FAST FACTS

Industry: Construction & Engineering

Geography: Headquarters and global operations is based in the Principality of Liechtenstein

Business Challenge: To migrate all SAP® business-critical applications from a
discontinued legacy UNIX environment to a scalable and reliable platform and to eliminate vendor
lock-in

Migration Path: HP Tru64 UNIX on Alpha Servers to Red Hat Enterprise Linux 4 on x86_64
commodity based hardware

Software: Red Hat Enterprise Linux 4, Red Hat Cluster Suite, Red Hat Global File System
(GFS), ATIX Open-Sharedroot Extension, SAP applications including SAP Business Suite, SAP ERP and SAP Customer Relationship Management (SAP CRM), and the SAP NetWeaver® technology platform

Hardware: 185 HP ProLiant servers, the largest machines have 32 CPUs with 128GB RAM

Benefits: Eliminated vendor lock-in; achieved an estimated 50 percent overall cost reduction; increased performance by more than 100 percent; provided ease of management and reliable uptime; reduced electricity costs and carbon footprint; and provided the company with a long-term platform strategy that will help retain expert knowledge and enable the team to be highly responsive to the increasing requirements of a global manufacturing, sales and service organization

Download the PDF case study This case study is also available in: German.

“We migrated to SAP applications on Red Hat Enterprise Linux on HP ProLiant servers after evaluation and testing. Since migrating, we have have experienced increased performance of more than 100 percent. We also reduced IT costs with a commoditized architecture. Having a tight level of integration between Red Hat and SAP support organizations gives us the confidence to continue along this path and further reduce our costs by migrating all of our SAP environment to Red Hat Enterprise Linux.”
– Martin Petry, CIO at Hilti

BACKGROUND
Hilti Corporation, headquartered in Schaan in the Principality of Liechtenstein, is a world-renowned manufacturer of leading-edge technology for the global construction industry. Hilti’s high-performance drills, chisels, cutters, fastening, and measuring systems are used by construction workers around the world. The company’s sales and service organization of 20,000 global employees works directly with customers in more than 120 countries around the world, handling more than 200,000 customer contacts every day.

Hilti boasts an internal IT center, including its own in-house SAP solution-based landscape, and places a strategic focus on technology leadership and innovation.

BUSINESS CHALLENGE
An integral part of Hilti’s company philosophy is to sell directly to end customers and provide outstanding service at construction sites worldwide. The company’s sales and service department make up two thirds of its worldwide staff. Besides high product quality and the constant innovation of Hilti’s engineers, the responsiveness and flexibility of its sales and service organization are the main factors that make Hilti stand out from the competition and have ensured the company a loyal customer base.

As part of a family-owned company with a philosophy of sustainable and long-term investment decisions, Hilti’s IT organization aims to meet business requirements with agility. Fundamental investment decisions in IT are made with two key requirements in mind. First, a technology or platform choice needs to be scalable to allow for growth, be able to meet business challenges, and adapt to changes the organization will encounter over the next 15 to 20 years. Second, the platform choice must enable Hilti to retain necessary expertise to manage and develop systems over as many years as possible.

In the early 1990s, Hilti chose to build its mission-critical IT infrastructure on HP’s Tru64 UNIX operating system running on Alpha Servers. With this decision, the company was one of the first global operations to migrate to a 64-bit operating system.

In 2004, HP announced it would discontinue development and support for Tru64 UNIX and Alpha Server. This vendor decision required Hilti to develop a new sustainable technology roadmap and select a software and hardware combination that would comply with its 15-20 year strategy for technology investments.

“We were experiencing a classic case of vendor lock-in and the expertise our team had built over many years was suddenly depreciating rapidly,” explains Michael Hagmann, head of Enterprise Server Technology at Hilti. “Trying to extend the lifecycle of the Tru64/Alpha platform was not an option, as we would quickly run into maintenance and hardware issues. We started evaluating alternative platforms with our previous experience in mind.”

SOLUTION
When considering alternative platforms, Hilti’s enterprise server team assessed open source software from the start. The process started in 2005, shortly after HP’s end of life decision. Initially, Hilti wasn’t sure if x86-based hardware would be capable of handling the large amounts of data and tens of thousands of daily transactions its daily business produced. But the prospect of avoiding vendor lock-in completely by building the new infrastructure on open source software made Linux Hilti’s preferred operating system.

Hilti’s enterprise server team had only three months to evaluate hardware and software alternatives before making the final investment recommendation to the Executive Board. A strong argument in favour of choosing Red Hat was that all of Hilti’s application vendors have certified their products to run on Red Hat Enterprise Linux, which boasts an ecosystem of over 3,000 certified software applications. Hilti uses a broad array of SAP applications. Its largest and most critical systems rely on SAP ERP and SAP CRM, each with more than 5TB of data stored.

