| In many data centers, electricity now represents as | | | | * Rack level views of front and rear elevations |
| much as half of operating expenses. Data centers are | | | | * Device views that display consumption information |
| not using their resources effectively. The resultant | | | | Customizable Overlays and Status |
| power unleashed if operations were optimized could | | | | * The solution must be able to visually represent user |
| amount almost to a new data center that had been | | | | defined thresholds |
| "hidden". The average power consumption per server | | | | * Display coding by business service, department, or |
| rack has doubled in the past three years. | | | | customer will also be extremely useful |
| Over the next five years, 90% of businesses will see | | | | Documentation |
| their data center options halted by power failures or | | | | The system must document the data center physical |
| limits on power availability. A third of companies are | | | | infrastructure: |
| planning projects in the next 12 months...roughly 50% | | | | * Equipment layout, location, and placement |
| were planning (data center) expansions in the | | | | * Equipment Consumption of space, power, heat |
| following 12 months. Demand for data center | | | | generation, and weight |
| services will increase at 13% a year. You've heard | | | | * Power distribution system |
| the hype, we're ready to take action | | | | Relationships between all elements of the data center |
| Do we spend a lot of money on exotic solutions? Or | | | | must be tied back to business services, departments, |
| do we step back and take a more practical | | | | end users of IT services, or customers |
| approach? Here are some tips to help you approach | | | | * Devices, power connections, applications must be |
| this complicated problem and ensure the greatest | | | | identifiable by the end IT service they are associated |
| potential for success. | | | | with |
| The Outline for a Practical Approach | | | | * Each element should be able to be associated with |
| 1. | | | | multiple IT services |
| * Where it's being used in the data center | | | | * Current contact information for the end user of |
| * How effectively your power distribution and cooling | | | | the IT service should be maintained |
| systems are being used | | | | * Warranty, maintenance, service contract, service |
| 2. | | | | life, information should be maintained |
| * Plot the historical consumption levels | | | | Modeling and Analysis |
| * Then overlay plans for new services and their | | | | Basic Modeling |
| impact on consumption | | | | * Where is power be consumed and heat generated |
| 3. | | | | by (Location, Data center or room, Row, rack, or |
| * Isolate the "Hot Spots" and re-distribute the load | | | | device, etc) |
| optimize the cooling system | | | | Advanced Modeling |
| * Balance the power load across the power | | | | * Adding or removing devices debits or credits the |
| distribution system | | | | power system for available power (Circuits, PDU, |
| 4. | | | | UPS, Power Source) |
| * Now that you know where you are, you can plot a | | | | * Analyzes which devices are redundantly powered |
| course forward | | | | * PDU, Circuit and Equipment Levels |
| * Compare and contrast relevant technologies and | | | | * Simulation of failing of power infrastructure |
| solutions | | | | components: Determines effected devices, |
| * Build a business case! | | | | applications, business services |
| Selecting the Right Tool for the Job | | | | Historical and Trend Reporting |
| * You could do this on spreadsheets but, you might | | | | * Must be able to produce reports that graph the |
| grow old doing it | | | | consumption trends in the data center |
| * Better to use a solution that will automate as much | | | | * Should be able forecast future consumption based |
| of the process as possible | | | | on past consumption and expected expansion |
| * What are the requirements for such a solution? | | | | existing services and addition of new services |
| Where's the Data? | | | | Management |
| Centralized data storage & controlled access | | | | Now that we know where we've been and where |
| * It should put all of the data in a relational database | | | | we're likely to go we can analyze how new |
| such as SQL | | | | technologies might help us |
| * Once data is entered into the database and future | | | | It's about the business case |
| changes should be changed controlled | | | | * The solution must be able to use the existing |
| * It must have role based security | | | | infrastructure model as a base line to compare |
| * Read only access should be provided through a thin | | | | contrast against a proposed solution |
| client, i.e. web browser | | | | * Example: If we replace the legacy servers in the |
| The goal is to make the information: | | | | data center with blade servers |
| * As accessible as possible, while maintaining it's | | | | * Basic Analysis: |
| accuracy | | | | How much energy is saved?, Does heat generation |
| * Create and maintain relationships between the data | | | | go up or down? |
| that will allow it's modeling and analysis | | | | Is heat concentrated in specific locations vs. being |
| Visualization | | | | evenly distributed? |
| The ideal solution should provide visual access to the | | | | Is space saved? |
| dataset Data that can not be easily understood or | | | | * Advanced Analysis: |
| accessed has little value Get the big picture and then | | | | Economic impact |
| drill through to detail | | | | Analysis of the relationship between CAPEX and |
| * A top level view should be provided that | | | | OPEX |
| summarizes all of your datacenter assets | | | | Return on Investment (ROI) |
| * That view should provide configurable summary | | | | Business case and purchase justification |
| and status information | | | | To save the world and be green you must first: |
| * The user should be able to drill through to more | | | | 1. Know at a detailed level where you are |
| detail | | | | 2. Optimize what you have |
| What views should it afford? | | | | 3. Look where you've been so you can determine |
| * A geographical view that shows me all of my | | | | where you are going |
| locations | | | | 4. Consider likely growth in determine the future |
| * Location views that display top down views of my | | | | 5. |
| data centers | | | | |