Re-Hosting Mainframe Applications on Intel Xeon Processor-Based Servers

(1)

Re-Hosting Mainframe Applications on

Intel® Xeon® Processor-Based Servers

A technical discussion of methodology, design, and benefits based

on a successful migration by one of the world’s largest banks

Executive Summary

Many enterprises rely on mainframes to support core business operations, yet the cost and complexity of these monolithic systems can become roadblocks to innovation in today’s increasingly fast-paced business environment. Such was the case for HSBC, one of the world’s largest banks. When heavy workloads from a new suite of loan applications created performance and reliability issues for the HSBC mainframe in Mexico City, the company had a choice to make: either upgrade the mainframe or migrate existing workloads onto an alternative platform.

After a thorough review of alternatives and consultation with Intel, HSBC Mexico chose to re-host the applications on Intel Xeon processor-based servers running Linux. The resulting migration met or exceeded all of HSBC’s goals. Performance and uptime were improved, delivering a better experience to customers and branch office personnel. Mainframe workloads were reduced by 2,000 MIPS and monthly service charges by 70 percent. The migration also improved business flexibility, making it easier to scale and adapt applications as business requirements continue to grow.

Two factors were particularly important to the success of the HSBC mainframe migration. • The latest two-, four-, and eight-socket Intel Xeon processor-based servers

provide major advances in scalability and reliability for mission-critical workloads, while retaining the high value and interoperability of standards-based computing platforms. HSBC had considerable flexibility in designing its solution, and was able to meet all requirements using a small number of powerful, cost-effective servers.

• HSBC took advantage of experienced vendors and a proven methodology to accelerate the migration and reduce risk. The company followed a step-by-step approach to setting goals, assessing the current environment, and designing, testing, rolling-out, and integrating its new solution. The result was a smooth migration with predictable costs, timelines, and benefits.

Although every mainframe migration is unique, the basic principles required for success are well established. This paper describes the overall migration approach and methodology employed by HSBC. It outlines a practical strategy that other companies can use to plan and implement a high-value, low-risk mainframe migration that improves business agility, while significantly reducing total costs.

Business Growth

and Mainframe Constraints

HSBC has more than 7,500 offices in 87 countries and territories around the globe. For many years, the company relied on a mainframe in Mexico City to support its end-to-end banking solutions throughout Latin America. Recently, HSBC Mexico added a suite of 12 internally developed, Java-based solutions to support loan application, processing, workflow management, business modeling, and related services for its online customers and branch offices.

The new applications were a huge success, fueling more demand for loans and related banking services. Transaction rates rose to an average of approximately 10 million per day, with up to 600,000 transactions per hour during peak periods. This added significantly to mainframe workloads, creating performance bottlenecks and reliability issues not only for the new loan applications, but also for other core banking services.

HSBC needed to resolve these issues to maintain the quality of the customer experience and enable additional growth. However, decision-makers were concerned about the costs of upgrading the

mainframe environment and wanted to explore options for migrating workloads to a more cost-effective computing platform. Since moving mission-critical applications off the mainframe represented a fundamental shift in HSBC’s strategy, a comprehensive and systematic approach was needed to ensure the chosen solution would fully address both current and future needs.

A Proven Methodology for

Mainframe Migration

Mainframe migrations have increased substantially in recent years, and there are now many experienced regional and global system integrators, independent software vendors (ISV), and hardware suppliers with proven methodologies, tools, and professional services to assist with migrating workloads to Intel® processor-based servers. HSBC worked with experienced vendors and took advantage of a well-established migration methodology throughout its assessment, planning, and deployment efforts (Figure 1). To ensure business and technical

issues were thoroughly explored, an interdisciplinary team of managers, business analysts, system and application architects, and IT operational staff was formed. As a first step, the team participated in an in-depth migration planning and technical workshop to make certain all participants had a clear understanding of how to conduct an efficient and effective migration. (see the sidebar, Understanding Mainframe Migration)

Figure 1 . A migration methodology based on documented best practices helps to reduce cost and risk and accelerate time to benefits.

1 2 3 4 Workshop Assessment Migration Methodology

Approach OverviewSolutions Financial OverviewBusiness &

Evaluate &

Analyze DefinitionPlan SolutionDesign

Proof of

Concept SolutionTest OptimizationSystem

Infrastructure

Deployment ApplicationRoll-Out Operations &Monitoring

Deployment

UndERSTAndInG MAInFRAME MIGRATIon

Most IT organizations will benefit from an in-depth technical workshop provided by a vendor with extensive experience in mainframe migration. The entire life cycle should be discussed, including best practices for each of the following steps.

• Reviewing the current system, solutions, and business requirements • Developing and documenting the

proposed solutions and designs • Evaluating and testing the proposed

system architectures and infrastructure • Developing plans for proof of concept

(PoC) testing

• Migrating business applications and solutions

• Upgrading the core software infrastructure as required

• Developing detailed migration plans, schedules, and contingencies

• Integrating IT operations and management tools with migrated applications • Managing the project to ensure

milestones and deadlines are met

(3)

Establishing Business and Technical Goals

Following the initial workshop, the team determined the goals of the migration based on business and technical priorities. Defining and documenting goals early in the project is important to provide a framework for weighing alternatives throughout the planning process. The goals included:

• Providing a better foundation for business growth. A key objective was to enable cost-effective scaling of performance and capacity to handle projected increases in user volumes and application requirements, while matching or exceeding the reliability, security, and responsiveness of the existing mainframe solution.

• decreasing total IT costs. Reducing overall IT capital and operational expenditures (CAPEX and OPEX) would free up capital to accelerate growth and innovation. It was clear that moving key workloads off the mainframe onto lower-cost systems would reduce hardware, maintenance, and software licensing costs. The team would work to maximize the savings without sacrificing other key requirements.

• Improving IT efficiency and agility. The flexibility of open system computing solutions would simplify growth. However, excessive variability in a distributed environment can offset this advantage by increasing complexity. To maximize efficiency and agility, it would be important to establish standards and define and implement IT operational best practices.

• Maintaining global governance. Designs would need to meet company-wide policies, including IT requirements for reliability, availability, serviceability, and security. They would also need to be compliant with internal and external regulatory and operational standards, including local, national, and worldwide requirements.

Assessing the Current Environment Once goals were established, the HSBC team began a comprehensive assessment of its current business, technical, and operational environments, including: • Physical, logical, and solution

architectures across production, staging, disaster recovery, quality assurance (QA), and development systems (Figure 2). • All key software applications,

middleware, and utilities including the product versions, releases, and updates. In many cases, migrating applications from one platform to another will require upgrading the base software for the targeted environment to the latest or supported versions.

• Performance, functionality, and interoperability across the full hardware and software solution stack. This included identifying bottlenecks that were impacting transaction performance. • Key mainframe interfaces, including

solution and systems interdependencies.

• Future performance requirements based on growth in workloads, transactions, the number of users, and the integration of new services.

The IT industry has developed a number of infrastructure assessment and inventory tools, utilities, and programs to help with hardware and application migrations. These tools range from simple spreadsheets to advanced programs that may include automated data center discovery. Depending on the selected tool, it may measure system workloads, resource utilization, and dependency mapping. It may also discover physical and virtual IT assets, applications, and the relationships between them. The resulting assessment typically includes detailed reports and graphs, and can be useful for system design and for identifying opportunities to virtualize infrastructure and reduce overall costs.

Figure 2 . A comprehensive and integrated assessment of the existing infrastructure provides a foundation for planning an efficient, low-risk migration.

Integrated Assessment

Business Solutions & Applications

Staging & Development Systems Backup Systems Storage Devices Network I/O

Database Application Services Commercial and Custom Applications

Operating System Enterprise Infrastructure So ftwar e Utilities, IT Oper ational T ools IT Oper

ations & Managemen

t IT Managemen t Fr ame w ork Har dwar e De vices, So ftwar e Driv ers In fr as tructur e P erf ormance System Performance

Integrated Solution and Infrastructure Assessment

In tegr ated S ys tem Assessmen t 3

(4)

The new HSBC loan solution is a suite of 12 Java 2 Platform Enterprise Edition* (J2EE) applications running on IBM WebSphere* Application Server. This solution, along with a number of other banking solutions and the IBM DB2* database, was running in two logical partitions (LPAR) on the mainframe (Figure 3). The workloads for the loan applications were balanced across the LPARs based on user groups. Transactions for the loan applications, as well as access to other banking solutions, were managed using a customer information control system (CICS) transaction server.

Figure 3 . Physical and logical architecture for the HSBC mainframe environment prior to migration. The suite of 12 loan applications was running in two LPARs on the mainframe. The IBM DB2* database was also running on the mainframe.