Hilti’s enterprise server team wanted to continue managing its clustered systems as one single “root disk.” An add-on called “Open-Sharedroot” from Munich-based ISV and consulting company Atix, which specializes in Linux environments and complex clustering projects, provided that for Red Hat Enterprise Linux.

“In the end, we had the perfect partners for our migration,” said Hagmann. “We had Red Hat Enterprise Linux, the possibility to continue working with a shared root cluster after the migration, plus the commitment from Atix and Red Hat to support our project from start to finish. A migration of this scale had never been done before anywhere, and many believed we would face challenges, but we were confident that the solution’s benefits and performance were enterprise-ready.”

As an SAP customer, Hilti wanted to consolidate, standardize, and expand its SAP software environment to improve its business performance and enhance its systems reporting capabilities.

“Our business-critical systems like SAP ERP and SAP CRM are all centralized at our headquarters,” explained Hagmann. “Our sales and customer service employees around the world rely on these SAP applications to be up and running 24/7, so migration-related downtime was not an option.” Hilti started with moving less mission-critical applications to the new platform in winter 2006/2007. SAP CRM was migrated at the end of 2008 and has been fully
operational and stable since January 2009. As the last step, SAP ERP was migrated and ready for production in May 2009.

ATIX and Red Hat supported Hilti’s migration with a dedicated Technical Account Manager and two members of support staff who had access to duplicate test systems at Red Hat and guaranteed quick responses to support tickets. “With such a large-scale, and mission-critical migration, we built our relationship with Red Hat early and our decision to utilize a Technical Account Manager saved a considerable amount of time,” said Hagmann. “ATIX and Red Hat’s dedication to making our lives easier and pulling this huge migration project off together was an indispensable asset to us.”

BENEFITS
With the new enterprise server environment, Hilti’s business-critical IT infrastructure is scalable and vendor-independent. A key benefit of the migration is that the knowledge about the Red Hat Enterprise Linux-based infrastructure can be retained in Hilti’s IT organization over many years to come, enabling the company’s own experts to scale the systems to match future business needs.

“We migrated to SAP applications on Red Hat Enterprise Linux on HP ProLiant servers after evaluation and testing. Since migrating, we have experienced increased performance of more than 100 percent. We also reduced IT costs with a commoditized architecture. Having a tight level of integration between Red Hat and SAP support organizations gives us the confidence to continue along this path and further reduce our costs by migrating all of our SAP environment to Red Hat Enterprise Linux,” said Martin Petry, CIO at Hilti.

“Considering that we lost a lot of know-how as our legacy Tru64/Alpha servers were switched off, it’s very reassuring to know that Linux and x86_64 processors will be around for many more years to come, independent of the fate of individual vendors or their product decisions. The expertise we have gained is here to stay,” said Hagmann.

The new Red Hat-based platform has additionally brought performance gains to Hilti. Its IT infrastructure is now capable of handling more computing requests with the same number of CPUs as its legacy systems, but now uses less rack space. Lower electricity consumption and a “greener” footprint of the IT infrastructure are additional benefits.

“Running our SAP applications on Red Hat Enterprise Linux has delivered significant improvements in performance,” said Hagmann. “And the servers are very power-efficient, which means cost savings and a reduced carbon footprint.”

A large portion of Hilti’s cost savings result from the elimination of software licensing fees with open source software. “While it is still early to determine the exact total cost of ownership as we’ve just completed migration, our estimates show that our cost benefits are likely to exceed 50 percent compared to our previous UNIX platform,” said Hagmann.

“As a company driven by innovation and passionate engineers, we’re very happy to have made this big step to standardizing on SAP applications on Red Hat Enterprise Linux and industry-standard servers,” said Hagmann. “Now we can benefit from faster innovation cycles of standard hardware and be assured that all the software we need, be it for the operating system or the applications, is actually available on-demand. This clearly gives us a competitive edge in our business operations, as the IT department is able to meet new requirements very quickly.”

If you would like to start planning a platform migration to Linux with minimal downtime and want to learn more about how SAP solutions on Linux could enhance and integrate into your current platform strategy, please email: mds@sap.com.

To learn more about migrating your SAP applications to Red Hat Enterprise Linux, please email: SAP@redhat.com.

HOPS is a leading provider of data analytic tools and products, utilizing its unique data analysis engine to quickly get clients to places they have never been before. The HOPS data analysis engine is an innovative solution for maintaining and analyzing massive amounts of data—hundreds of terabytes and greater. Combining unparalleled speed, compression and sorting capabilities, HOPS is one of the fastest data analysis tools ever produced blasting through the scalability issues that limit traditional database solutions. What’s more, HOPS is delivered using a software-as-a-service (SaaS) model, enabling customers to gain all the benefits of a high-performance, highly secure data center that’s accessible 24×7, without incurring the capital and operating expenses associated with such an environment.

