ITA Capacity

In Planning > Asset Provisioning, add or move servers and other equipment to the proposed best location in a rack and U-space. Use the options in the Installation Requirements pane and view the Provisioning Map and Provisioning Match.   To provide the system with the best foundation for the calculations, first add appropriate rack PDUs, UPS, and PDUs, and configure power paths.   Select a server in Genomes to add a new server or drag an existing server from the Navigation pane to move an existing server into the layout. Review the asset information in the Installation Requirements pane, assign tags, select the required redundancy and network configuration, and select the amount to add as required. Select Best Rack to have the system automatically calculate and select the rack and U-position that is best suited for the new asset based on the capacity values (StruxureWare Data Center Operation: Capacity feature). Click Show location to highlight the best location for the selected equipment in the layout. Select Add Asset or Move Asset to allow the system to add or move the equipment to the best location in the layout.   The color-coded layout highlights rack positions as follows: Green: Optimum location for the equipment Yellow: Not optimum but possible location (may result in stranded capacity) Red: Location violating your setup (may result in power overload)   To investigate why it is not recommended to place equipment in the racks, use the Provisioning Match pane and the tool tips in the layout.   102237459_360036715033.png   See also ITA: Capacity management Working with Asset Provisioning Enabling provisioning match recommendations

You can generate capacity reports showing an overview of the capacity and capacity over time.   Power Capacity report   You can generate a report with power data, for example, available estimated load and planned estimated load.   Capacity History report   You can generate a report with a graph showing capacity data from a specified time period, e.g. the total load of all connected equipment (calculated by the system) or the measured peak load on the UPS devices (actual measured data form associated UPS devices).   The Capacity History report is available if you have a valid IT Advisor: Capacity license. Note: The report is based on data collected over the last 24 hours, so the server must have been running for at least 24 hours.   When you select the report filter options, select what is appropriate according to the data types. For example, when you are generating a report of available U-space, select a rack (not a UPS), and for a report of measured peak load, select a UPS (not a rack).   The generated report displays in the Report View. You can now use the Report View toolbar to navigate the report pages on the screen (arrow icons), or to print or export the report to a file in any of the available formats (Export icon, down arrow for selection of format).   Capacity Forecast report   You can generate a Power Capacity Forecast Report which is based on the power configuration from the modelled rooms and the planned changes within the rooms.   The report has a few parameters where Forecast Date will take work orders planned until and including the date selected. Redundancy Selector will be used in calculating the capacity for both actual and forecast. The report will show the results both at room, location and global level.   Rack PDU Power Overview report   For new rack PDUs with outlets and banks definition user can generate detail report with power capacity values on banks, outlets. Report can also show measured peak value for each bank, outlet if 'Show measurements' checkbox is enabled. For better phase recognition user can set individual color in the options section. Report can be generated directly for rack or rack PDU in the floor overlay by selecting it from context menu.   102237362_360036174534.png See also Working with Reports

When talking about data center capacity, we talk about power, space or cooling capacity.   You may experience one or more of these typical power, space or cooling capacity issues in your data center (see screenshot).   When you have modelled the data center in IT Advisor, use the add-on capacity management module to design to solve these issues.   IT Advisor: Capacity makes daily tasks significantly easier for you by applying advanced calculations on your design. The asset provisioning functionality will guide you to obtain the optimum data center design, such as suggesting the best location for your next equipment because of the right amount of free space, cooling, power, and redundancy available.   Note: Using the Capacity module requires a preconfigured and valid power path, including an upstream connection to one of these power sources: UPS Input breaker on a PDU/Power Panel (assuming it is connected to the mains supply)   ITA_3_capacity_problems_1_360036714733.png   How does ITA know what the available capacity is?   The system calculates how much capacity is used to guide you on how much capacity is available. This happens automatically and you don't need to set anything up for the system to provide you with guidance, other than modelling your power equipment in the system, including correct breaker values, and a valid power path.   