8.1 Enterprise.
This document is provided for information purposes only and the contents hereof are subject to change without notice. This document is not warranted to be error-free, nor subject to any
other warranties or conditions, whether expressed orally or implied in law, including implied warranties and conditions of merchantability or fitness for a particular purpose. Oracle specifically
disclaims any liability with respect to this document and no contractual obligations are formed either directly or indirectly by this document. This document may not be reproduced or
transmitted in any form or by any means, electronic or mechanical, for any purpose, without our prior written permission.
Intels compilers may or may not optimize to the same degree for non-Intel microprocessors for optimizations that are not unique to Intel microprocessors. These optimizations include SSE2,
SSE3, and SSSE3 instruction sets and other optimizations. Intel does not guarantee the availability, functionality, or effectiveness of any optimization on
microprocessors not manufactured by Intel. Microprocessor-dependent optimizations in this product are intended for use with Intel microprocessors.
Certain optimizations not specific to Intel microarchitecture are reserved for Intel microprocessors. Please refer to the applicable product User and
Reference Guides for more information regarding the specific instruction sets covered by this notice.
Intel technologies may require enabled hardware, specific software, or services activation. Check with your system manufacturer or retailer.You may not use
or facilitate the use of this document in connection with any infringement or other legal analysis concerning Intel products described herein. You agree to
grant Intel a nonexclusive, royalty-free license to any patent claim thereafter drafted which includes subject matter disclosed herein.
Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1-800-5484725, or go to: www.intel.com/design/literature.htm.
Software and workloads used in performance tests may have been optimized for performance only on Intel microprocessors. Performance tests, such
as SYSmark* and MobileMark*, are measured using specific computer systems, components, software, operations and functions. Any change to any of
those factors may cause the results to vary. You should consult other information and performance tests to assist you in fully evaluating your contemplated
purchases, including the performance of that product when combined with other products. For more information go to http://www.intel.com/performance.
Copyright 2014 Intel Corporation. All rights reserved. Intel, the Intel logo, VTune, and Xeon are trademarks of Intel Corporation in the U.S. and other
countries. Oracle and Java are registered trademarks of Oracle and/or its affiliates. *Other names and brands may be claimed as the property of others.
0914/JR/MESH/PDF
330669-001US
1