Approach
Deploy HOPS data analysis engine on HP Integrity rx6600 server with Red Hat Enterprise Linux

Business technology improvements
• Increased the speed of database scan by 97 percent
• Ensured high system availability and reliability
• Gained Linux training and certifications to easemigration from UNIX

Business outcomes
• Enabled dynamic reporting on billions of data records
• Met highly demanding service-level agreements (SLAs)
• Gained eight times the performance at one quarter the cost

The key to delivering blazing-fast data analysis results over the web is running HOPS on a high-performance and highly reliable infrastructure. For years, HOPS relied on a RISC-based HP AlphaServer system with Tru64 UNIX. As this system approached end of life, the company searched for a new system that could take HOPS to a whole new level. After a thorough evaluation of leading systems, HOPS chose HP Integrity servers and Red Hat Enterprise Linux

Download [PDF]

FAST FACTS

Company: CQUniversity Australia (Central Queensland University

Industry: Higher Education

Geography: Australia

Opportunity: To migrate from an HP Tru64 Unix environment to a cost-effective and performance-boosting platform within Central Queensland University’s budget

Migration Path: HP Tru64 Unix to Red Hat Enterprise Linux and Red Hat Satellite Server on HP 7000c Blade and BL460c Blade Servers

Software: Red Hat Enterprise Linux, Red Hat Network Satellite, Oracle 10g RAC, PeopleSoft Enterprise Applications (Finance & Student Administration), Talent2 Works (HR)

Hardware: HP 7000c Blade System and BL460c Blade Servers

Benefits: Experienced increased cost savings, performance, ease of management, and reliable ongoing support

Download the case study [PDF]

BACKGROUND
CQUniveristy Australia (Central Queensland University) was founded in 1967 as the Queensland Institute of Technology (Capricornia). It became the Capricornia Institute of Advanced Education in 1971 before a transition phase as the University College of Central Queensland in 1990. It achieved University status in January 1992 and today offers a wide range of undergraduate and postgraduate courses on-campus and through distance education.

CQUniversity has campuses and learning sites in Rockhampton, Bundaberg, Gladstone, Mackay, Emerald and Noosa, on the Sunshine Coast and operates international campuses in Sydney, Brisbane, Melbourne, and the Gold Coast. Some of the Univeristy’s programs are also delivered in Singapore.

With about 20,000 students and approximately 1,200 staff, the daily functioning of the University relies heavily on the reliability and stability of its IT environment, which powers critical finance, human resources, and student-administration systems. With campuses located across various time zones, CQUniversity requires an infrastructure that supports standard operation for at least 13 hours per day.

OPPORTUNITY
For a number of years, CQUniversity serviced its student body and faculty with an extensive HP Tru64 UNIX deployment. As the University continued to expand and demand more from its technology infrastructure, CQUniversity’s IT department knew it was quickly outgrowing its existing IT system, consisting of a four-node cluster for production with two additional web servers, and a six-node cluster with another three web servers for test development, and needed a reliable alternative.

“We knew the time was coming for a change, particularly when we knew we would need more than eight CPUs for a single database server,” said Bruce Young, Peoplesoft Systems Engineer, CQUniversity. “HP Tru64 UNIX was a discontinued product, and we didn’t have unlimited funds to throw at the problem, so we were forced to look at more cost-effective alternatives for expanding our capacity.”

CQUniversity searched for an updated, scalable, and performance-boosting solution that could fit inside its limited IT budget.

SOLUTION
In late 2006, CQUniversity’s IT team decided that from both a hardware and software perspective, Red Hat Enterprise Linux was its operating system of choice. “We looked at a few different configurations, including Solaris on Sun T2000 servers and Red Hat Enterprise Linux on Dell, but it was quite obvious that the HP Blade Servers together with Red Hat Enterprise Linux provided a winning combination that met our needs,” said Young.

Today, CQUniversity runs 12 HP Blade Servers with Red Hat Enterprise Linux for its databases. Ten additional Blades run CQUniversity’s middleware solutions to support its applications.

“Already, Red Hat Enterprise Linux has proven to be a fundamental component of our infrastructure, and while we’re running more environments than before, we’re running them a lot faster,” said Young.

The University has also extended Red Hat Enterprise Linux to its research students and staff, using Red Hat Network Satellite (RHN), Red Hat’s reliable systems management solution, to efficiently manage the deployment and keep it updated without additional, costly resources. Red Hat Satellite Server offers a centralised management tool that enables users to manage server systems more efficiently. With RHN Satellite, CQUniversity has been able to reallocate resources to focus on proactive, high-value roles, instead of reactively maintaining complex systems and lifecycles.

BENEFITS
Since putting its production systems into operation in April 2007, CQUniversity has completed the initial stage of its migration. To date, the University has realised immediate savings of approximately $100,000 when compared to Solaris on Sun T2000 servers.