Mainframe

Network (Redundant Connections)

Original System and Solution Architecture

Physical Architecture

Mainframe

Logical Architecture

Mainframe

Z LPAR

IBM WebSphere* Application Server

Loan App 1 http JDBC CTG Loan Database (IBM DB2*) Loan App 2 http Loan App 3 http Loan App 4 http Loan App N http

Loan App AOR 1: AOR Loan App AOR 2: AOR

Loan App TOR: TOR

Mainframe

Logical Architecture

(5)

Creating the Proposal and developing the Solution Architecture

Following the initial analysis, the team reviewed options for meeting current and future needs including upgrading the existing mainframe and offloading selected applications. Technical feasibility was evaluated for each option, and detailed return on investment (ROI) and total cost of ownership (TCO) analyses were developed.

The result of the analyses was a detailed recommendation to re-host the suite of 12 loan applications on Intel Xeon processor-based servers running Linux, while leaving the database and other core applications running on the mainframe. This would eliminate the current performance bottlenecks, reduce CAPEX and OPEX, and make it easier to scale and adapt HSBC’s loan solutions going forward. It would also preserve the value of the company’s mainframe investments while avoiding both the high costs of upgrading the mainframe and the risks associated with a more extensive migration.

The plan included a phased migration, which further reduced risk and increased overall IT operational flexibility and efficiency. The plan also included specific recommendations to address scalability and redundancy requirements, support efficient integration with the existing IT operational environment, and ensure adherence to security and governance requirements.

The team took advantage of the migration project to upgrade to the latest version of IBM WebSphere Application Server, so technical staff would have access to the latest features and functionality.1_They

also moved the remaining mainframe applications into a SYSPLEX* environment with data sharing to provide better overall resilience for the mainframe solution. The detailed solution architecture and migration plan were developed by a technical team that included a project manager, system and application architects, data center IT operations personnel, database administrators, and quality assurance staff. Business unit stakeholders were also included to

ensure the proposed architecture was well aligned with current and projected business needs.

This team was responsible for ensuring: • Integration with the existing mainframe

and IT operational environments • Selection of mission-critical systems to

support current and planned growth • Adherence to security and

compliance requirements

• Migration, testing, and deployment, including PoC testing

• Identification and mitigation of potential risks to the success of the project • Detailed business and financial analyses

to estimate costs and determine the ROI • Alignment of development and testing

resources between the system stabilization and platform migration initiatives. (It was important for HSBC to share the knowledge and resources among teams while doing the migration.)

Selecting the Servers

The team reviewed the performance and functional capabilities of a variety of Intel Xeon processor-based servers to ensure the proposed design and solution architecture could fully address business and technical needs including compute performance; reliability, availability, and serviceability (RAS); and overall systems scalability.

Two-socket servers based on the Intel Xeon processor 5600 series were selected for the Web serving tier. As demonstrated by published SPECweb2005* benchmark scores (Figure 4), these servers can support up to 104,422 simultaneous user sessions in a modern web environment. That is more than 20 times the capacity of older servers based on single-core processors. By balancing Web server workloads across two or more servers, customers can meet the most extreme performance and availability requirements simply and at relatively low cost.

Figure 4 . Web server performance has increased substantially for two-socket Intel® Xeon® processor-based servers. The latest systems based on the Intel® Xeon® processor 5600 series are ideal for processing large numbers of concurrent user requests.1_{(The SPECweb*2005 benchmark} score of 104422 is the geometric mean of the following results: SPECweb2005_Banking = 162000, SPECweb2005_Ecommerce = 177000, SPECweb2005_Support = 88000).

60,000

Intel® Xeon processor 3.80 GHz (2 MB L2, 800 MHz FSB, single core)

4555 15193 29591 104422 83198 30,000 0 120,000 SPECweb*2005 Score (Higher is better) 90,000

Intel® Xeon processor 5160 (3.0 GHz, 4 MB L2, 1333 MHz FSB, dual core) Intel® Xeon processor X5460 (3.16 GHz, 2x6 MB L2, 1333 MHz FSB, quad core) Intel® Xeon processor X5570 (2.93 GHz, 8 MB L3, 6.4 GT/s, quad core) Intel® Xeon processor X5680 (3.33 GHz, 12 MB L3 cache, 6.4 GT/s, six core)

(6)

Four-socket servers based on the Intel® Xeon® processor 7500 series were selected for hosting HSBC’s suite of 12 Java-based loan applications, the workflow solutions, and the IBM WebSphere Application Server. These larger servers deliver scalable performance for heavy, compute-intensive workloads. With increased memory capacity and a high-speed, point-to-point interconnect system, these servers support exceptionally fast processing for high-volume transaction workloads. The tightly-integrated interconnect system also provides RAS for mission-critical environments through features such as advanced error correction and automated link recovery.

