T-Systems creates cloud-based big data solution with Intel® Xeon

T-Systems creates cloud-based big data solution with Intel® Xeon
CASE STUDY
Intel®Xeon®Processor E3 Product Family
Big Data Analytics
Cloud Computing Solutions
T-Systems creates cloud-based big data solution
with Intel® Xeon® processor-based servers
The Intel®Xeon®processor E3 product family enables T-Systems to bring big data to smaller businesses
Companies now realize that the data passing through their businesses can have strategic value. To help them unlock it, T-Systems launched
a new cloud-based big data analysis service, based on the Intel Xeon processor E3 product family.
CHALLENGES
• Enable big data: T-Systems needed to deploy a cloud-based big data solution for its customers, with minimal operating costs.
SOLUTIONS
• Intel-based servers: T-Systems deployed 95 physical servers based on the Intel® Xeon®
processor E3 product family, Intel® Server Board S1200BTLR and 1 Gigabit Intel® Ethernet
Converged Network Adapters.
• Big data software: Hadoop*, running on the CentOS* operating system, was deployed for
big data analysis.
TECHNOLOGY RESULTS
• Energy efficiency: T-Systems’ own tests revealed that a single Intel Xeon processor E3
product family consumes one-third less power than the previous technology, while almost
matching its performance.
• Energy optimized: T-Systems has optimized its server power consumption to less than
100W at full load.
• Lower costs: T-Systems’ own tests found that the new processor lowers its total cost of
ownership so it can pass savings on to its customers.
“We built our big data infrastructure
on the Intel® Xeon® processor E3
product family and Intel® Server
Boards because our own research
showed they have higher energy
efficiency and a lower total cost
of ownership.”
Sven Löffler
Business Development Executive
BI & Big Data
T-Systems
BUSINESS VALUE
• New services: T-Systems has extended its offering to existing clients through the launch
of the cloud-based big data analysis service, which enables customers to cost-effectively
analyze complex and large data sets.
• New customers: T-Systems can now work with customers that could not previously afford
big data analysis services.
The rise of big data
Across all sorts of industries, companies are
starting to realize that data they’ve long neglected can have strategic value. The discipline
of big data is all about taking this information,
typically unstructured, and creating meaning from it, so businesses can make better
decisions, create efficiencies and, in some
cases, invent new services and products.
In the automotive industry, for example, data
from network-enabled vehicles can deliver
insights into driver behavior and vehicle
performance. In procurement, having more
real-time data can help to inform better
buying decisions in terms of both price and
timing. For marketers, a real-time understanding of how a campaign is performing
makes it possible to steer the campaign in
the right direction, redirecting advertising
resources and refining campaign materials
in line with the market segment’s response.
Media organizations can become intelligent
news organizations if they can mine their
archived content to find valuable insights.
It is these four use cases that T-Systems
wanted to address when it started to offer
big data analytics solutions to its customers,
adding to the SAP*-based business intelligence solutions it already provides. Headquartered in Frankfurt, T-Systems provides
integrated IT services and consultancy to an
international client base. It employs over
50,000 staff, and generated over EUR 10
billion of revenue in the 2012 financial year.
T-Systems’s own research shows it cut server
power consumption by one-third while maintaining
constant performance
T-Systems relies on being able to offer innovative, technology-led solutions that successfully address the practical requirements of
customers and help support business growth.
Its new big data solutions will be delivered as
cloud services, opening up big data capabilities to customers who could not previously
afford them.
T-Systems chooses Intel
To enable the new cloud-based big data services, T-Systems deployed 95 physical servers
based on the Intel Xeon processor E3 product
family, Intel Server Board S1200BTLR, and
1 Gigabit Intel Ethernet Converged Network Adapters. Each node has six two-terabyte disks for storage.
These servers run the CentOS operating
system and the Cloudera* Hadoop implementation. T-Systems developed its own
Linux*-based software for software deployment, patch management, and hardware
monitoring. According to T-Systems, the new
solution can support a total of 1.2 petabytes
of data, with 400 terabytes of usable data
at any one point due to how the Hadoop
platform manages data.