In addition to the power information available in each rack (physical breaker limits), the system relies on the enabled power capacity strategy for calculating estimated load. Again, you don't need to set anything up, and in most cases, you can go with the default settings. However, an option to change strategy is available if you want to explore other options.   Power information on equipment in the racks   The available capacity is defined by the physical breaker limits on the rack PDUs and receptacles in the racks, and upstream limitations (UPS capacity), or you can choose to introduce additional design limit configurations if you like to plan capacity based on custom predefined values on racks and rack PDUs. Breaker size (derated capacity of the breakers in the power path) UPS capacity (upstream limitation) Design limits (on racks and rack PDUs)   Capacity strategies See Understanding estimated load and how it is calculated   Where are the calculations used?   The system calculates power consumption for estimated load and reserves capacity for advanced setups, including failover load, PSU uncertainty, and distribution redundancy. The tooltips provide an easy overview of power values relevant to your specific setup, showing the numbers and a graph (including phase balancing bars).   Examples of capacity calculations in various redundancy setups   These simple examples illustrate how the capacity is calculated in various redundancy setups.   Tooltip with capacity details ITA_tooltip_capacity_details_360036714733.png   Capacity Management Finding the best location for a new server   Tags Creating tags Assigning tags   Cooling Working with Capture Index Designs with perforated tiles on a raised floor Working with the 3D temperature map   Reports Capacity reports   Getting live data from external systems Associating Live Data from an External System to the Layout   Impact Analysis on Virtual Machine Hosts VMware integration enabling migration of virtual machines impacted by alarms Locating virtual machines hosts with impact in the layout   IT Impact Simulation Simulating impact Configuring simulated impact layout Filtering and finding equipment in simulated impact list Exporting List View table

The estimated load is a value used by the system to calculate how much capacity is used to be able to calculate and guide you on how much capacity is available. By default estimated load calculations are based on adjusted nameplate values. However, they can be different depending on the power information available to the system in each rack and the power capacity strategy you have selected as the main strategy in Preferences > Power Capacity.  The values and graphics in the Power overlay illustrate how close your estimated load is to the breaker capacity or design limit based on your predefined thresholds. See the introduction to the different strategies below to learn how they work and see the consequences of each strategy to the calculation of your estimated load here.    Adjusted nameplate You can reduce the manufacturer's nameplate value to the adjusted value when you know the actual power consumption of the product and allow the system to perform more precise calculations. The adjusted nameplate value is more precise than the manufacturer's nameplate as it is the actual known peak power draw of a server rather than the maximum load that the manufacturer guarantees the server will not exceed. If the system doesn't have this value available, it will fall back on the manufacturer's nameplate. Use the adjusted nameplate strategy when you know the adjusted nameplate values of your equipment and you want more precise calculations than what the manufacturer's nameplate enables you to get (less pessimistic). The adjusted nameplate values must be entered manually in the equipment's property settings. If the system does not know about the phase connections, it will assume the server draws the load on all phases in a 3-phase setup and triple the load to ensure the UPS/PDU can handle the load. To avoid this, configure how the servers are connected to the rack PDUs.   Predicted power Power capacity calculations based on measured peak data from DCE or other external system integrations allow the system to predict power based on already collected measurements. You can customize the default implementation of 30 days trending and fixed minimum and maximum values (Preferences > External Systems > Measurement Trending). Use the predicted power strategy if you have measured power values from your equipment and you want the capacity calculations to be based on the measured values rather than nameplate values. See more about predicted power here.   Contracted power (colo) The contracted power for the rack is distributed on the rack PDUs or receptacles and used as estimated load. The contracted power is defined in the Colo overlay. You cannot actively select this strategy in the system. It is used if you are a colocation provider and you haven't set up connected equipment with measured values from the contracted equipment. Thresholds in Preferences > Power Capacity allow you to define the percentage of the contracted power to be used as estimated load.

