COTS lifecycle challenges; or how to choose the best COTS suppliers

In the last eight to 10 years the use of commercial off-the-shelf (COTS) electronic hardware and software has experienced a quick start but a tough implementation from the COTS supplier's view and the defense contractor's view.

By Edmond Hennessy

In the last eight to 10 years the use of commercial off-the-shelf (COTS) electronic hardware and software has experienced a quick start but a tough implementation from the COTS supplier's view and the defense contractor's view.

While the initial goals of COTS — such as cost reduction, performance improvement, and accelerated development cycles — have become reality to some degree, using COTS often creates as much complexity as it has created simplicity.

Across a program's life, developers must contend with ever-shorter product life cycles from the COTS vendors, which in turn introduce technology refresh, insertion, and obsolescence management issues.

Further, the number of organizations involved in each procurement provides additional complexity. A typical program involves the procuring government agency, the defense contractor/integrator, and numerous commercial suppliers — each of which has a different view of program and product life cycles and business models.

The typical COTS supplier's product life cycle is two to three years. The typical defense contractor's program life cycle is seven to 15 years. The typical government program lifecycle and sustainable platform life is 25 to 40 years. These disparities have forced the defense contractors and the COTS supplier base to look more closely at the costs of technology refresh, insertion, and obsolescence management. One example of the life of a military platform is the P-3 Orion which entered military service in 1962 and is still in use today and is still being upgraded.

Leaders of the U.S. Department of Defense (DOD) realize they must revise how they view program life cycle by building in shorter design and technology-insertion cycles. Successive refinements, rather than a series of point-designs, are the approach that best fits the COTS model. Historically, low startup costs and the need for continuity between generations of a product design were among the reasons that COTS succeeded. COTS brought rapid prototyping, accelerated implementation, and reduced complexity. These now are regarded as the real benefits of COTS.

Choosing a COTS supplier
In a typical program life, defense contractors incur 60 to 70 percent of their costs after initial deployment in maintainability, reliability, and supportability programs. Defense contractors have always considered these life cycle issues when choosing COTS suppliers. However, most COTS suppliers are not familiar enough with life cycle costing and programmatics to provide more than initial product and maintenance pricing.

Further, while hardware costs across the life cycle are easily quantified, the majority of "hidden" costs are software-related. The selection criteria of COTS suppliers needs to extend beyond hardware evaluation to the evaluation of software tools that can provide quick application implementation and a standard application programming interface (API) that enables application code to migrate easily to new technology without expensive recoding and re-optimization.

Government agencies and defense contractors cooperate in choosing the COTS supplier on any project. The government has established standard cost criteria that include several elements in how to chose COTS suppliers. This information provides a realistic lifecycle cost model, which provides the government and military with solid insights into projected cost behavior and estimated cost of ownership. Often these elements are involved in the selection of COTS suppliers:

  • hardware related costs;
  • initial product price of development;
  • initial price of commercially packaged COTS product;
  • initial product price of deployment;
  • initial price of deployable/rugged COTS product;
  • installation costs;
  • initial cost to install and test equipment;
  • maintenance costs;
  • system installation and MTB unit repair costs;
  • training costs;
  • learning costs, including training manuals;
  • costs of operation;
  • lifecycle replacement costs;
  • fault isolation and redundancy costs;
  • support costs;
  • integrated logistics costs; and
  • software costs;
  • technology refresh costs;
  • costs of porting and re-optimization of software; and
  • supplier performance risks

Defense contractors and the government use a similar set of evaluation criteria, yet profit motive and competitive pressures distinguish the supplier model from the government model. The long-range upgrade plan for technology insertion — including the latest hardware, software development, and maintenance costs — have the most substantial influence on profits.

Developing these costing elements makes the COTS supplier much more valuable to the defense contractor and builds long-term relationships between the two organizations that can continue through out the program life.

Surprisingly, most COTS suppliers are not able to provide life cycle statistics; some lack any experience in military deployment/production. Many companies are able to offer defense contractors initial pricing models but are unable to get realistic predictions of product and cost performance over the program lifecycle. In today's competitive electronics market place, only a few companies — including DY 4 Systems in Kanata, Ontario, and Sky Computers in Chelmsford, Mass. — have made a commitment to life cycle planning and analysis for the defense community. If a defense contractor cannot get these critical life cycle predictions from their COTS supplier, they need to look elsewhere for a supplier.

From development to deployment
A program's requirement for rugged packaging for deployment introduces a new set of complexities. Each COTS supplier has a different approach to ruggedizing its products. Historically suppliers Like DY 4, Radstone Technology in Towcester, England, and Vista Controls in Santa Clarita, Calif., developed products for extremely harsh environments. The specialized technologies of these companies result in products that are expensive and typically that do not keep pace with the industry standard price/performance ratio (Moore's Law).

In terms of performance per dollar, per watt, and per square inch, their design goals focus ultimately on environmental requirements that are most extreme. A standard axiom in this segment is the trade-off of conformance to environmental requirements with cost and performance.

Today some companies are introducing a family of products that use common technology for each phase of the program cycle. One source of commercial-to-rugged technologies simplifies the development to deployment dilemma. With the recent announcement of its Xtreme product line, Sky Computers provides several levels of packaging for harsh environments, thereby providing the government and defense contractors with one source of multiprocessor technology.

Technology refresh and insertion
Over the course of a program's life cycle are several technology refresh and technology insertion points. Driving many of these upgrades are changing application requirements, and such upgrades demand flexible and powerful technologies that far exceed the capabilities of the original design.

In addition to the considering impact of hardware changes to a program upgrade, systems designers cannot underestimate the impact of software migration and supportability costs across a program's lifecycle. Software development can represent a $15 million to $20 million multigenerational investment with software maintenance tallying another $2 million to $3 million. Choosing a COTS supplier with a standard API and tools for quick application implementation can reduce the program's overall software development and optimization effort by as much as 30 to 50 percent. This savings provides "quick program turns" to support military preparedness/readiness.

Managing upgrade costs by using portable software is one way to keep the software lifecycle costs in check, and is another key element in COTS supplier selection. COTS suppliers must focus on methods for containing upgrade costs and providing a path for continuous performance upgrades and cost reductions over the program lifecycle.

There are two key considerations: code portability, and code optimization. Companies such as Sky Computers that provide open architectures and standard APIs can support a technology upgrade with tools that easily port software from one product generation to the next. With just a few instructions, the code ports to the new target platform. Sky refers to this as "compile and run" and ensures 100 percent code compatibility. As importantly, the code is migrated to the new platform fully retaining its optimization. No additional time-consuming hand coding and performance tweaking is required.

Edmond Hennessy is the worldwide vice president of sales and marketing at Sky Computers in Chelmsford, Mass. For more information, contact Sky Computers by phone at 978-250 1920, by fax at 978-250-0036, or on the World Wide Web at http://www.skycomputers.com/.

More in Communications