Skip to main content
バージョンタグがご希望のコンテンツと一致しているかご確認ください。 または、最新バージョンをご利用ください。

PDCを使用したアプリケーションパフォーマンスの分析

Pega Predictive Diagnostic Cloud(PDC)は、Pega Cloud®サービス上で動作するSaaS(Software as a Service)ツールです。 すべてのアクティブなPega Platform™アプリケーションからリアルタイムのパフォーマンスおよび健全性の指標をアクティブに収集、監視、分析します。

Pega Platformは、分析に必要なクリップボードパラメーターデータのみを使用することで、既知の安全なコンテンツをPDCに送信します。 監視対象となるPega Platformの実装へのパフォーマンスの影響を最小限にするため、データは非同期で送信されます。 個人を特定できる情報を含むデータ(例:ケースデータ)は、プライバシーの観点から送信されず、PDCが監視対象となるアプリケーションからの情報を要求することもありません。

Pega PlatformはPDCのために以下のデータを収集します。

  • Alerts:アラートのタイプおよびインタラクションで取得されたメタデータを特定します。
  • Parameter page:現在のパラメーターページのパラメーターについて、監視対象のアプリケーションで実行される関数についての重要なコンテキスト情報などについて説明します。 残りのパラメーターはすべてフィルタリングされて除外されます。
  • Database alerts:ビジネスデータ値を除いたデータベースクエリーの詳細情報が含まれます。
  • Exceptions:アラート用に送信されたフィールドのサブセットであるコンテキストデータが含まれています。
  • Performance statistics:平均応答時間とユニークユーザー数の統計が含まれます。 PDCはこれらの統計情報を基に、監視対象アプリケーションの全体的なパフォーマンスやパフォーマンスの傾向を把握します。
  • Database indexes:データベースのインデックス情報が含まれます。 PDCは、データベースのインデックス情報を利用して、クエリーのパフォーマンスを向上させるための推奨事項を生成します。

PDCダッシュボード

PDCはPega Platformアプリケーションから生成されたデータを分析および集約し、トレンドダッシュボードを作成します。 トレンドダッシュボードでは、毎日の平均レスポンス使用量、インタラクションボリューム、ユーザーセッション数などが図表で表示されます。 トレンドグラフでデータを表示することで、経時的な動きを表し、問題の監視、切り分け、特定、予測に役立てることができます。 システムパフォーマンスやビジネスロジックの潜在的な問題を予測し、修正案を提示することで、ビジネスのステークホルダーやIT管理者が予防的なアクションを取れるよう促します。

PDCはアラート、使用およびリソースの状況、システム健全性パルスからのデータを分析および集約して、次の図のようなトレンドグラフを生成します。

Overview of PDC

PDCを使用して、オンプレミスおよびクラウドベースの複数のPega Platformアプリケーションを監視できます。 Pega Cloud上で動作するシステムには、すでにPDCが統合されており、Pega Cloudが示すURLからPDCにアクセスできます。

補足: PDCに関するアクセスと使用の詳細については、「Logging in to Pega Predictive Diagnostic Cloud」を参照してください。

以下のインタラクションで理解度をチェックしてください。

Issue Identification and research with Pega Predictive Diagnostic Cloud

Pega Predictive Diagnostic Cloud™ (PDC) helps you identify areas that you need to work on to improve your system's performance. Use Improvement Plan to view a summary of issues that affect the performance of your monitored system. Use Event Viewer to conduct an in-depth investigation

The following articles guide you through common use cases for identifying issues in your system:

Resolving issues in your system with Pega Predictive Diagnostic Cloud

Resolve common performance issues and find the root cause by using the Pega Predictive Diagnostic Cloud (PDC) problem-solving tools. With Event Viewer, you can access diagnostic data from all the nodes in your system.

To start resolving a performance or stability issue, consider the following actions:

  • In Improvement Plan, open a case, and then follow the recommended next steps.
  • In Event Viewer, view the log data from the past 14 days, and then analyze the issues that occurred during that time.

For example, if you know when your system failed, you can verify the alerts and exceptions that were generated at that time. By analyzing these events, you can determine the problem that caused the selected system to fail.

  • To quickly find the root cause of an issue, organize diagnostic data with customizable sets of filters.

For example, narrow the possible sources of diminished performance to a single user, node, or application.

For more information about event types in PDC, see Cases in Pega Predictive Diagnostic Cloud.

以下のインタラクションで理解度をチェックしてください。

Notifications in Pega Predictive Diagnostic Cloud

Pega Predictive Diagnostic Cloud (PDC) provides numerous measurements of your system that you can receive as messages to help you address any issues that might occur. With these messages, called notifications, you can immediately react to important issues in your system or counteract potential stability problems.

To receive only the measurements that are relevant and useful for you, determine what data you need to receive on a regular basis and what information is useful only when you troubleshoot a specific issue. The type and frequency of notifications that can help you the most depend on your user role.