Depending on the the nature of the modeled equipment, redundancy settings, and power path setup in IT Advisor, the capacity calculations can get quite complex. If you have a similar setup as one of these simplified examples, they may help you understand where the numbers in the tooltips come from.   Example 1: Multiple power supplies The data center solution includes: 1 large server, 3000 W load 6 power supplies Capacity values: In previous versions of ITA, the Estimated Load would be very high. In the current version, it now works as follows: Estimated Load: 500W on each connection (even load) Estimated Load: 1000W on each connection (uneven load) Power Supply Unit Uncertainty: 500W  (uneven load) If the server only uses half of the PSUs, the system must reserve up to: (3000W) / (6/2) = 1000W on each connection. This implies a Power Supply Uncertainty of 500W (because 1000W - Estimated Load = 500W). To minimize the impact of PSU Uncertainty on the PDU, the PSU Uncertainty be only be carried upstream to the PDU in case of phase shifts. Blade solutions: If a blade enclosure includes bays powered by specific PSUs, the enclosure may only be using e.g. 2 out of 6 PSUs because the other bays are empty. In this case, the PSU Uncertainty reservation done by the system might be too small!   Example 2: Classic 2N and multiple power supplies The data center solution includes: 1 large server, 3000 W load 6 power supplies 3 connections to each of two separate rack PDUs Capacity values: Estimated load: 500 W on each connection Failover load: 500 W on each connection If the failover load can carry the full power supply uncertainty (as in this example), power supply uncertainty is not added to the calculations.   Example 3: Reservation for Distribution Redundancy The data center solution includes: 1 server, 1000 W load 2 connections to two different PDUs (L1 and L2) Both PDUs are connected to the same parent PDU Capacity values: Load overview Load type and phase PDU 1 PDU 2 Parent PDU Estimated Load, L1 500 W 0 W 500 W Estimated Load, L2 0 W 500 W 500 W Failover Load, L1 500 W 0 W - Failover Load, L2 0 W 500 W - Reserved for Distribution Redundancy, L1 - - 500 W Reserved for Distribution Redundancy, L2 - - 500 W   On the parent PDU there is no failover load because additional load will not move to this PDU in case of a downstream failure. However, a phase shift might happen in case of a downstream failure because of the mixed phases on the server. (If the server would have been connected to only one single phase, there would be no need for reserving extra capacity.) Example 4: Reservation for Distribution Redundancy with opposite effects The data center solution includes: 1 server, 1000 W load 2 connections to two different PDUs (L1 and L2)   1 server, 600 W load  2 connections to two different PDUs (L2 and L1)   Capacity values: Load overview Load type and phase PDU 1 PDU 2 Parent PDU Estimated Load, L1 500 W 300 W 800 W Estimated Load, L2 300 W 500 W 800 W Failover Load, L1 500 W 300 W - Failover Load, L2 300 W 500 W - Reserved for Distribution Redundancy, L1 - - 200 W Reserved for Distribution Redundancy, L2 - - 200 W   The two servers partly cancel out the need to reserve extra capacity.    

Power measurements can come from two sources: Data Center Expert The ETL import database. This includes importing manual measurements by uploading an Excel sheet. The measurements tracked in ITA are: Peak power Average power Peak amps Predicted power, calculated based on the historical peak power measurements The most interesting value for managing the data center is the peak power measurement observed within the last 30 days (configurable). This is the value typically shown in the user interface. Predicted power is also actively used throughout ITA as estimated load. You can set this under Power Capacity; set Estimated Load Strategy to Predicted Power. When this strategy is used, power measurements are prioritized in capacity planning calculations. The flow of power measurements is designed around the integration with Data Center Expert. The ETL integration tries to mimic the same behavior. Polling for measurements After ITA polls DCE for peak power measurements every 5 minutes (configurable), two things can happen: If it has been less than 24 hours since a full update of the power measurements was done, only the peak measurements are updated. If it has been more than 24 hours since a full update of the power measurements was done, there is a full update (peak and average for the last 24 hours). Peak update of power measurements ITA maintains values for the highest peak measurement observed over the last 30 days, configurable for the integration with DCE. During a peak update, this value is updated if a larger peak measurement was found, that is, recent measurements