Deploying servers based on live data, not guesswork
(6) Before commissioning, you need to be sure. So you run what-if scenarios to confirm maximum rack densities and to predict impacts on power, cooling, space, and connectivity. This includes analyzing aggregated impacts along the power chain to assess potential impacts (based on live data, not guess-timates). You confirm there is no adverse impact to filling out the racks.
(7) You create Work Orders for the various operations teams, with visual diagrams showing exact placement of servers and provisioning of power and connectivity. iTRACS outputs these Work Orders automatically. Each work order includes explicit instructions for both power and network connectivity. For power, the work orders graphically explain which power strip outlets to plug the left and right power supplies into, and which slots on the PDU to place the breaker to maintain phase balance. For network, they detail the entire patching scheme for each port on each server.
(8) You dispatch teams to complete the project free of errors or delays, since the Work Orders tell them exactly what to do.
The bottom line – 60 percent higher rack densities
Real-time power data gives you new levels of insight and informed decision-making. The benefits, as I’ve said, are palpable:
- Increased capacity – you now have 60% more servers in your racks than if you had settled for using inaccurate derated values.
- Higher energy efficiency – you’ve eliminated “stranded power” in under-utilized racks.
- Extended data center life – you’ve extended the life of the data center and delayed or eliminated the need for capital expenditures.
1. Nameplate values indicate max of 3 servers in these racks
2. Derated values indicate max of 5 servers
3. Intel real-time readings show max of 8 servers – 60% more. (Images courtesy of iTRACS.)
Industry Perspectives is a content channel at Data Center Knowledge highlighting thought leadership in the data center arena. See our guidelines and submission process for information on participating. View previously published Industry Perspectives in our Knowledge Library.
Pages: 1 2