“We’re definitely impressed with the initial cost savings and the boost to server performance, and we also expect further long-term savings on licensing costs,” said Young. “The reality is that for our budget, Red Hat Enterprise Linux has enabled us to do what what we set out to complete. With Red Hat solutions, we didn’t have to resort to a poor band-aid job, as would have happened if we had chosen a proprietary solution.”

In addition to carving out costs, CQUniversity has experienced a boost in server performance, which has increased by between 20 and 30 percent, and enjoyed an easy-to-deploy solution. “In terms of ease of management, Red Hat Enterprise Linux was the best choice, and the closest fit to what we were familiar with when using Tru64,” said Young.

For the future, CQUniversity has plans to transition additional systems to its new Red Hat Enterprise Linux platform.

Download this video: [Ogg Theora]

FAST FACTS

Industry: Travel

Geography: Global

Business Challenge: To increase performance, reduce costs, and expand scalability for a mission-critical, rapidly evolving IT architecture requiring 100 percent uptime

Migration Path: Unix to Red Hat Enterprise Linux

Software: Red Hat Enterprise Linux 5

Hardware: HPx86 and Intel

Benefits: Achieved system reliability, 90 percent cost savings, 3x performance gains, and unmatched scalability, while standardizing its IT infrastructure on Red Hat Enterprise Linux

Download the case study [PDF]

BACKGROUND
Sabre Holdings is a privately held world leader in the travel marketplace that is owned by Silver Lake and TPG. It merchandises and retails travel products and provides distribution and technology solutions for the travel industry. Headquartered in Southlake, Texas, the company maintains approximately 9,000 employees located across 59 countries.

In 2006, Sabre Holdings earned $2.8 billion in revenue as it supported travellers, travel agents, corporations, and travel suppliers around the world through its three companies: Travelocity, the most popular online travel service offering consumer-direct services for the leisure and business traveller; Sabre Travel Network, responsible for marketing and distributing travel-related products and services through its travel agency and corporate channels; and Sabre Airline Solutions, the world’s largest provider of software products with over 200 airlines relying on its passenger solutions and consulting services.

The first automated passenger reservation system offered by Sabre Holdings was installed in 1960 and marked a dramatic technological leap forward for the airline industry. Since then, Sabre Holdings has pioneered technological advances for the industry in areas such as revenue management, pricing, flight scheduling, cargo, flight operations, and crew scheduling.

BUSINESS CHALLENGE
For years, Sabre Holdings utlized proprietary solutions to support its advanced IT architecture. Internet-based travel bookings were gaining rapid acceptance among a growing segment of customers and travellers demanding more services and better options were driving up costs for the worldwide travel leader. To capitalize on its enormous growth opportunity, the company needed an infrastructure that was faster, cheaper and more scalable. Looking to increase performance and reduce technology costs associated with proprietary solutions, Sabre Holdings decided to investigate open source alternatives to help it face the relentless conditions of a fast-moving, challenging travel industry.

“While operating the largest travel distribution service in the world that processes tens of thousands of transactions per second, it’s always a peak business hour somewhere and we need to be available 24×7,” said Robert Wiseman, chief technology officer at Sabre Holdings. “This creates heavy demands on our IT infrastructure and presents challenges, but with Red Hat we are able to build stronger and smarter systems with our global customers in mind.”

SOLUTION
Enabling more than 32,000 transactions per second, Sabre Holdings’ IT infrastructure demanded 100 percent uptime. If its systems were to experience downtime, the result would be disastrous with planes grounded, passengers stranded, and trips unable to be booked.

To improve the stability, performance, and cost-effectiveness of its IT infrastructure and to ensure reliability for its own customers, Sabre Holdings turned to open source solutions. It evaluated the leading Linux offerings on the market, but ultimately chose Red Hat for its established market leadership and renowned support offerings. With Red Hat, Sabre Holdings also discovered the power of the developer community that could provide reliable, consistent updates and fixes that could be incorporated in realtime.

“We selected the IT solution that would allow us to run our business most effectively. Our customers use our services for the content, speed, price, and presentation, not for the operating system our business runs on,” said Josh Hartman, chief technology officer at Travelocity. “Red Hat provides a proficient and effervescent experience for our customers, who don’t even realize that Red Hat is there. And that’s the way it should be.”

With millions of online customers completing tens of thousands of transactions per second, Sabre Holdings has deployed Red Hat Enterprise Linux on thousands of HP x86-based servers and it continues to expand its use of Red Hat solutions today.

“Red Hat Enterprise Linux is our corporate standard – as we purchase hardware and build systems, they are required to work seamlessly with Red Hat Enterprise Linux,” said Wiseman. “Our systems now perform three times faster while costing 90 percent less when compared side-by-side to other providers.”

When Sabre Holdings recently acquired lastminute.com, a leader in European online travel, its IT system included over 1,000 servers running numerous free open source operating systems that required significant time and resources to operate and maintain. Today, Sabre Holdings is in the process of migrating all of lastminute.com’s servers to its corporate standard, Red Hat Enterpise Linux 5.