affect the peak value. If a larger peak value was found, the predicted power values are recalculated. Full update of power measurements Measurements during the last 24 hours are polled, then peak and average measurements are stored as historical values in the database. If there are no measurements for devices, and there are already historical measurements, the historical value is copied and reused it as the measurement for the last 24 hours. The highest observed peak values for both amps and power are recalculated. A job is started that recalculates the predicted power values. Special cases There is a special case when there have been no measurements for several days. This typically happens right after you set up ITA, or when the server was stopped for multiple days. ITA performs a catch-up process to get measurements from as far back as 30 days (not configurable). Catch-up means that ITA looks at the latest measurement, up to 30 days back, and the time this measurement was taken. Then the measurements for 24 hours immediately after that are polled. If no measurement is found, the existing measurement is duplicated with a new timestamp. This process is repeated when there is less than 24 hours between the current time and the last measurement time. Examples There are no historical measurements because the IT server was just installed and started: 30 day catch-up is performed The server was shutdown for 10 days: 10 day catch-up is performed When are measurements reset? The 30 days that the peak calculations look back is configurable for the DCE integration. Sometimes, when making changes to the database, the historical peak values are no longer interesting. There are two cases where ITA marks measurements older than some timestamp as invalid. When equipment is removed from a rack When you remove equipment from a rack, the historical measurements are no longer valid. ITA saves a timestamp for the most recent time when equipment was deleted from a rack, and does not look further back than this timestamp for the measurements it uses. Just after a delete, no measurement is available, so ITA shows the previous measurement until a new one is available. When the measurements comes from an integration with DCE, this typically happens within a few minutes. If no measurement is available within the next 24 hours, the previous measurement will be used for that day, even if it was taken before the timestamp, since no other measurement is available. This is important to note when importing manual measurements through Excel, or when using the ETL database in a non-automated way. When the assigned customer changes The process described above happens for relevant equipment when you update the customer assigned to a rack or cage. Another thing to note is when equipment is planned to be added using work orders (requires the Change module), ITA tries to include the equipment in the capacity calculations. Since the equipment is not installed yet, it has not yet affected the measurement; when using the predicted power capacity strategy, the adjusted nameplate of planned equipment is added to the estimated load. What happens when the highest observed peak was 30 days ago? Since the user interface shows the highest peak over the last 30 days, it is interesting to understand what happens when the highest peak measurement becomes older that 30 days. In the typical setup with DCE, there are unique measurements for each day; when the previously highest peak becomes 31 days old, the new highest peak from the last 30 days is shown. How does this work with manual breaker panel measurements? This is, in principle, the same thing that happens when you have manual breaker panel measurements uploaded in an Excel sheet. However, with manual measurements, a new measurement is typically not made every day. This means the previous measurement is duplicated for each day you do not have a measurement. If you observe and import a measurement of 10 amps on day 1, and import an updated measurement of 5 amps on day 15, the measurement from day 1 is duplicated for each day from 2 to 14. Only after 30 days, on day 45, will the 10 amps measurement leave the history and the value in ITA change to 5 amps.