Published benchmarks for SPECjAppServer*2004 and

SPECjbb*2005 demonstrate the ability of these servers to support demanding enterprise workloads (Figure 5). Both benchmarks measure performance for Java application servers in a typical three-tier environment. The first provides a comprehensive, full-fledged, multi-tier benchmark. The second simplifies the Java environment to focus more intensely on the application tier (it does not stress the database or network).

Results for both benchmarks show that servers based on the Intel Xeon processor 7500 series delivered a major leap in scalability versus

previous-generation servers, increasing performance by 2.5 to 3.5 times. They also demonstrate excellent scalability in large server configurations. Eight-socket systems supported 1.8 to 1.9 times the workload of four-socket systems, indicating that IT organizations can scale up or out efficiently, depending on their workloads and preferred strategy. Customers can expect to see even higher performance using the more recent Intel Xeon processor E7 family, which provides significantly more computing resources and is pin-compatible with the Intel Xeon processor 7500 series. (To view the latest performance benchmarks as they become available, visit http://www.intel. com/performance/server/index.htm.)

Figure 5 . Four-socket, eight-socket, and larger Intel® Xeon® processor-based servers deliver scalable performance for demanding Java* applications, as demonstrated by published results on the SPECjAppServer*2004 and SPECjbb*2005 benchmarks. Based on workload requirements, HSBC selected four-socket servers for hosting its suite of 12 Java-based loan applications.

16,000

Intel® Xeon processor X7350 (8 M cache, 2.9 GHz, 1066 MHz FSB) 3340 4410 11057 20092 8,000 0 SPECjAppServer*2004 JOPS@Standard (Higher is better) 24,000

Intel® Xeon processor X7460 (16 M cache, 2.66 GHz, 1066 MHz FSB Intel® Xeon processor X7560

(24 M cache, 2.26 GHz, 6.40 GT/s Intel® QPI) Intel® Xeon processor X7560

(24 M cache, 2.26 GHz, 6.40 GT/s Intel® QPI)

Intel® Xeon processor X714OM (16 M cache, 3.40 GHz, 800 MHz FSB) Intel® Xeon processor X7350 (8 M cache, 2.93 GHz, 1066 MHz FSB) Intel® Xeon processor X7460 (16 M cache, 2.66 GHz, 1066 MHz FSB) Intel® Xeon processor X7560

(24 M cache, 2.26 GHz, 6.40 GT/s Intel® QPI) Intel® Xeon processor X7560

(24 M cache, 2.26 GHz, 6.40 GT/s Intel® QPI)

217334 537116 633897 8-Socket 4-Socket 8-Socket 4-Socket 3816799 2021535 SPECjbb*2005 BOPS (Higher is better) 4,000,000 3,500,000 3,000,000 2,500,000 2,000,000 1,500,000 1,000,000 500,000 0

Intel® QPI - Intel® QuickPath Interconnect

(7)

other design Considerations The technical team identified and evaluated a number of design options for the new infrastructure and worked closely with all participants to ensure the chosen solution would meet business and technical requirements. This close collaboration was fundamental to the success of the project, enabling the team to make better decisions faster by taking advantage of their combined expertise. Four of the most important design decisions were focused on ensuring: • Scalable performance. The IBM Redbook

entitled “WebSphere Application Server V6 Scalability and Performance Handbook” provides a number of detailed and proven design options for deploying both horizontally- and vertically-scaled solutions on Intel Xeon processor-based servers. After examining several options, the technical team chose an intelligent load balancing solution to distribute requests among Web servers running on multiple physical servers. This strategy delivers the required performance, scalability, and security and allows applications to scale readily as business requirements change.

• Efficient transaction management. The new solution would have to support fast, reliable, and secure interactions among the loan applications and the mainframe environment. The team reviewed a number of design, deployment, and configuration options based on IBM CICS* Transaction Gateway* (CICS TG) and IBM WebSphere MQ*. An MQ solution was selected because it provided a better fit with HSBC IT standards. The chosen configuration includes MQ clients and clustered MQ servers. Applications can access MQ running on one server and the queue manager running on a different server. This design has the benefit of reducing the number of required servers because there is no need to implement a full version of MQ. Accessing CICS using

MQ was also found to be a good fit for IBM WebSphere Application Server and HSBC security requirements.