BENEFITS
Since migrating to Red Hat Enterprise Linux, Sabre Holdings has benefited from system reliability, cost savings, increased performance, and scalability.

Red Hat Enterprise Linux has enabled Sabre Holdings to standardize its IT architecture, consisting of thousands of systems worldwide, across all three of its companies’ IT infrastructures. “Red Hat has allowed us to horizontally scale our systems with the ability to drive thousands of transactions per second, 24×7, 365 days a year,” said Hartman. “Red Hat is the leader in the marketplace and offers the performance and stability to drive our high-transaction business at a high economic value.” In addition to the performance improvements and cost savings realized since migrating to Red Hat Enterprise Linux, Sabre Holdings also further strengthened its position as an innovative technology leader and employer. Working with open source technology partners has allowed Sabre Holdings to recruit and retain the best talent.

“Our team at Sabre Holdings is unparalleled in the industry, we are the leading travel services provider and our use of open source software attracts IT leaders to our company,” Wiseman said. From building the first computer reservation system in the 1960s, to blazing the trail for online travel in the 1990s, to delivering the sophisticated, cutting-edge travel solutions of today, Sabre Holdings has made innovation its business. “With decades of experience in the travel service industry, we built a culture of uptime and reliable performance. Standardizing on Red Hat Enterprise Linux has allowed us to focus on our core business competencies,” said Wiseman.

“Sabre Holdings relies on Red Hat as our corporate standard operating system for our mission-critical IT infrastructure.” As it continues to evolve its IT architecture, Sabre Holdings plans to incorporate additional Red Hat solutions and is currently assessing Red Hat integrated virtualization technology. “We are beginning to take a close look at Red Hat’s virtualization capabilities and have been impressed with its recent virtualization management technology development. We’re also interested in flexible deployments of Red Hat on Amazon’s elastic cloud,” said Wiseman. “We value Red Hat as a long-term partner in our development of Sabre Holdings’ corporate IT goals because of the visionary, leadership role Red Hat plays in the open source community.”

FAST FACTS

Industry: Utilities, Government

Geography: Western Australia

Business Challenge: Decrease technology costs by avoiding expensive hardware replacement fees and improving performance and manageability

Migration Path: Tru64 Unix to Red Hat Enterprise Linux

Software: Red Hat Enterprise Linux 4.6, Red Hat Enterprise Linux 5.1, Oracle 9i
Hardware: 30 x HP DL385 (AMD Opteron)

Benefits: Increased performance by nearly 500 percent, achieved significant cost savings of approximately $140,000 per machine and gained reliable support from a trusted vendor

Download the case study [PDF]

BACKGROUND
Western Power is an electricity networks corporation owned by the Western Australian Government. It transports electricity from power stations to towns and cities and then distributes it to homes and businesses. This is done via a large network of power lines known as the South West Interconnected System (SWIS), designed to meet the needs of customers and developers and to bring electricity to new areas.
While companies that produce electricity as generators, and companies that sell electricity as retailers, have access to the network, Western Power is responsible for maintaining this network and restoring power after interruptions.

“Western Power supplies energy to more than 840,000 households and businesses throughout the state,” said Julian Rouse, systems administration manager, Western Power.
“Our mission is to conduct a safe, efficient, and importantly reliable operation. As the backbone of the business, our technology infrastructure is paramount to that.”

CHALLENGE
Western Power had been a long time Compaq Tru64 Unix user, but with HP’s acquisition of Compaq, the organisation was prompted to re-evaluate its existing server environment. It was faced with aging hardware, but needed to keep server replacement costs to a minimum.
In designing a new strategic platform for the long term, improving current software performance and improving manageability were key priorities for Western Power.
“Reassessing our strategic platform initially involved a comparison between HP-UX and Compaq TRU64 offerings, but when we looked at what our major energy application provider was doing, our attention quickly turned to Linux,” said Rouse.

Western Power’s ENMAC energy management solution, developed by GE, is responsible for providing centralized, controlled, secure, safe access to the distribution network, in real-time, for maintenance and fault restoration purposes. With GE testing and then porting the software application onto Linux, it made perfect sense for Western Power to consider Linux for its new platform. This would ensure that it could achieve the best possible performance from the solution.

SOLUTION
Western Power investigated Novell SUSE and Red Hat Enterprise Linux for its infrastructure upgrade. With the knowledge that GE was already testing ENMAC on Red Hat, the organisation soon realised that only Red Hat could guarantee optimum performance, ease of management and first-class support.
“While we evaluated the leading Linux offerings on the market, Red Hat had already established itself as the enterprise Linux market leader, and that gave us a great deal of confidence,” commented Rouse.
As Rouse recalled, another factor in choosing to work with Red Hat was Red Hat Network, an easy-to-use systems management platform that brings greater efficiency to package updates and other ongoing maintenance tasks.

“With Red Hat, we also knew that we had instant access to the right support if and when we needed it, rather than being referred to downloads from a public website. Plus, the power of the developer community meant that the Linux kernel was constantly evolving and being further improved – and that was an attractive proposition.”

By August 2007, Western Power commenced its migration to a Red Hat Enterprise Linux platform as part of an overall ENMAC update. The migration started with 18 HP DL385 (AMD Opteron) test and development servers and extended to a further 12 production servers, running a mix of Red Hat Enterprise Linux 4.6 and Red Hat Enterprise Linux 5.1. A total of more that 30 servers in development, test and production.
Through its migration to Red Hat Enterprise Linux, Western Power became one of the first GE customers to run the ENMAC energy management solution for distribution on Linux.

BENEFITS
For Western Power, the migration to Red Hat Enterprise Linux resulted in a migration to higher performance and lower running costs. Unix loads typically run on Linux using commodity-class hardware that costs up to 75 percent less than what is required by Unix systems. Through experience, Western Power has been able to realise these cost savings.

According to Rouse, “Running Red Hat Enterprise Linux on more cost-effective commoditised hardware has seen the kernel outperform any previous build. This is great because the difference in cost between the HP servers we’re now able to use, compared to TRU64 Unix servers, is in the order of several magnitudes.”
“Furthermore, we’ve been able to halve our hardware cost per node, because the 64-bit kernel has enabled us to tune our hardware for both excellent memory and disc performance, allowing us to be able to run both Oracle and EMAC on the same box, whereas previously this required separate tiered machines” Rouse added.

Rouse estimates that, to date, Western Power has experienced CAPEX cost savings of approximately $140,000 per machine, along with a 50 percent increase in total system headroom due to an average CPU usage drop and an impressive performance gain of 500 percent.
Western Power also looks forward to further savings on an ongoing basis. With the move to commodity hardware and virtualisation, it will be able to decrease annual hardware maintenance costs by up to 80 percent.

“There’s no doubt that Red Hat Enterprise Linux is our operating system of choice, and as we embark on overhauling our middleware application, our next focus will be on extending our investment in open source with the introduction of JBoss into our environment,” said Rouse.
This entry was posted under Government, APAC, AMD, Oracle, Red Hat Enterprise Linux 5, Tru64 to RHEL. You can follow any responses to this entry through the RSS 2.0 feed. Responses are currently closed, but you can trackback from your own site. Edit this entry.

Industry: Financial Services

Challenges: Handle highly volatile user traffic and accomodate rapid growth. Seamlessly migrate from legacy UNIX-RISC environment. Improve network performance, while simultaneously reducing costs. Eliminate downtime

Solution: Software: Red Hat Enterprise Linux, Oracle 9i Database & Order Routing System

Benefits: Enterprise Linux on x86 servers generate 10 times more performance than the earlier UNIX-RISC infrastructure. Linux’ inherent security provides a hassle free solution.

This story is available in the following languages: [  ]

Introduction

Indiabulls is India’s leading retail financial services company with a network of 135 outlets spread across 95 cities. Indiabulls offers a complete range of financial services and products, including equities, insurance and personal loans. The company has a vision of using technology to create a world of smart investors. Indiabulls offers real-time electronic trading capabilities to over 1,30,000 investors across the country, generating a total turnover of Rs 10,000 million daily in cash and derivatives.

Challenges

Indiabulls offers investors across the country a secure, reliable way to carry out their transactions remotely. Using an installable client application provided by Indiabulls, investors can log in to the Indiabulls Professional Network from a remote PC. After their identity has been verified, investors can proceed to send transaction data to the Stock Exchange. A confirmation receipt is returned to the investors after the data has been transmitted. The application server runs an Order Routing System, which manages the transaction lifecycle.

“This end-to-end transaction round trip take place in approximately two seconds; and it’s the Internet that takes up a bulk of the delay,” says Tejinderpal Singh Miglani, CTO, Indiabulls Group.

With approximately 8,000 concurrent investors logged on to the Indiabulls Professional Network at any given time, there is absolutely no room for error. During peak loads, investors generate up to 2,000 requests per minute for which 4,000 responses are triggered in return. This puts the total peak volume of online transactions at 6,000 per minute. The database server logs the entire length of the transaction, handling an average of 25 queries per transaction. Under a peak load of 6,000 online transactions, this translates to approximately 1,50,000 queries per minute.

During volatile market days, there are sudden spikes in usage, which adds a further strain to the database and application server. “We needed a high performance platform that could match our demanding, continuous availability needs,” explains Miglani.

Approximately 45-50 percent of Indiabulls’ total revenue is generated online, which totals to about Rs 1 crore daily. A downtime of a single minute could generate a loss of Rs 50,000 for the company. “Our database and application servers are the lifelines of our online share trading business. Even a performance delay of a few seconds translates into lost revenue for us,” adds Miglani.

Originally, both the Oracle 9i Database and Order Routing System were hosted on HP Tru64 Alpha. As the UNIX platform headed towards uncertainty, Indiabulls decided to migrate to the latest platform available.

As the number of investors grew, Indiabulls was expecting the workload to increase significantly. Indiabulls was also keen to introduce redundancy into its infrastructure to provide reliable services to its customers. The enterprise had an option of continuing with proprietary UNIX-RISC machines, which presented an expensive proposition and an uncertain future. Indiabulls wanted a low cost platform that could run on commodity hardware without any compromise in performance or reliability. “A non-proprietary platform running on x86 servers would give us complete freedom from vendor lock-ins and eliminate the expensive operating costs that proprietary UNIX-RISC machines command – i.e. expensive upgrades, updates, patches and maintenance,” says Miglani.

Solution

After an extensive evaluation phase, Indiabulls found Enterprise Linux to be the perfect choice. “The performance of Enterprise Linux overshadows Windows under benchmark results. Enterprise Linux offers a reliable, robust architecture for which support is easily available. Also, since we were planning to implement Oracle RAC in future, Red Hat was the perfect platform, as it offers the maximum pre-certified configurations for RAC technology,” says Miglani.

On the hardware front, as the 64-bit RISC technology faced limited and extremely slow development over the years, Indiabulls found high-speed, 32-bit x86 servers an attractive proposition to invest in. “Migrating away from the proprietary RISC-UNIX bundle to Red Hat running on inexpensive x86 servers provided the best value proposition for us,” says Miglani.

The backend infrastructure consisting of an Oracle 9i database and an Order Routing System are both powered by Red Hat Advanced Server 2.1. The personal loans system which falls under Indiabulls Credit Services has also been hosted on Red Hat Advanced Server 2.1. As the Internet Share Trading and personal loans applications were developed in Java, porting to Enterprise Linux was virtually seamless.

Benefits

By using Enterprise Linux, Indiabulls could extract high performance from low cost, 32-bit x86 servers, which were available at less than half the price of their 64-bit UNIX-RISC counterparts. Enterprise Linux running on the x86 architecture has empowered Indiabulls to purchase more servers and generate 10 times more performance than its earlier UNIX-RISC infrastructure.

“Linux’ inherent security and lower vulnerability to viruses has delivered a hassle free solution,” claims Miglani. With absolutely no downtime or performance delays, Enterprise Linux ensures that daily revenues remain intact at Indiabulls.

Future Roadmap

Indiabulls is now looking at standardizing on Linux and is also considering migrating to Red Hat Enterprise Linux v.3 to further increase performance. With growing popularity of Linux on the desktop, Indiabulls is considering to port its client application to Linux. In doing so, it hopes to meet the needs of investors who prefer to use Linux desktops from their home or office in connecting to their services network.

Conclusion

Indiabulls has found a robust and reliable solution by banking its entire online Share Trading platform on Enterprise Linux. With its mission critical needs satisfied by Enterprise Linux running on low cost servers, Indiabulls has managed to procure a cost effective alternative to proprietary technologies.

32 node and 96 node clusters power national research projects

Industry: Academia

Challenges: Modernize and scale proprietary, SMP-UNIX infrastructure. Set up a high performance computing environment that can handle cutting edge research projects. Simplify management and system administration.

Solution: Software: Red Hat Enterprise Linux
Hardware: AMD64 and Intel 32-bit servers

Benefits: Improvement in performance and overall computation time through 64-bit x86 servers. Significant TCO reduction by eliminating the need for proprietary licenses and RISC based hardware. Freedom to modify and work with the source code

This story is available in the following languages: [  ]

Introduction

Indian Institute of Technology (IIT), Kanpur is one of the premier institutions established by the Government of India. The institute has a goal to conduct original research of the highest standard and to provide leadership in technological innovation for the growth of the country. Established in 1959, IIT Kanpur boasts of a sprawling residential campus that spans across 420 hectares. With path-breaking innovations in both its curriculum and research, the institute is rapidly gaining a legendary reputation across the world.

Challenges

In addition to offering formal undergraduate and post-graduate programs, IIT Kanpur is a leading contributor to national research & development. The institute faces a formidable challenge in maintaining a state of the art research facility that enables both students and faculty to participate in national science and technology projects.

The projects typically involve development of complex applications in diverse areas like fluid dynamics, molecular modeling, speech recognition, etc. The conventional method employed by the institute to run these resource hungry applications was to use large RISC based, SMP servers running different flavors of UNIX.