In Tools > Preferences > Capacity History, you can set up the system to collect and save Key Performance Indicators (KPIs) for a specified period of time. This is an IT Advisor: Capacity feature. Once you have enabled the feature and configured when you want the system to collect the data, and for how long you want it to be saved, the system stores the data on the server and makes it possible for you to access and analyze critical data over time. You can use the data for creating custom trending reports, display on the dashboard and integrate with external systems through web services. Read about the API For example, a standard built-in report allows you to create a report with the current amount of available U-positions in one or more data centers. However, you may want to know the development over time in available u-space. The saved KPI data allows you to access this data. 102237487_360036714973.png 123 Steps to... Setting up the system to save KPI data In Tools>Preferences>Capacity History, define the KPI data recordings. In Record data, select the interval between the recordings, e.g. Daily. In Clear data, select the interval between clearing the data, e.g. After 1 year, or select Never to keep storing the data indefinitely. You can start using the data for a trending report after more than one day. Creating a trending report In Analytics>Reports, select Capacity History in the list of available reports and define the filter options. 1. Select a rack and a Date Range, for example the last week. 2. Select to include the data type, for example Available U-space, and click to generate the report. The capacity history trending report shows with a graph of the amount of available U-positions per day in the Report View and can be printed or exported. Tip If you are planning to reuse the same selections, e.g. for a monthly report showing the development in space capacity, click the Save Filter As icon and name the report template, e.g. Monthly U-space, to have it preset and ready for you to push the generate button once a month. Exporting capacity history When you have generated a report, you can export and use the data in other applications, e.g. Excel or .csv format by clicking the Export icon at the top of the report view. See also IT Advisor: Capacity management Capacity history KPI description Working with Reports Working with Report Filter Templates

  Capacity history data captured in EcoStruxure IT Advisor on a daily basis:   Capacity History Data Type (KPI) Area/license Description Data Equipment Level Available U-space (AVAILABLE_USPACE) Capacity The currently available rack u-space Location, Room, Rack Reserved U-space (RESERVED_USPACE) Capacity The rack u-space currently reserved by planned changes Location, Room, Rack Total used U-space (USED_USPACE) Capacity The currently used rack u-space Location, Room, Rack Available copper (AVAILABLE_COPPER_NETWORK) Capacity The number of available copper network ports in layer 2 & 3 network equipment Location, Room, Rack Available fiber (AVAILABLE_FIBER_NETWORK) Capacity The number of available fiber network ports in layer 2 & 3 network equipment Location, Room, Rack Available network ports (AVAILABLE_NETWORK_PORTS) Capacity The number of available network ports in layer 2 & 3 network equipment Location, Room, Rack Used copper (USED_COPPER_NETWORK) Capacity The number of used copper network ports in layer 2 & 3 network equipment Location, Room, Rack Used fiber (USED_FIBER_NETWORK) Capacity The number of used fiber network ports in layer 2 & 3 network equipment Location, Room, Rack Used network ports (USED_NETWORK_PORTS) Capacity The number of used network ports in layer 2 & 3 network equipment Location, Room, Rack Ambient temperature ((AMBIENT_TEMPERATURE) Capacity The calculated ambient ("background") temperature. This is the temperature in the room when not in the immediate vicinity of racks, coolers etc. Room Cold aisle capture index (COLD_AISLE_CAPTURE_INDEX) Capacity The cold aisle capture index UPS, Rectifier, Rack Cooler capacity (COOLING_CAPACITY) Capacity The total capacity of coolers Location, Room Cooler load (COOLING_LOAD) Capacity The calculated load of coolers Location, Room, CRAC Hot aisle capture index (HOT_AISLE_CAPTURE_INDEX) Capacity The hot aisle capture index UPS, Rectifier, Rack Inlet temperature (INLET_TEMPERATURE) Capacity The calculated inlet temperature UPS, Rectifier, Rack Auxiliary load connected (AUXILIARY_LOAD_CONNECTED) Capacity Get the estimated load for auxiliary equipment. This is equipment that is not in racks, such as coolers Room Estimated Load (ESTIMATED_LOAD) Capacity Get the estimated load Location, Room, Cage, UPS, Rectifier, PDU, Rack, Rack PDU, Internal Breakers Estimated failover load (ESTIMATED_TRANSFERRED_LOAD) Capacity Estimated failover load Location, Room, UPS, Rectifier, PDU, Rack PDU Estimated PSU uncertainty (RESERVED_FOR_PSU_UNCERTAINTY) Capacity Estimated power reserved for PSU uncertainty Location, Room, UPS, Rectifier, PDU, Rack, Rack PDU Estimated distribution redundancy (RESERVED_FOR_DISTRIBUTION_REDUNDANCY) Capacity Estimated power reserved for distribution redundancy Location, Room, UPS, Rectifier, PDU, Rack, Rack PDU Measured peak load (MEASURED_PEAK_LOAD) Capacity Get the measured peak power Location, Room, UPS, Rectifier, Rack, Cage, PDU, Rack PDU, Internal Breakers, Power Outlets Measured failover load (MEASURED_TRANSFERRED_LOAD) Capacity Measured failover load Location, UPS, Rectifier, PDU Measured peak load per phase(MEASURED_PEAK_LOAD_PER_PHASE) Capacity  Measured peak load per phase for three phase equipment. UPS, Rectifier, PDU, Rack PDU Power capacity (POWER_CAPACITY) Capacity Power capacity Location, Room, UPS, Rectifier, PDU Total estimated load (TOTAL_ESTIMATED_LOAD) Capacity Get the estimated load for equipment in racks and cages Location, Room, Rack, Cage Total load connected (TOTAL_LOAD_CONNECTED) Capacity Get the total nameplate load of all equipment Location, Room, Rack Estimated load per top redundancy item (ESTIMATED_LOAD_PER_TOP_LEVEL_ITEM) Capacity Estimated load for each top level redundancy item of a rack Rack Estimated failover load per top redundancy item (FAILOVER_LOAD_PER_TOP_LEVEL_ITEM) Capacity Estimated failoverload for each top level redundancy item of a rack Rack  Estimated PSU uncertainty per top redundancy item (RESERVED_FOR_PSU_UNCERTAINTY_PER_TOP_LEVEL_ITEM) Capacity Estimated power reserved for PSU uncertainty for each top level redundancy item of a rack Rack  Measured average load per top redundancy item (MEASURED_FOR_TOP_LEVEL_ITEM_AVERAGE) Capacity Measured average load for each top level redundancy item of a rack Rack Measured peak load per top redundancy item (MEASURED_FOR_TOP_LEVEL_ITEM_PEAK) Capacity  Measured peak load for each top level redundancy item of a rack Rack Free outlets (FREE_POWER_OUTLET_CAPACITY)  Capacity Number of free rack PDU power outlets grouped by outlet type Location, Room, Rack, Rack PDU, Internal Breakers Used outlets (USED_POWER_OUTLET_CAPACITY)  Capacity Number of used rack PDU power outlets grouped by outlet type Location, Room, Rack, Rack PDU, Internal Breakers Total number of outlets (TOTAL_POWER_OUTLET_CAPACITY)  Capacity Total rack PDU power outlets capacity grouped by outlet type Location, Room, Rack, Rack PDU, Internal Breakers Cancelled work orders (CANCELLED_WORK_ORDERS) Change Total number of cancelled work orders Location, Room Work orders not started  (NOT_STARTED_WORK_ORDERS) Change Not started work orders, which are not overdue Location, Room Work orders in progress (IN_PROGRESS_WORK_ORDERS) Change In progress work orders Location, Room Overdue work orders (OVERDUE_WORK_ORDERS) Change Overdue work orders Location, Room Completed work orders (COMPLETED_WORK_ORDERS) Change Total number of completed work orders Location, Room Total work orders (TOTAL_WORK_ORDERS) Change Total number of work orders Location, Room Revenue space (ROOM_CAGES_AREA) Co-Lo Room cages area. Location, Room Sold space (ROOM_CLOSED_AREA) Co-Lo Room area sold to customers Location, Room Power contracted with customers (ROOM_CLOSED_POWER_SOLD) Co-Lo Power already sold to customers for all equipment in a room. Location, Room Closed racks (ROOM_CLOSED_RACKS) Co-Lo Racks in room sold to customers. Location, Room Floor-mounted area (ROOM_FLOOR_MOUNTED_AREA) Co-Lo Room floor standing equipment total area Location, Room Internal space (ROOM_INTERNAL_AREA) Co-Lo Room internal area. Location, Room Internal racks (ROOM_INTERNAL_RACKS) Co-Lo Racks in room for internal usage. Location, Room Open space (ROOM_OPEN_AREA) Co-Lo Room area open for new customers. Location, Room Open racks (ROOM_OPEN_RACKS) Co-Lo Racks in room open for new customers Location, Room Racks utilization (ROOM_RACKS_UTILIZATION) Co-Lo Room racks utilization percentage. Racks utilization is sum of open, closed and reserved racks divided by total number of racks. Location, Room Reserved space (ROOM_RESERVED_AREA) Co-Lo Room area reserved for potential customers. Location, Room Power reserved for customers (ROOM_RESERVED_POWER_SOLD) Co-Lo Power reserved for potential customers for all equipment in a room Location, Room Reserved racks (ROOM_RESERVED_RACKS) Co-Lo Room area reserved for potential customers. Location, Room Sellable area (ROOM_SELLABLE_AREA) Co-Lo Sellable area: sum of Open, Closed and Reserved area. Location, Room Space utilization (ROOM_SPACE_UTILIZATION) Co-Lo Sellable area as a percentage of total room area. Location, Room Closed racks (standalone) (ROOM_STANDALONE_CLOSED_RACKS) Co-Lo Racks in room sold to customers. Only standalone (outside cages) racks are counted. Location, Room Internal racks (standalone) (ROOM_STANDALONE_INTERNAL_RACKS) Co-Lo Racks in room for internal usage. Only standalone (outside cages) racks are counted. Location, Room Open racks (standalone) (ROOM_STANDALONE_OPEN_RACKS) Co-Lo Racks in room open for new customers. Only standalone (outside cages) racks are counted. Location, Room Total racks (standalone) (ROOM_STANDALONE_RACKS) Co-Lo Room total racks number. Location, Room Racks utilization (standalone) (ROOM_STANDALONE_RACKS_UTILIZATION) Co-Lo Room racks utilization percentage. Racks utilization