• High-speed data access. The IBM WebSphere Application Server instances running on the Intel Xeon processor-based servers require Java database connectivity (JDBC) drivers to support interactions among the loan applications and to provide access to the DB2 database running on the mainframe. Using the appropriate type of driver is important to optimize performance and security. The team reviewed and tested Type 2 and Type 4 JDBC drivers and ultimately selected the Type 4 driver. • Availability. IBM WebSphere Application

Server Cluster was used to support JVM failover so transactions are guaranteed to complete successfully in the event of a server failure.

Using commercially available applications can simplify mainframe migrations, since they can typically be deployed in far less time than custom application code that must be developed from the ground up. HSBC technical teams found Linux versions of all key software components were available for Intel Xeon processor-based servers, including IBM WebSphere Application Server, IBM WebSphere MQ Client, and IBM WebSphere MQ Server. PoC Testing

HSBC technical teams conducted PoC testing to explore crucial design options, address potential issues, test components, and ensure that the proposed solutions met business and IT requirements. The QA mainframe system was used to access and test the DB2 database and MQ connectivity. PoC testing included: • Testing Java-based deployment utilities

on a subset of loan applications to estimate the time and effort required to deploy the full suite.

• Testing the proposed architecture for end-to-end performance, throughput, and scalability. System performance and load-testing measurements were performed with Secure Sockets Layer (SSL) enabled. Metrics included average response time, number of transactions, number of sessions, Web server

requests, database connections and pool size, JVM memory, CPU and I/O utilization, system paging, and base memory configuration. The team used CA Wily Application Performance Management*, Introscope*, and Customer Experience Manager* to monitor application and transaction performance. These and similar applications are valuable for monitoring performance and determining root causes for bottlenecks in distributed and heterogeneous environments. • Testing all critical components related

to connectivity, configuration, and interoperability between the new distributed platforms and the mainframe. Especially close attention was paid to the JDBC and to the MQ environment (WebSphere MQ client, WebSphere MQ server, and WebSphere MQ manager). The testing confirmed that 64 GB of memory per server was an optimum configuration.

The PoC results demonstrated that the proposed solutions met or exceeded all current and projected requirements. Such targeted PoC testing can be extremely valuable, even essential, when migrating custom applications. Testing allows technical teams to discover and address potential performance and operational issues early in the design process. PoCs may not be necessary when migrating commercial applications to Intel Xeon processor-based servers, since ISVs can often provide reference designs and sizing guides as well as best practice recommendations that have been proven across numerous customer implementations.

(8)

The HSBC Solution

Solution Architecture

The high-level solution architecture for the HSBC loan application infrastructure is shown in Figure 6. Transaction requests from customers and more than a thousand branch office personnel are received and distributed across a pair of two-socket Web servers based on the Intel Xeon processor 5600 series. This load-balanced architecture provides primary failover for high availability and can be scaled through the addition of servers as workloads grow. Transactions from the Web tier are distributed across four Intel Xeon processor 7500 series-based servers running IBM WebSphere Application Server on the Linux operating system (OS). Each of these servers is configured with four processors and 64 GB of memory. Any transaction can be received by any Web server and handled by any application server. With this architecture, a failed server or network link will not cause service disruption, and the solution can be scaled horizontally with additional servers to increase scalability and availability. Integration with the mainframe

environment is provided via the MQ clients, which enables an application to connect remotely or locally to an MQ queue manager in the MQ cluster. The MQ cluster allows multiple instances of the same service to be hosted through multiple queues to provide high availability, failover, and scalability for the messaging connections between the mainframe and client systems.

Security was a key requirement in re-hosting the loan applications. The SSL protocol supports secure communication for MQ message passing, and the combined use of Java, IBM WebSphere Application Server, and IBM WebSphere MQ was crucial for enabling a quick migration while maintaining strong security.

Migration Experience

Once the internal migration and initial system testing were complete, the new solution was integrated into the production environment using a phased deployment strategy. Two HSBC branch offices were selected for pilot deployments using a subset of the 12 loan applications. When these deployments were shown to meet performance and business criteria, additional loan applications were released to a larger number of branch offices.

During the pilot implementations, the IT operational tools, scripts, and utilities were integrated into the existing IT management and operational framework to seamlessly manage and support the new infrastructure and environment. The mainframe QA test environment, deployment, and staging systems were extended to account for the new systems. The success of this phase of the project can be attributed to the very close collaboration among multiple groups, including the mainframe (UNIX*) and server (Linux) teams and the IT operational team; in particular, the direct support and participation of the application software engineering teams proved to be invaluable.