Over the years, IIT Kanpur had amassed a significant number of standalone SMP servers that were quickly reaching their threshold limits. In addition to requiring the purchase of expensive compilers and proprietary software components, the flexibility offered in a closed environment was limited. The availability of local support staff was another concern, as the infrastructure had to be up and running on a 24×7x365 basis. In the event of a failure, sourcing support from outside the campus was a painstaking process.

A state of the art High Performance Computing Cluster (HPCC), capable of processing large amounts of parallel or sequential data became the need of the hour.

“For our high performance computing needs, we were looking for an operating environment that could deliver top performance and stability, while maintaining a high degree of sustenance for running complex jobs,” says Dheeraj Sanghi, Professor, CSE Department, IIT Kanpur.

“Some of the complex applications developed on campus require a continuous runtime in excess of a month. Imagine the frustration and loss if a crash were to happen on the Nth day. It would not only put the research project behind schedule, but would also mean a complete wastage of weeks of computation time. We required a popular platform that had campus wide support, which would reduce our dependence on outside vendors to a bare minimum,” adds Sanghi.

Solution

IIT Kanpur’s tryst with Linux on the HPCC front began in late 2002. After setting up its first experimental Beowulf cluster with 16 Pentium III class servers on Red Hat Linux 7, the benefits of using an open source platform became apparent. Initially, the cluster was meant to be only a test bed for running parallel applications.

“Before this, we had used Red Hat Linux only on the non-critical services side to run applications like Web servers, mail servers , proxies, DNS, etc.,” claims Brajesh Pande, Senior Computer Engineer, Computer Centre, IIT Kanpur.

“However, the satisfactory results that we soon witnessed on the HPCC test bed justified the investment in state of the art Linux clusters for our research facility,” he adds.

In mid 2004, the institute setup its first Beowulf cluster for production use on the Red Hat Linux 9 platform, powered by 32 low cost, x86/32-bit servers. With the availability of low cost AMD 64-bit processors in the market and the significant 64-bit capabilities built into the Red Hat Enterprise Linux platform, the institute began to look at purchasing new 64-bit machines to add a second, more powerful cluster.

“Since Red Hat offered a stable, base OS that was compatible and certified across a wide range of 64-bit hardware, it was a natural choice for the second cluster project. Plus, it offered full support for different compiler platforms through GCC,” explains Sanghi.

In 2005, the institute setup its second, 96 node cluster with 98 AMD64 Opteron servers powered by Red Hat Enterprise Linux 3. Two dedicated engineers employed by the institute manage the two clusters 24×7x365.

“Instead of investing in expensive RISC-UNIX 64-bit servers, low cost AMD64 Opteron machines powered by Red Hat Enterprise Linux were a smarter choice for us. As Red Hat is the most popular Linux flavor in the market, everyone on campus was already familiar with its intricacies. In fact, we have been both using and actively following Red Hat’s progress right from the release of Red Hat Linux 5,” adds Sanghi.

Benefits

With proprietary UNIX-RISC servers, scaling the HPC infrastructure was an expensive proposition for IIT Kanpur. With Enterprise Linux running on non-RISC hardware, the institute has found a scalable, low cost solution that delivers the same performance, without any compromises.

“With Red Hat’s popularity on campus, both students and faculty now have the flexibility to develop applications using open standards on their Fedora or Red Hat machines. When these applications are ready, they can seamlessly move them to the HPCC environment for computation,” says Sanghi.

“Red Hat has also given us tremendous freedom to tinker with the OS, which wasn’t true in the earlier proprietary environment. In the event of a crash, skills are available throughout the campus to resolve issues immediately. With Red Hat, our dependence on support calls to proprietary vendors has been eliminated completely,” adds Sanghi.

“Another benefit of using a certified platform like Enterprise Linux is that third party HPCC software for resource allocation and node management run without any problems,” adds Pande.

A critical requirement at the time of setting up the two Linux clusters was scalability of usage. The institute currently has 100 users who access the HPC laboratory. In the next six months, the figure is expected to scale to 300.

“While designing the solution, we wanted to make sure that we could extract maximum throughput, to the last drop. Enterprise Linux coupled with our high speed servers have allowed us to support multiple applications and users simultaneously,” explains Sanghi.

With the new 64-bit HPCC environment powered by Red Hat Enterprise Linux, computation time has been reduced by 50%. “Applications that typically used to take 3-4 weeks to run earlier, now take less than two weeks,” he adds.

Future Plans

With a 6 TB SAN already in place to support the 64-bit cluster, IIT Kanpur’s storage requirements are in the process of scaling rapidly.

“Red Hat’s Global File System would be a nice filesystem to have and we are actively considering it. Also, new technologies like Xen virtualization, Stateless Linux, SystemTap, etc. are exciting to look forward to in the next Red Hat releases,” adds Sanghi.

Conclusion

The freedom and flexibility offered by Red Hat Enterprise Linux has allowed IIT Kanpur to setup two of the most popular Linux HPC clusters in the country, which have contributed significantly to next generation research projects.