is sum of open, closed and reserved racks divided by total number of racks. Only standalone (outside cages) racks are counted. Location, Room Reserved racks (standalone) (ROOM_STANDALONE_RESERVED_RACKS) Co-Lo Room area reserved for potential customers. Only standalone (outside cages) racks are counted. Location, Room Room area (ROOM_TOTAL_AREA) Co-Lo Room total area. Location, Room Total contracted power (ROOM_TOTAL_POWER_SOLD) Co-Lo Total power sold to customers for all equipment in a room. Location, Room Total racks (ROOM_TOTAL_RACKS) Co-Lo Total number of racks in a room Location, Room Above temperature threshold (ABOVE_TEMPERATURE_THRESHOLD) Cooling Optimize Relative time that the temperature for a rack exceeds its threshold (with a higher value) Rack Below temperature threshold (BELOW_TEMPERATURE_THRESHOLD) Cooling Optimize Relative time that the temperature for a rack exceeds its threshold (with a lower value). Rack PUE Energy Efficiency The PUE This is for energy systems only Location Total network ports (TOTAL_NETWORK_PORTS) Operations The total number of network ports in layer 2 & 3 network equipment Room, Rack Total U-space (TOTAL_USPACE) Operations The total rack u-space Location, Room, Rack Total weight (TOTAL_WEIGHT) Operations The total weight of equipment Rack Server quantity (SERVER_QUANTITY) Operations The number of servers mounted in racks Location, Room, Rack Total IT Assets (TOTAL_ITOASSETS) IT Optimize Total number of servers and blades in a room Location, Room Not associated IT Assets (UNPLACED_ITOASSETS) IT Optimize Number of not associated ITO assets Location, Room IT Asset CPU utilization (UTILIZATION_CPU_AVG) IT Optimize Get the average CPU utilization (average of associated) Location, Room IT Asset Power utilization (UTILIZATION_POWER_AVG) IT Optimize Get the power utilization (sum of associated averages) Location, Room Consolidation Candidates (NUM_CONSOLIDATION_CANDIDATES) IT Optimize Number of retirement candidates (server utilization between 2% and 10%) Location, Room Retirement Candidates (NUM_RETIREMENT_CANDIDATES) IT Optimize Number of retirement candidates (server utilization below 2%) Location, Room Consolidation Candidates costs (POWER_COSTS_CONSOLIDATION_CANDIDATE_PRICE) IT Optimize The power costs for consolidation candidates Location, Room IT Asset costs (tag) (ENERGY_COST_PER_TAG) IT Optimize Power costs for all servers with a given tag, per room Location, Room IT Asset costs (POWER_COSTS_PRICE) IT Optimize The power cost setup for a room Location, Room Retirement Candidates costs (POWER_COSTS_RETIREMENT_CANDIDATE_PRICE) IT Optimize The Power Costs for retirement candidates Location, Room     Capacity history data discontinued in version 8.0: Capacity History Data Type (KPI) Area/license Description Data Equipment Level Measured A-Feed average (MEASURED_A_FEED_AVERAGE) Capacity Get the measured average power on the A-feed UPS, Rack Measured A-Feed peak (MEASURED_A_FEED_PEAK) Capacity Get the measured peak power on the A-feed UPS, Rack Measured B-Feed average (MEASURED_B_FEED_AVERAGE) Capacity Get the measured average power on the B-feed UPS, Rack Measured B-Feed peak (MEASURED_B_FEED_PEAK) Capacity Get the measured peak power on the B-feed UPS, Rack Total reserved capacity (TOTAL_RESERVED_CAPACITY) Capacity Get the reserved capacity cf. capacity groups plus auxiliary equipment such as coolers Location, Room

Power cable connections between rack PDUs and servers are automatically calculated by default. Turning off the automatic cable connections allows you to overrule the default settings to manually configure cable connections. This is useful for example when configuring an already running data center with a power configuration that is not the most efficient. Once you have modeled the data center in IT Advisor, you can use the capacity features to identify and optimize the design. Configuration Toggling cable connection settings is available in Tools>Preferences>Cable Connection if you have IT Advisor: Capacity installed. Open Tools > Preferences > Cable Connection. Define the default power connection setup for a server when it is added to the layout from Genomes. In Cable Connection when moving a server, select Auto calculate cable connection or No cable connection. Define the default power connection setup for a server when it is added to the layout from another position in the Rack Layout or Navigation. In Cable Connection when moving a server, select Auto calculate cable connection or No cable connection. 102237469_360036714873.png Usage When you configure manual cable connections, it is important that you remember that any server you add or move to a rack must be manually connected to a rack PDU immediately. Otherwise, you will see load values that are too high. If the system does not know about the phase connections, it will assume the server draws the load on all phases in a 3-phase setup and triple the load to ensure the UPS can handle the load. Note: If you add equipment using the rack advisor functionality in the Installation Requirements pane, cable connections are automatically calculated regardless of the selection of cable connection preferences. See also ITA: Capacity management Reviewing system preferences