“We built our big data infrastructure on the
Intel Xeon processor E3 product family and
Intel Server Boards because our tests
showed it had a lower total cost of ownership
and higher energy efficiency and we knew it
had a strong track record, having been available for some time,” said Sven Löffler, business development executive for business
intelligence and big data at T-Systems.
Making a big splash with big data
The type of data being analyzed in the new
architecture varies depending on the specific use cases required by end customers,
and includes both structured and unstructured data. It includes location and performance information sourced from vehicles;
social media and other written online content; archived news articles, video and audio
reports; and supplier, product, and market
segment data.
Customers are delighted with the offering,”
said Löffler. “They particularly appreciate
the low ramp-up time and the high cloud
availability and security standards, which
ensure their data is available for them when
they need it and protected from unauthorized access. We are targeting both new and
existing customers with this new service
and are excited by the opportunity we have
to bring big data analytics to customers who
could not previously afford it.”
T-Systems carried out its own research, using
the LinX* benchmark to compare the singlesocket Intel Xeon processor E3 product family
(based on 22nm architecture) with two of the
previous processors T-Systems had been using,
the Intel Xeon processor 5000 series (based
on 32nm architecture). “We found we could
achieve nearly the same performance with
just one of the new processors, compared
to two of the older processors, and yet still
cut power consumption by one-third at maximum load. Our further Hadoop benchmarks
correlated with these results. We’ve been
able to optimize the power consumption of
the setup to less than 100W at full load with
12 terabytes of hard disk space,” said Löffler.
Lessons learned
By using servers based on the Intel Xeon
processor E3 product family, T-Systems
was able to use a single processor in place
of two and cut energy consumption by
one-third while maintaining near-constant
performance levels. The new servers have
enabled T-Systems to cost-effectively
implement big data solutions so it can
expand its business with new services to
existing clients and reach new clients
who could not previously afford big data
analysis.
He adds: “The new platform enables us to
deliver insights much faster and with greater
potential to scale up to serve a larger customer base. It enables us not only to develop
new analytics services, but also to implement
them with a low total cost of ownership, so
we can offer them to customers at a lower
price point and expand our potential customer base. We don’t foresee any limits on
how many customers we can simultaneously
serve from our new big data infrastructure.”
Find the solution that’s right for your organization. Contact your Intel representative,
visit Intel’s Business Success Stories for IT
Managers (www.intel.co.uk/Itcasestudies)
or explore the Intel.co.uk IT Center
(www.intel.co.uk/itcenter).
Copyright 2013 Intel Corporation. All rights reserved. Intel, the Intel logo, Intel Xeon, and Xeon inside are trademarks of Intel Corporation in the U.S. and other countries.
This document and the information given are for the convenience of Intel’s customer base and are provided “AS IS” WITH NO WARRANTIES WHATSOEVER, EXPRESS OR IMPLIED,
INCLUDING ANY IMPLIED WARRANTY OF MERCHANTABILITY, FITNESS FOR A PARTICULAR PURPOSE, AND NONINFRINGEMENT OF INTELLECTUAL PROPERTY RIGHTS. Receipt or
possession of this document does not grant any license to any of the intellectual property described, displayed, or contained herein. Intel® products are not intended for use in
medical, lifesaving, life-sustaining, critical control, or safety systems, or in nuclear facility applications.
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
Intel does not control or audit the design or implementation of third party benchmark data or Web sites referenced in this document. Intel encourages all of its customers to visit
the referenced Web sites or others where similar performance benchmark data are reported and confirm whether the referenced benchmark data are accurate and reflect performance of systems available for purchase.
*Other names and brands may be claimed as the property of others.
0813/JNW/RLC/XX/PDF
329410-001EN
©
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertising