ProductsDesktop Server Red Hat Enterprise Linux OpenStack Platform For IBM POWER For IBM System z For SAP Business Applications Satellite Management For Scientific ComputingExtended Update Support High Availability High Performance Network Load Balancer Resilient Storage Scalable File System Smart Management Extended Lifecycle SupportAccelerate Automate Integrate Red Hat JBoss BPM Suite Red Hat JBoss Developer Studio Portfolio Edition Web Framework Kit Application Platform Web Server Data Grid Portal Fuse Red Hat JBoss A-MQ BRMS Red Hat JBoss Fuse Service Works JBoss Operations Network JBoss Community or JBoss enterprise Red Hat JBoss Data Virtualization
SolutionsWhy Red Hat Why open hybrid cloud? The new IT Public cloud Cloud resource library Private cloud Infrastructure-as-a-Service (IaaS) Platform-as-a-Service (PaaS) Cloud applications and workloadsSolaris to Red Hat Enterprise Linux Migration overview Migrate from your UNIX platform How to migrate to Red Hat Enterprise Linux Upgrade to the latest Red Hat Enterprise Linux release JBoss Enterprise Middleware Benefits of migrating to Red Hat Enterprise Linux Migration services Start a conversation with Red Hat
TrainingPopular and new courses Red Hat JBoss Administration curriculum Core System Administration curriculum Red Hat JBoss Middleware Development curriculum Advanced System Administration curriculum Linux Development curriculum Cloud Computing, Virtualization, and Storage curriculum
ConsultingSOA and integration Business process management Cloud and virtualization Custom Software Development Enterprise Data and Storage Systems management Migrations
ANSES Recognizes Cost Savings and Greater Flexibility with JBoss Enterprise Application Platform
April 4, 2011
Customer: Administración Nacional de la Seguridad Social (ANSES)
To find a solution that would address multiple challenges including: Business recording changes in laws without impacting transnational systems; allowing non-IT professionals to make legal and regulatory changes without manipulating source code; separating management and disbursement processes from applications; and guarantee the seamless operation of systems at all times.
JBoss Enterprise Application Platform 4.3, JBoss BRMS (Business Rules Management System) 5.0, JBoss Data Services Platform (DSP) 5.5
SUN M9000 with 8 Quad Core Sparc VII processors of 64 bits and 192 Gbytes of Ram, Storage Tek 6740 and SO Solaris 10.
The Administración Nacional de la Seguridad Social (ANSES, in its Spanish acronym) is a decentralized gov- ernment agency within the scope of the Social Security Secretariat. ANSES reports to the Ministry of Labor, Employment and Social Security of the Government of the Argentine Republic. ANSESs’ main function is to manage and disburse all social security contributions in Argentina destined for active and retired workers and their families, including national retirement and pension funds, funds of employees under an employment relationship and self-employed workers, subsidies, family allowances, and unemployment benefits. ANSES has 14,000 employees and 600 offices spanning the Argentine territory connected through a modern multi-protocol label switching (MPLS) network. Its systems are centralized in a datacenter that has two primary functions, running the online applications that support 600 customer service locations, and running web applications that the social security system beneficiaries access through a web portal. Each month these systems perform about 170 million transactions. The processes engaged in the settlement of payments involve over 1.5 million monthly records. In order to support this entire infrastructure, ANSES’ IT department has a permanent staff of 490 employees, 180 of which are developers. This area is also assisted by 100 to 300 independent developers depending on the demand.
Over the years, ANSES employees have developed a broad knowledge base and expertise on the processes and rules used to settle and manage social security payments in Argentina. During the last 50 years, not only have new concepts and methods of settling benefits been created, but laws have been subject to numerous changes as well.
Significant efforts have been made to translate and incorporate the expertise of employees into systems that ensure a single source of information at the time of making material decisions, like the right to benefit payment. However, in the case of entitlements, evidence has demonstrated that the same situation could be successfully handled differently at various customer service locations based on the expertise of individuals serving in specific regions. As such, ANSES needed a system that could both systematically account for rules and changing laws and be flexible enough to accommodate specific situations. In 1996, an attempt was made to organize the regulation of the social security system according to a business rules format based on the proportional logical writing method. At that time technology was incapable of supporting the 60 MB of spreadsheets resulting from such work.