The estimated load calculations always fall back on the best available information in the rack. Depending on the available information in the rack, your calculations may switch strategy. Adjusted nameplate This illustration shows how, if the Adjusted Nameplate strategy is the main strategy (default), information available in the rack (left) defines which strategy is enabled (right). 105474238_360036174434.png   Predicted power This illustration shows how, if the Predicted Power strategy is the main strategy, information available in the rack (left) defines which strategy is enabled (right). 105474239_360036174434.png

What is predicted power and how is it calculated? Predicted power is a capacity strategy defining how estimated load is calculated. The calculations run in the background and affect capacity values all over the ITA interface, such as impact and failover load. Past trends in your equipment's measured peak power data from DCE or other external system integration are used to predict the future power demand. By default, the system bases the calculations on historic data from the last 90 days and predicts 30 days in to the future. You can adjust these settings in Preferences > External Systems > Measurement Trending. Note: The graphs here serve as examples only. They do not reflect any graphs available in ITA or in reports. The first graph illustrates the predicted power concept and the rest are examples of special scenarios to explain why your calculations behave the way they do.   105471637_693x250_360036174414.png Predicted power (measured peak power consumption on rack PDU) How is the prediction affected when adding the first server to an empty rack? Due to the added load, the trend would become unrealistically steep over a period of 30 days. Therefore, a maximum factor of 1.5 (by default) is used to keep the trend within a more appropriate range. Note: When adding the new server, you may see some unexpected values while you're editing. This happens because the system falls back on adjusted nameplate calculations until you save your changes or, if you have IT Advisor: Change, until you execute the pending change.   105471638_683x250_360036174414.png Predicted power (when adding the first server to a rack) How is the prediction affected when your measurements are stabile? If the measurements have been stabile for the period of time the system has data for this rack, the system does not rely on this data to be realistic until there's measured data for at least 90 days (default). The trend line would become almost flat. Therefore, a minimum factor of 1.2 (by default) is used as a safety margin to keep the trend line within a more appropriate range. A lower threshold of 1 also prevents the trend line from becoming negative. However, if the measurements have been stabile for an even longer period of time (more than 90 days by default), the data is assumed to be highly reliable and the safety margin is disregarded. 105471639_670x250_360036174414.png Predicted power (when measurements are stabile) How is the prediction affected when removing a server from a rack? Due to the reduction in load, the trend would drop unrealistically over a period of 30 days. When you have removed a server from a rack and saved your changes, the trend line will no longer include values from the last 90 days but only values since the server was removed to exclude values from removed equipment no longer relevant to you.

Power Supply Unit (PSU) Uncertainty indicates an uncertainty in the power supply path as to which phase is carrying the load from a server or chassis. Many servers distribute the load evenly across multiple power supplies; some power supplies are most efficient when operated at a higher load. This gives an incentive to turn off power supplies for servers or blade chassis with multiple power supplies, meaning the real consumption may be higher on some power supplies than if the load was evenly distributed. Since ITA doesn't model power supplies, it is unknown which power supplies may be turned off by the server or blade chassis. If the server or blade chassis is connected to different phases, there is uncertainty as to which phases are carrying the load. To make sure you do not exceed your capacity, ITA reserves some of the capacity for PSU uncertainty which you can see expressed as PSU uncertainty in the power tool tips.