Notifications are available only for events that are important and require immediate attention or additional action.

PDC supports the following types of notifications:
Daily digest notifications
Periodic summaries of issues that PDC detects in your system. These notifications focus on specific categories, for example, Improvement Plan TOP 5 Web events or Improvement Plan TOP 5 Overall issues. You can configure PDC to send daily digests in any range from daily to weekly notifications. This type of notifications is most useful for manager roles for a general overview of your system health.
Event-based notifications
Messages that PDC sends when a specific number of the same events occur during a certain timespan. You can configure the event threshold and the time interval for each notification. This type of notifications is most useful for operations team members who can use them to swiftly react to stability and performance issues.

You need to create separate notifications for each of your systems that PDC monitors.

Choosing daily digest notifications

As part of system governance, subscribers receive the following scheduled digests to keep them up to date about the status of a monitored system:
Cases Summary
A summary of the active cases in PDC for the selected systems. Cases require at least one alert in the last 7 days to be considered active. The data is collected each day at 0:00 GMT. The report contains the active cases for a 7-day period, the new cases that were created in the last 24-hour period, and the number of unassigned cases.
To see the details of the new cases, see the New items daily report.
CDH Usage Summary
A summary of Pega Customer Decision Hub™ usage metrics with information on the month to date, the current day, the total usage, and the trend in usage data.
Event Summary
A summary of events that occurred over a 24-hour period on the selected systems.
Improvement Plan Top 5 Background
A report on up to the top five most urgent issues that affect background processing for each of the categories on the selected systems (Database, Connectors, General Performance, Stability, Runtime Guardrails, Custom Cases, and Decisioning). The report displays the amount of time the system spent processing background work and shows the percentage of time spent last week compared to the time spent this week.
Improvement Plan Top 5 Overall
A complete Improvement Plan for the selected system with up to the top five most urgent issues overall for each of the categories on the selected systems. The report displays the amount of time that the system spent processing web, service, and background work and shows the percentage of time spent last week compared to the time spent this week.
Improvement Plan Top 5 Service
A report on up to the top five most urgent issues that affect services for each of the categories on the selected systems. The report displays the amount of time that the system spent processing service requests and shows the percentage of time spent last week compared to the time spent this week.
Improvement Plan Top 5 Web
A report on up to the top five most urgent issues that affect the web for each of the categories on the selected systems. The report displays the amount of time the system spent processing web requests and shows the percentage of time spent last week compared to the time spent this week.
New items daily
A list of new issues that have been created in the last 24 hours.

Choosing event-based notifications

Note: To view all events for which PDC can send notifications, see List of notifications available in Pega Predictive Diagnostic Cloud.

Each notification informs you about an issue in your system, such as unresponsive components, exceeded time thresholds, or low resources. PDC can detect hundreds of thousands of events, and you cannot react to each event individually. Therefore, subscribe only to notifications about the events that are most relevant to your system. Because every application uses different components in different ways, you choose notifications manually.

As a best practice, review the following notification types:
  • Focus on the events that can affect the stability of your system. These events are marked as Recommended.
  • Check the list for events that require urgent attention and subscribe to the events that apply to your system. These events are marked as Urgent.
  • Check the list for infrastructure events that help you to manage resources inside your system. These events are marked as Infrastructure resources.
  • Look at other notifications and subscribe to the types that are most relevant to your system.

After you use PDC for some time, you can revise the subscriptions to stay informed about the events that are most relevant to your system and application.

During subscription, operators can add more than one account to each notification.

By default, PDC sends notifications by email. However you can configure PDC to send individual notifications by SMS, notification API, or to ServiceNow.

For more information about configuring notifications, see Managing notifications in Pega Predictive Diagnostic Cloud.

Notification API

You can further customize how you receive notifications using the notification API feature in PDC. With the notification API, your system can receive notifications directly from the API.

To connect your system to the notification API, the system must support the OAuth protocol and be able to relay notifications to end users. During configuration, PDC generates the following authentication data:
  • Endpoint URLs for token authorization
  • Client ID
  • Client secret
You then use this data to access the API. For more information, see Subscribing to Pega Predictive Diagnostic Cloud notifications by using a REST API.

Onboarding for PDC

You can learn more about PDC through the following articles:

  1. Fundamentals of PDC
  2. Event Viewer in PDC
  3. Notifications in PDC
  4. Cases in PDC
  5. Improvement Plan
  6. Integrations with project management tools

以下のインタラクションで理解度をチェックしてください。


このトピックは、下記のモジュールにも含まれています。

トレーニングを実施中に問題が発生した場合は、Pega Academy Support FAQsをご確認ください。

このコンテンツは役に立ちましたか?

改善できるところはありますか?

We'd prefer it if you saw us at our best.

Pega Academy has detected you are using a browser which may prevent you from experiencing the site as intended. To improve your experience, please update your browser.

Close Deprecation Notice