CUNA Mutual
From Devwiki
CUNA Mutual Group, Member Services Enterprise (MSE)is a credit union insurance and financial services company. It serves credit unions in 57 countries with a menu of over 150 products and services. No other company in the world offers so many credit union-specific products. CUNA Mutual Group has 4,500 employees and assets of more than $6.5 billion.
The pioneers of the credit union movement established the CUNA Mutual Insurance Society in 1935. CUNA Mutual Insurance Society is the parent organization of all the companies that together form CUNA Mutual Group. As a mutual company, CUNA Mutual belongs to its credit union and credit union member policy owners, and operates to serve their best interests.
The confidence and trust credit unions have in CUNA Mutual is reflected in the fact that it protects, with one or more coverages, 99 percent of 11,700 credit unions in the United States. In addition to achieving positive financial results, CUNA Mutual believes it is essential to provide, on a competitive basis, every service that can contribute to the success and security of credit unions and the members they serve.
In addition to insurance and services for credit unions, CUNA Mutual's Member Services Division also provides financial products or insurance coverage to approximately 30,000,000 credit union members across the United States.
Contents[hide] |
[edit] Objectivity Case History
[edit] Important
This information is an archive, so any use of the present sense in the text should be taken in the historical context, generally determinable from the Status section below.
[edit] Customer Information
- Customer: CUNA Mutual Insurance Society, Member Services Enterprise
- Project: Member Service Relationship Complex (MSRC)
- Location: WI
- Territory:
- Industry Verticals: Complex Financial
- Technologies: Very Large Database, Knowledge Management
- Application Domain: Consolidation of Membership Services Relationships.
- Market Characterization: Relationship Hunting
- Number of developer licenses:
- Runtime license volume and type:
[edit] Status
- First Contact: 1998
- Lead came from: ParcPlace
- Evaluation Start Date: 1998
- Evaluation Finish Date: 1998
- First Purchase Date: 1998
- Deployment Date: 1999
- Current Status: Fully deployed.
- Can we talk about this customer and the product/project? Yes.
- Referenceable?: Yes.
[edit] Environment
- Hardware: Sun
- Operating System: Solaris
- Precision: 32 bit.
- Development language: Smalltalk, evolving to Java
- Compiler: Visualworks Smalltalk. Sun JDE.
- Third Party vendor tools: ENVY
- Open Source tools: Linux (Trial use only)
[edit] The Project/Product
[edit] Project Background
CUNA Mutual spent several Million Dollars trying to implement a consolidated member database using IBM DB2, but the system was struggling because of the variety of data types and multiple join operations. They had built the system in Smalltalk and looked around for an alternative database.
In January of 1998 CUNA set out to help member credit unions offer better services to their members by creating a consolidated member database with Objectivity. Previously data has been spread out over disparate, policy number keyed databases rather than one with a member focus. This new database was important to sell more services to member credit unions so they could better serve (and market to) their customers. In other words, this would add more dollars on the bottom-line for CUNA.
Several issues had to be resolved. This included figuring out consolidation of the multiple databases, loading Objectivity and designing high-performance queries. The data needing consolidation consisted of over 50 million records from over 5000 sources in over 4000 file formats representing over 30 million individuals.
The system is running successfully and saves CUNA Mutual over $1.5 Million per year.
[edit] Project/Product Description
The Member Service Relationship Complex (MSRC) is the core of a technical infrastructure that enables member centric business activities. The MSRC creates and maintains knowledge of credit union members relationships with CUNA Mutual Group, facilitate the consistency of this knowledge across its legacy applications, and provides enabling services to business activities through application level programs, such as a contact management system.
So far they have been able to consolidate 50 million old database records into an object database that is nearly one Terabyte in size. By taking advantage of parallel processing against multiple Objectivity Databases they were able to load 18,000 objects and 18,000 hash table entries per second while achieving lookup times of .001 second.
The initial load of the database consists of loading the records and creating sorted proxies. A sophisticated rule based match of proxies is then used in conjunction with rule based source selection (determining, for example which address and phone number for John Smith is the most recent) for business object creation. The final step is to create lookup data structures. The data structures are multi-level hierarchies that hash into a container. These provide enough information on a member to display and select from which the full set of member information can be accessed via object ID. Additional processes have been created for ongoing updates.
The MSRC is a large and complex undertaking that was developed using a divide and conquer strategy. The objective was to implement the strategic components using a logical, well-engineered approach. The next step in building the MSRC involved developing a business object model that captured the significant relationships with individual credit union members. The goal was to capture a substantial portion of these complex relationships and store them in an object database. Some database statistics:
- 100 Million records from 5000 sources.
- 30 Million customers.
- 4000 different data formats.
- Updated nightly and used by almost all North American Credit Unions.
[edit] Buying Criteria
[edit] Business Priorities
- Stable vendor
- Pricing
[edit] Technical Priorities
The following were priorities during the decision process.
- Scalability. The Objectivity database is greater than a Terabyte in size, holds more than 100 million records and six (6) billion objects. The customer realized early on that no RDBMS could scale to this size and offer performance with as many relationships.
- Performance. In an environment of this size, the sheer number of tables that would be required by an RDB would most certainly drag RDB performance down. This is a large-volume, scalable, high-performance environment. How high? During a recent benchmark the Objectivity database delivered sub-second response time. Even more impressive, the benchmark was run against a cold database, one that had just been populated but never viewed. In this transactional test the first member's profile (with all active financial product offerings displayed) was brought from NT server to user screen with a response time of LESS than ONE second.
- Language heterogeneity was another key priority. The customer has a mixed language application. Thus, they wanted the ability to create an object in C++ and view that object via the Smalltalk binding. Or, create in Smalltalk and view via the C++ binding. With Objectivity they found complete language independence. No other product that this customer evaluated offered equal capabilities across language bindings.
[edit] Competitors/Alternatives
- GemStone, ObjectStore and Versant
[edit] Why They Chose Objectivity
- A pilot project quickly verified the robustness and scalability of Objectivity/DB.
- Lack of strong Smalltalk support from Versant and no support from ObjectStore.
[edit] Partners
- ParcPlace
[edit] Collateral
- Press Releases: July 11, 2005
- Flyers: None
- White Papers:
- Case Study: Thumbnail sketch
- Other:
[edit] Contact Information
- Objectivity Rep:
- Customer Contact:
- Customer Phone:
- Customer Email:
- URL: CUNA Mutual
