How does load balancing work in a dual-redundant mission computer with two CPUs, and what criteria determine failover direction?

How does load balancing work in a dual-redundant mission computer with two CPUs, and what criteria determine failover direction?

• A. Distributes tasks evenly across CPUs by a fixed partition.
• B. Runs all tasks on the primary CPU and ignores the secondary.
• C. Distributes tasks; if one CPU fails, tasks migrate; failover decisions based on health indicators, error rates, and performance metrics.
• D. Failover direction is random.

Answer: (C)

Load balancing in a dual-redundant mission computer relies on dynamic task distribution and health-based failover. The system allocates work across both CPUs to use resources efficiently and keep response times predictable. When everything is healthy, tasks are spread to maintain balance and meet performance goals; if one CPU shows signs of problems or becomes overloaded, tasks are migrated to the other CPU to maintain operation and meet timing requirements. Failover decisions are driven by concrete signals: health indicators (such as heartbeat status and watchdog status), error rates (fault counts, retries, and incorrect results), and performance metrics (CPU utilization, latency, throughput, memory pressure, and thermal conditions). This combination ensures that critical functions stay online and the transition is smooth, rather than relying on a fixed partition or random selection.

Load balancing in a dual-redundant mission computer relies on dynamic task distribution and health-based failover. The system allocates work across both CPUs to use resources efficiently and keep response times predictable. When everything is healthy, tasks are spread to maintain balance and meet performance goals; if one CPU shows signs of problems or becomes overloaded, tasks are migrated to the other CPU to maintain operation and meet timing requirements. Failover decisions are driven by concrete signals: health indicators (such as heartbeat status and watchdog status), error rates (fault counts, retries, and incorrect results), and performance metrics (CPU utilization, latency, throughput, memory pressure, and thermal conditions). This combination ensures that critical functions stay online and the transition is smooth, rather than relying on a fixed partition or random selection.