Results and Business Benefits To date, the migration and re-hosting of the HSBC loan application suite has proven highly successful. Mainframe workloads were reduced by 2,000 MIPS and all of the goals established at the outset were achieved, including:

• Higher performance. Throughput and average response times have improved, providing a better user experience for customers and branch office staff. • Higher availability. Since the

migration, HSBC has experienced no unplanned downtime.

• Cost-effective scalability. The solution has been successfully scaled to accommodate increasing workloads by adding Intel Xeon processor-based servers to the application tier of the distributed architecture. The upgrades were performed without interruption to application users.

• Lower TCo. Maintenance fees have been reduced by approximately 70 percent, from USD 420,000 to USD 80,000 per month, delivering annual savings of more than USD 4 million.

• Flexibility. The new architecture allows business units, application development teams, and IT operational teams to enhance the loan applications as needed, with little or no impact on the performance and availability of the mainframe production environment. In addition, the new systems have enabled business units to do online modeling without slowing the performance of production systems.

• Increased resilience. Migrating and re-hosting the loan applications helped to improve both the solution design and end-to-end operations. Workloads are balanced with intelligent and robust load balancing, which does not depend on predefined user groups. This enhanced design provides for greater scalability and increased system and application availability.

The value of the new solution extends beyond the immediate technical, business, and financial benefits of the re-hosted loan applications. The success of the project has further solidified HSBC’s intention to standardize on Intel Xeon processor-based servers and the Linux OS. The company has started migrating other applications off the mainframe in Mexico City and will use the re-hosting architecture in other regions to provide similar benefits throughout the company’s global operations. The experience gained in the initial migration is helping to accelerate timelines, improve ROI, and reduce related risks for these additional projects.

(9)

Physical Architecture

Logical Architecture

SYSPLEX* Z LPAR 1

Mainframe

Hogan and Core Banking Applications LPAR MQ* LPAR MQ JDBC IBM MQ Servers 4-socket servers Running Linux (Intel® Xeon® Processor 7500 Series) 2-socket servers Running Linux* (Intel® Xeon® Processor 5600 Series)

IBM WebSphere* Application Server Cluster

Server 1 Server 2 Server 3 Server N

JDBC

MQ MQ

Loan App 1

http

Database and

Applications SYSPLEX Z LPAR 2

Database (IBM DB2 Z)

Loan App 2

http http Loan App 3 http Loan App N

Loan App 1: AOR

Loan App N: AOR

Loan App Batch (Cobol)

GSI: CICS

GSI: CICS Loan App 2: AOR

Loan App 1: AOR

Loan App N: AOR

Loan App Batch (Cobol) Loan App 2: AOR

GSI: CICS GSI: CICS* IBM DB2 for z/OS IBM DB2* for z/OS JDBC J Load Balancer

IBM Mainframe Logical Architecture (database and applicators)

Figure 6 . The new, load-balanced Intel® Xeon® processor-based infrastructure integrates seamlessly with the mainframe environment and will scale easily and cost-effectively to handle future growth.

(10)

Conclusion

Migrating modern workloads from a mainframe to Intel Xeon processor-based servers can deliver fundamental

advantages, including better performance and scalability, improved business and IT agility, and lower capital

and operating costs. HSBC has realized all of these advantages in re-hosting its suite of mission-critical loan

applications. Online customers and branch office personnel are enjoying a better experience, and the company

can now scale and adapt its mission-critical loan applications more easily and without disrupting its production

environment. Since the migration, HSBC has experienced no downtime and monthly service fees have been

reduced by 70 percent, delivering annual savings of approximately USD 4 million.

Today’s Intel Xeon processor-based servers deliver the scalable performance and advanced reliability needed

to support a broad range of workloads currently running in mainframe environments. A successful migration

requires detailed planning, commitment, a team effort by multiple business and technical teams, and direct

support from senior management. A proven migration methodology, such as the one used by HSBC, can help

IT organizations deliver desired benefits quickly, while minimizing cost and risk. Professional support from an

experienced vendor is also recommended, and is available from leading system integrators, ISVs, and hardware

suppliers around the world.

