It seems like a simple choice. You need to upgrade a platform’s computing capabilities—whether on a ground vehicle, a fast-delivery ship, a signal’s intelligence airplane or in a server room—but some of the existing hardware still is salvageable. Rather than do a complete upgrade from scratch, it is possible to leverage much of the existing technology and retain existing racks, power supplies and mass storage in the retrofit.

It makes perfect sense: Why throw away parts that seem to be working? But a closer inspection might reveal a different answer. Let’s peel back a few layers and see why—and when—it might make sense to throw away existing equipment and start over.

You never know what’s behind the walls

As with a home contractor, when vendors perform a retrofit, there’s always one line left open on the contract because they don’t know what they’re going to find. The contractor might find termites, faulty plumbing or other damage. In the high-performance computing industry, initial testing to determine the viability of existing equipment can be time-consuming and costly as the customer pays for design services and fixes.

 

---

Ben Sharfi

CEO and Chief Architect
General Micro Systems, Inc.

This email address is being protected from spambots. You need JavaScript enabled to view it.

---