How to manage such a sprawling system

• What metrics should/could be watched?

• Subsystems should have monitoring built into them

  • SNMP not good enough

• Centralized view

  • real time vs after the fact

Previous slide

Next slide

Back to first slide

View graphic version