(11)

System interfaces, interdependencies, networking, and storage

Software solution compatibility, business applications and solutions, operating system (OS), middleware, Java* Virtual machine (JVM), database and database connections

Application, database, and transaction performance, including current bottlenecks

Future performance requirements with projected growth in workloads, transactions, and new services Security, compliance, and IT management and operational framework

dESIGnInG THE nEW SoLUTIon And SYSTEM ARCHITECTURE

develop the solution architecture, including physical and logical systems that account for:

Software stack, including business applications and solutions, enterprise service bus (ESB), application server and JVM, database systems, and OS Compute, network, and storage systems, including topology for load balancing and for horizontal and vertical scaling

Performance and scalability for current and projected workloads System availability and maintainability

Security, compliance, and integration with IT management framework Establish solutions for connecting to the mainframe environment, including:

Java database connectivity (JDBC) between the application server (in this case, IBM WebSphere* Application Server) and mainframe databases Transaction management (in this case, IBM CICS* Client Gateway)

Message passing connectivity (in this case, WebSphere MQ* Client and WebSphere MQ Server) PLAnnInG FoR MIGRATIon, dEPLoYMEnT, And IT oPERATIonS

develop detailed plans for migration, including: A comprehensive functional design

Systems and solutions for production, development, test, and staging environments Detailed migration timelines, schedules, and resources

Disaster recovery and failover solutions Proof of concept and testing

Application migration, including tools and utilities Final deployment and go-live

Establish testing tools and procedures for:

Performance testing and infrastructure validation Stress and end-to-end system functionality testing Create plans for deployment and operations, including:

On-going support and maintenance

Integration with IT management, operational tools, and procedures Security and governance

ASSESSInG And MITIGATInG RISK

Identify essential resources and potential risks to the success of the project, including: Key stakeholders, collaborators, and resources

Internal and external team dependencies and requirements, roll-back plan QUAnTIFYInG CoSTS And BEnEFITS

Create detailed reports of benefits, costs, and risks, including:

Business and cost benefits, including license, systems, maintenance, and operational costs

Analyses of total cost of ownership, return on investment, capital expenses, and operational expenses, including projected cost savings

(12)

SPECjAppServer2004 JOPS@standard. Source: www.spec.org/osg/jAppServer2004/results/res2010q1/jAppServer2004-20100310-00140.html. • Four-socket Intel Xeon processor X7560 (24 M cache, 2.26 GHz, 6.40 GT/s Intel QPI). Hewlett-Packard ProLiant DL980 G7 server platform with eight Intel Xeon processors X7560 (24 M cache, 2.26 GHz, 6.40 GT/s Intel QPI), Oracle WebLogic Server Standard Edition Release 10.3.3. Referenced as published at 20,092 SPECjAppServer2004 JOPS@ standard. Source:www.spec.org/osg/jAppServer2004/results/res2010q3/jAppServer2004-20100825-00145.html. Server configurations for Java* performance on the SPECjbb*2005 benchmark. SPECjbb is a trademark of the Standard Performance Evaluation Corp. (SPEC). Comparison based on best published/submitted results on www.spec.org as of September 21, 2010.

