[[["易于理解","easyToUnderstand","thumb-up"],["解决了我的问题","solvedMyProblem","thumb-up"],["其他","otherUp","thumb-up"]],[["很难理解","hardToUnderstand","thumb-down"],["信息或示例代码不正确","incorrectInformationOrSampleCode","thumb-down"],["没有我需要的信息/示例","missingTheInformationSamplesINeed","thumb-down"],["翻译问题","translationIssue","thumb-down"],["其他","otherDown","thumb-down"]],["最后更新时间 (UTC):2025-09-04。"],[],[],null,["# Project performance use cases\n\nThe following sections describe various ways that Performance Dashboard\ncan help you. This page describes project performance use cases. For\nGoogle Cloud performance, see\n[Google Cloud performance use cases](/network-intelligence-center/docs/performance-dashboard/concepts/use-cases-google-cloud).\n\nCurrent performance diagnostics: Is it the network or the application?\n----------------------------------------------------------------------\n\nPerformance Dashboard gives you live visibility into the\nnetwork performance underlying your project. It helps you to determine whether\napplication issues are the result of software or network problems. If you see\nsignificant packet loss or high latency, it is possible that a\nGoogle Cloud network problem is at least part of the issue. If packet loss\nand latency look normal, it is likely a problem with the application.\n\n### Scenario: Investigate an issue happening right now\n\nYou open Performance Dashboard and see a big spike in the **Packet loss\nsummary** chart within the last hour. Because this chart summarizes packet loss\nacross all zones, you don't yet know where the packet loss took place.\n\n#### Select the regions and zones\n\n- To view data for specific regions, you select the regions from the **Locations**\n menu. You can select up to five regions.\n\n- To view zone-to-zone data for a region pair, you click the square in the\n regional heatmap that corresponds to the region pair.\n\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-pl-now-blue.png) Current packet loss (click to enlarge)\n\n#### View the data for the selected time\n\nTo investigate further, you click the time that the spike took place. Doing so\nopens a heatmap specific to the selected time.\n\nYou can adjust your selection by dragging your cursor on the time axis of the\nsummary chart.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-pl-now-click-blue.png) Click a time to see details for that time (click to enlarge)\n\nThe heatmap that is displayed shows you data for the time that you selected.\nThe heatmap squares are color-coded. As summarized in the legend to the left of\nthe heatmap, each color reflects a different percentage of packet loss.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-pl-heatmap-blue.png) Packet loss heatmap for a specific time (click to enlarge)\n\nPacket loss is measured in one direction only. A square showing packet loss\nindicates packet loss *from* the zone indicated in the source axis *to* the zone\nindicated in the destination axis.\n\nTo see a chart specifically for a zone pair, you click the blue square for\nsource zone `us-central1-c`. The details\nchart keeps your time selection from the previous page.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-pl-details-blue.png) Packet loss for a selected timeframe (click to enlarge)\n\nYou see two lines on the chart, one for each direction of data flow. In this\nexample, the purple line shows packet loss for traffic from source zone\n`us-central1-c`.\n\n#### View further details\n\nThe chart shows that this spike in packet loss is an outlier. You can change the\ntime window of data displayed for this zone pair by clicking the time selector\non the top right. You can view up to six weeks of data. In this example, you\nclick **7 days** to see the packet loss trend for the selected zone pair.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-pl-details-7-days-blue.png) Packet loss for a 7-day timeframe (click to enlarge)\n\nHistorical performance diagnostics\n----------------------------------\n\nPerformance Dashboard lets you view the network performance\nwithin your project over a specific period of time in the past. You can view\nup to six weeks worth of historical data.\n\n### Scenario: Investigate an issue that happened in the recent past\n\nYou are investigating an issue with latency that happened earlier this week.\n\n#### Identify the zone with latency issues\n\nYou can use the historical performance data in the Performance Dashboard\nto examine the zone in question.\n\nTo change the view, you click the **Latency** tab.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-latency-tab-blue.png) Latency tab (click to enlarge)\n\n#### View the latency data for the selected time\n\nTo adjust the time window of the **Latency summary** chart, you use the time\nselector on the top right. In this example, it is set to **1 hour**. To see the\nheatmap for latency at a certain time, you click that time on the chart's time\naxis.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-lat-chart-blue.png) Select a time (click to enlarge)\n\nBecause there are consistently higher values on the left of the chart, you click the\ntime axis to see the heatmap for latency at that time.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-lat-chart-heatmap-blue.png) Latency heatmap (click to enlarge)\n\nThe dark blue squares in the heatmap show that the latency was 102\nmilliseconds (ms) between zones `us-central-1c` and `europe-west1-b`. To\ninvestigate further, you click a dark blue square. The latency details chart\nthat opens keeps your time selection from the previous page showing the blue\npin.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-lat-time-selection-blue.png) Latency spike (click to enlarge)\n\nThe blue line shows a spike in latency for traffic traveling from\n`us-central-1c` to `europe-west1-b`.\n\nTo zoom in on the spike, you drag your mouse.\n\nYou can now see that the spike lasted for two to three minutes, reaching a peak\nat 7:18 PM.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-lat-spike-blue.png) Latency spike details (click to enlarge)\n\n#### View further details\n\nYou can also see the details for the VMs, their projects, and\nthe aggregate hourly latency in the Latency between VMs and zones table.\nThus, you can identify exactly which VMs are contributing to the spike and\nreconsider your existing configuration or deploy new resources to address this\nissue.\n[](/static/network-intelligence-center/docs/performance-dashboard/images/pd-lat-table-blue.png) Latency between VMs and zones table (click to enlarge)\n\nWhat's next\n-----------\n\n- [Google Cloud performance use cases](/network-intelligence-center/docs/performance-dashboard/concepts/use-cases-google-cloud)\n- [Planning for performance across geographies use cases](/network-intelligence-center/docs/performance-dashboard/concepts/use-cases-performance-across-geographies)\n- [Planning for performance use cases](/network-intelligence-center/docs/performance-dashboard/concepts/use-cases-workload-optimization)\n- [Troubleshoot Performance Dashboard](/network-intelligence-center/docs/performance-dashboard/support/troubleshooting)"]]