Due to the high volatility of business rules that are based on regulations and laws, keeping the systems in perfect working order is extremely difficult. The successive changes in the Argentine retirement system over the years caused many of the primary systems to become unpredictable given the complexity of their logic and their many outdated components. What makes this issue even more complicated is the need to link systems and rules to laws and regulations. Rules must be prioritized based on the administrative attributes (laws, opinions, requests for amparo - a form of constitutional relief found in the legal systems supporting them). This complexity is increased when laws amend other rules that must govern the set of benefits falling within the rule in force at the time of its issuance. Therefore, the systems must be capable of understanding and managing which set of rules are applicable to each case. This reality entails an intricate network of dependencies and hierarchies between laws and rules which is very difficult to organize in closed information systems. In order to appropriately and successfully handle the complexity, ANSES incorporated within its modernization program (PRISSA) a solution to the logical volume problem and the permanent growth of these business rules. The main goal of PRISSA was to cut system management time for the IT department by separating the tasks required to successfully maintain the system. Under the new structure, the Standards and Process Management team, which is the relevant legal authority, would control the definition of rules, while the IT department would be solely responsible for the definition of the architecture, data modeling, resource deployment and other tasks that traditionally fall to IT teams. This new process represented a paradigm shift, where developers and business specialists share the workload in order to more efficiency and effectively manage the entire architecture that controls the management of millions of transactions.
Additionally, the system needed an easy-to-understand graphical user interface that non-technical business analysts could manipulate. Attorneys, legal assistants and business managers at ANSES would need the basic knowledge of desktop tools. This new structure would allow rules to be deployed in an inference engine and executed by means of transnational systems.
As part of the modernization project, a solution called Business Rules Generation Assistant (APRN, in its Spanish acronym) was designed based on JBoss Enterprise Business Rules Management System. The APRN allows for the folllowing:
- The development, testing, and documentation of rules by people only knowledgeable about office automation tools by means of a Wizard Web interface
- Data discovery in order to eliminate loopholes when the law fails to indicate what to do and when the conditions being regulated have values different from those anticipated
- Aligning each decision rule to the regulation supporting it (law, rule, request for amparo, etc.)
- Data modeling and simulation of the behavior of the rule with different values of productive data, by answering the question “what if”
- Ensuring the accurate implementation of new rules without affecting transnational systems
ANSES leveraged the flexibility and robust combination of a number of solutions within the JBoss Enterprise Middleware portfolio. Because the standards-based open source middleware architecture sits between the application and the operating platform, it serves to seamlessly manage loading, prioritizing and rule simulation (APRN). Additionally, JBoss Enterprise Middleware serves as the transnational system infrastructure layer, operating and warehousing the historical data sources of the organization. ANSES deployed a combination of JBoss Enterprise Middleware solutions to create a flexible, efficient and simplified architecture including:
- Rule Generation Assistant-Seam-based web application running on JBoss Enterprise Application Platform 4.3, which generates a rules meta-code in MVEL
- Rule execution and testing-Rules in MVEL are executed for simulation in JBoss Enterprise BRMS
- Real-time rule and data simulation logical entities involved in the rules through APRN. These entities are materialized as Java objects and associated with a DSP virtual view by the IT staff. The virtual view is finally associated with the actual data required for the view and may reside in relational databases, web services, VSAM files or structured text. After rules are compiled and virtual views are associated, the Rule Generation Assistant allows simulation of the execution of a rule against a real set of organization’s data in real time.
The Rule Generation Assistant was used in virtual partitions of a Sun M9000 unit with 8 Quad Core Spark VII 64 bit processors and 192 GB RAM, Storage Tek 6740 and Solaris 10 OS, SQL Server and DB2 databases supported in Mainframe z/10 of 900 mips and connected through a DB2 connect. There are also over 250 servers running on various operating platforms including Linux and Windows.
A critical issue is the volume of information within the organization’s databases which contain records for the entire active and retired population of Argentina which totals about 4 billion. The records of social security beneficiaries amount to 1.4 billion and the history of the social security current account is 275 million.
At present there are 50 new applications being developed which will have their logic externalized in the rules engine, without their own business logic.
By unifying the decision criteria of the entire organization, ANSES was able to eliminate the potential for discretionary use and fraud furthering a reduction in the number of lawsuits (approximately 30,000 lawsuits are filed against ANSES each month), resulting in a direct impact on the solution ROI. The development team’s application maintenance workload was also reduced allowing resources to be allocated to the design of new solutions.
Response time to change was minimized due to the ability to take immediate action to build system solutions. Response time is further minimized when this benefit is coupled with the swift implementation of new programs and benefits by legal authorities. Ultimately, the result is an improvement in the quality of life of the beneficiaries.
ANSES will migrate from their existing Mainframe environment (4.5 million dollars on equipment cost plus 1.4 million dollars for maintenance and annual support) to a lower-cost open system. Using Red Hat’s open source technology reduced the total cost of the solution and has allowed for the allocation of a greater portion of the budget to train internal resources and develop a task force fit for the challenge. The Red Hat suite of solutions and services also allowed ANSES to realize other open source advantages, such as lower cost of acquisition, along with the support of proprietary systems and the guaranteed presence of a partner to accompany and ensure project success. Gains in efficiency were also realized due to access to open source industry and community experts and most notably, product experts.