• Four-socket Intel®_Xeon®_{processor 7140M (16 M cache, 3.40 GHz, 800 MHz FSB): HP ProLiant ML570 G4 platform with four Intel Xeon processors 7140M (16 M cache, 3.40 GHz,} 1066 MHz FSB), 32 GB memory PC2-3200, Microsoft Windows Server* 2003 R2 Enterprise x64 Edition with SP1, Oracle JRockit* 5 P27.1.0. Referenced as published at 217,334 BOPS. Source:www.spec.org/osg/jbb2005/results/res2006q4/jbb2005-20061121-00222.html. • Four-socket Intel Xeon processor X7350 (8M cache, 2.93 GHz, 1066 MHz FSB): Dell PowerEdge* R900 server platform with four Intel Xeon processor X7350 (8M cache, 2.93GHz, 1066MHz FSB), 64 GB memory, Microsoft Windows Server* 2003 Enterprise x64 Edition SP1, Oracle JRockit 6 P28.0.0. Referenced as published at 537,116 BOPS. Source:www. spec.org/osg/jbb2005/results/res2009q1/jbb2005-20090224-00592.html. • Four-socket Intel Xeon processor X7460 (16 M cache, 2.66 GHz, 1066 MHz FSB): Fujitsu PRIMERGY* RX600 S4 server platform with four Intel Xeon processors X7460 (16 M cache, 2.66 GHz, 1066 MHz FSB), 64 GB memory, Microsoft Windows Server 2003 R2 Enterprise x64 Edition, Oracle JRockit 6 P28.0.0. Referenced as published at 633,897 BOPS. Source:www.spec.org/osg/jbb2005/results/res2009q1/jbb2005-20090305-00663.html. • Four-socket Intel Xeon processor X7560 (24 M cache, 2.26 GHz, 6.40 GT/s Intel QPI): Cisco UCS* C460 M1 server platform with four Intel Xeon processors X7560 (24 M cache, 2.26 GHz, 6.40 GT/s Intel QPI), Intel Hyper-Threading Technology enabled, Intel Turbo Boost Technology enabled, 256 GB memory, 1x 73 GB disk drive, Microsoft Windows Server 2008 R2 Enterprise x64 Edition, IBM J9* JVM (build 2.4, JRE 1.6.0, SR5). Referenced as published at 2,021,535 SPECjbb2005 bops and 126,345 SPECjbb2005 bops/JVM. Source: www.spec.org/osg/jbb2005/results/res2010q2/jbb2005-20100331-00838.html. • Eight-socket Intel Xeon processor X7560 (24 M cache, 2.26 GHz, 6.40 GT/s Intel QPI): Hewlett-Packard ProLiant DL980 G7 server platform with eight Intel Xeon processors X7560 (24 M cache, 2.26 GHz, 6.40GT/s Intel QPI), IBM J9 VM, Windows Server 2008 Enterprise SP1. Referenced as published score of 3,816,799 SPECjbb2005 bops and 119,275 bops/ JVM. Source: www.spec.org/osg/jbb2005/results/res2010q3/jbb2005-20100816-00914.html. • 32-socket Intel Xeon processor X7560 (24 M cache, 2.26 GHz, 6.40 GT/s Intel QPI): SGI Altix UV* 1000 server platform with 32x Intel Xeon processors X7560 (24 M cache, 2.26 GHz, 6.40GT/s Intel QPI), 1,048,576MB memory, SUSE* Linux Enterprise Server 11 SP1, Oracle JRockit P28.0.0. Referenced as published score of 12,665,917 SPECjbb2005 bops and 98,952 bops/JVM. Source: www.spec.org/osg/jbb2005/results/res2010q3/jbb2005-20100616-00867.html. INFORMATION IN THIS DOCUMENT IS PROVIDED IN CONNECTION WITH INTEL®_{PRODUCTS. NO LICENSE, EXPRESS OR IMPLIED, BY ESTOPPEL OR OTHERWISE,} TO ANY INTELLECTUAL PROPERTY RIGHTS IS GRANTED BY THIS DOCUMENT. EXCEPT AS PROVIDED IN INTEL’S TERMS AND CONDITIONS OF SALE FOR SUCH PRODUCTS, INTEL ASSUMES NO LIABILITY WHATSOEVER, AND INTEL DISCLAIMS ANY EXPRESS OR IMPLIED WARRANTY, RELATING TO SALE AND/OR USE OF INTEL PRODUCTS INCLUDING LIABILITY OR WARRANTIES RELATING TO FITNESS FOR A PARTICULAR PURPOSE, MERCHANTABILITY, OR INFRINGEMENT OF ANY PATENT, COPYRIGHT OR OTHER INTELLECTUAL PROPERTY RIGHT. UNLESS OTHERWISE AGREED IN WRITING BY INTEL, THE INTEL PRODUCTS ARE NOT DESIGNED NOR INTENDED FOR ANY APPLICATION IN WHICH THE FAILURE OF THE INTEL PRODUCT COULD CREATE A SITUATION WHERE PERSONAL INJURY OR DEATH MAY OCCUR. Intel may make changes to specifications and product descriptions at any time, without notice. Designers must not rely on the absence or characteristics of any features or instructions marked “reserved” or “undefined.”Intel reserves these for future definition and shall have no responsibility whatsoever for conflicts or incompatibilities arising from future changes to them. The information here is subject to change without notice. Do not finalize a design with this information. The products described in this document may contain design defects or errors known as errata which may cause the product to deviate from published specifications. Current characterized errata are available on request. Contact your local Intel sales office or your distributor to obtain the latest specifications and before placing your product order. Copies of documents which have an order number and are referenced in this document, or other Intel literature, may be obtained by calling 1-800-548-4725, or by visiting Intel’s Web site at www.intel.com. Copyright © 2011 Intel Corporation. All rights reserved. Intel, the Intel logo and Xeon are trademarks of Intel Corporation in the U.S. and other countries. *Other names and brands may be claimed as the property of others.

Re-Hosting Mainframe Applications on Intel Xeon Processor-Based Servers