White Paper

3 The Need for Accurate Thermal Modeling So far we have covered the increased power dissipation that results from the utilization of advanced FPGA devices. We now turn our focus to designing FPGA modules taking into account these new constraints— specifically, we discuss accurate power modeling. If we are able to predict the power usage of the FPGA devices, we can design power supplies capable of providing adequate current as well as the cooling systems required to dissipate the heat. Neglecting to plan for the realistic levels of power dissipation causes integration issues down the road. While the module might operate with a very simple test algorithm, the end user might find that their algorithm results in a power dissipation that causes such a high thermal rise that the device reliability or performance is significantly degraded. When this type of shortcoming is discovered at the integration stage, the resolution is not only expensive, but causes major schedule delays. Traditional FPGA Power Modeling Standard design methodology uses the power model provided by the FPGA device manufacturer. However, in order to yield an accurate result, these models require detailed inputs that are not always known to the FPGA module designer—especially during the initial phases of the design. In order to deal with the lack of power modeling accuracy, the designer typically has two options. The first option is to over-engineer the power supply and cooling systems. Using the most conservative analysis and adding margin, the designer can select a power supply that far exceeds the expected current draw of the devices. Additionally, increasing the physical size of the module and including extra thermal conduction pathways will allow the module to handle the power dissipation—even if the actual power levels vary from the rough model. While this approach will likely result in a product that meets the power specifications, it will be larger and more expensive than needed. Since most systems are limited in size, weight and cost, an over-engineered solution will likely present problems. The second option is to perform multiple design iterations. In the first iteration the designer uses their best guesses for the size of the power supply and cooling architecture. After a prototype is built, the power is measured and a second iteration is designed to account for the actual levels. While also resulting in a product that meets the power specifications, this approach delays the schedule and increases the design cost. Additionally, in order for this approach to be successful, the FPGA module designer must have detailed knowledge of the specific end-users algorithm. Without an accurate representation of the algorithm, the FPGA module designer is unable to measure the actual power used by the prototype. A new method is needed that not only accurately models the FPGA power for specific types of algorithms but also can be applied during the early design phases. A Novel Approach to FPGA Power Modeling At Mercury Systems, our power estimation method divides the modeling process into different phases based on the stage of the design. With this approach, an early power prediction at the beginning of the development process provides the designer with the information for the power supply design as well as the mechanical cooling elements. Using the results of this early power prediction modeling, a prototype is built. After fabrication, the power is measured by installing our validation IP, which has been designed to realistically emulate customer algorithms, into the FPGA devices. The results of this testing are fed back into the model and used to improve the accuracy for future designs. POWER METHODOLOGY PRODUCT CONCEPT FPGA POWER ESTIMATES DESIGN POWER SUPPLIES BA SELINE THERMAL MODELING CREATE POWER LOAD DESIGNS TEST POWER SUPPLIES DESIGN BOA RDS DESIGN COOLING SOLUTION FABRICATE COOLING SOLUTION FABRICATE BOA RDS DEPLOYMENT THERMAL MODELING DEPLOYMENT GUIDANCE DESIGN COOLING SOLUTION FABRICATE COOLING SOLUTION COMPONENT POWER ESTIMATES POWER ESTIMATION AND MODELING KNOWLEDGE FPGAS COMPONENTS THERMAL MO DELING BA SELINE THERMAL TESTING DEPLOYMENT THERMAL TESTING LEGEND MECHANICAL HARDWARE FIRMWARE POWER ESTIMATION AND MODELING CORRELATION FPGAS COMPONENTS THERMAL MO DELING POWER ESTIMATION AND MODELING CORRELATION FPGAS COMPONENTS THERMAL MO DELING Standardized Universal PCBA AFB/LFT [VITA 48.X] AFB [VITA 48.7] AFT [VITA 48.8] LFT [VITA 48.4] CC [VITA 48.2] AC [VITA 48.1] Mercury's Power Modeling Process

• Cover