This topic describes the terms that are related to Enterprise Distributed Application Service (EDAS). This helps you better understand and use EDAS.

Alibaba Cloud Toolkit
Alibaba Cloud Toolkit is a free plug-in for IntelliJ IDEA, Eclipse, and Maven. Alibaba Cloud Toolkit allows developers to use Alibaba Cloud services in their own integrated development environments (IDEs). Developers can more efficiently develop, test, diagnose, and deploy applications on Alibaba Cloud services by using Alibaba Cloud Toolkit.
Ali-Tomcat is a Servlet container that is developed based on Apache Tomcat. Ali-Tomcat supports core features of Apache Tomcat and can automatically load a Pandora container at startup to isolate classes.
CPU sharing ratio
CPU sharing is an approach to improve the resource usage of individual instances in Docker. For example, if you set the CPU sharing ratio to 1:2 for a 2-core 8 GB host, a maximum of four 1-core 2 GB Docker instances can be created on the host. If you set the CPU sharing ratio to 1:4, a maximum of eight 1-core 1 GB Docker instances can be created on the host. Note that memory resources are exclusive and cannot be shared.
Dubbo is an open source distributed service framework that supports high-performance and transparent remote procedure calls (RPCs).
Elastic Compute Service (ECS) is an easy-to-use, efficient, and scalable computing service provided by Alibaba Cloud. ECS helps you build stable and secure applications.
EDAS is a platform as a service (PaaS) offering to host applications and manage microservices. EDAS provides full-stack solutions to develop, deploy, monitor, and maintain applications. EDAS supports various microservice frameworks, such as Dubbo and Spring Cloud, which help you migrate applications to Alibaba Cloud.
EDAS Agent
EDAS Agent is a daemon installed on ECS instances. An EDAS cluster can use EDAS Agent to communicate with the applications that are deployed on ECS instances in the cluster. You can use EDAS Agent to manage applications, report status statistics, and retrieve information. EDAS Agent also serves as a communication channel between the EDAS console and your applications.
EDAS Container
EDAS Container is a basic container to run High-speed Service Framework (HSF) applications on EDAS. EDAS Container consists of Ali-Tomcat and Pandora.
K8S Namespace
Internal system objects are allocated to different namespaces. This way, you can isolate projects, teams, or user groups based on your business logic. Different groups can be separately managed while share all resources in the cluster.
Pandora (taobao-hsf.sar) is a light-weight isolated container. Pandora is used to isolate applications and middleware services, and also isolate different middleware services. This way, applications and middleware services do not affect each other. Pandora for EDAS integrates middleware plug-ins to register services, push configuration updates, and support trace query. You can use these plug-ins to monitor, process, trace, analyze, maintain, and manage EDAS applications.
Pandora Boot
Pandora Boot is developed on top of Pandora and is more light-weight. The Pandora and FatJar technologies allow you to directly launch Pandora Boot in IDEs.
A pod is the smallest deployable and billable unit in Kubernetes. A pod includes one or more containers. If a pod contains multiple containers, these containers share the computing resources, storage, IP address, and port of the pod. You can limit the proportion of computing resources that are allocated to each container.
Server Load Balancer (SLB) is a load balancing service that distributes network traffic across multiple instances. SLB distributes network traffic to backend servers to improve the responsiveness and availability of your applications. You can use SLB to prevent service interruption caused by single points of failure (SPOFs) and improve the availability of applications.
Virtual Private Cloud (VPC) allows you to build a custom, private network on Alibaba Cloud. VPCs are logically isolated from each other. You can create and manage instances, such as ECS instances, SLB instances, and ApsaraDB for RDS instances, in your VPCs.
When you deploy, start, scale in or out applications, or perform other operations to manage the lifecycle of an application in EDAS, the business logic of the entire process is abstracted and saved in application change records. You can view these application change records.
Continuous integration
Continuous integration is a software development practice where developers integrate code into a shared repository frequently, and then run automated builds and tests.
Method tracing
Method tracing uses Java virtual machine (JVM) bytecode enhancement to record the elapsed time and call sequence for all the method calls in the selected method. This allows you to check the execution sequence while the execution is in progress.
Trace query
The trace query feature is used to query status of system traces, especially for slow or abnormal services.
Distributed tracing
EDAS EagleEye analyzes every service call, message submission, and database access within a distributed system. You can use EDAS EagleEye to precisely identify system bottlenecks and potential risks.
Distributed job management
SchedulerX is a distributed job scheduling service developed by the Alibaba Cloud middleware team. You must add dependencies of SchedulerX-Client to your application, create a scheduled job in the SchedulerX console, and then set the parameters. The job is scheduled after the application is started. A SchedulerX-Server cluster can distribute jobs to a SchedulerX-Client cluster. It ensures that jobs are stably and reliably triggered and scheduled.
Billing account
A billing account is an account that is used to purchase EDAS. A billing account can be bound to at most five Alibaba Cloud accounts. The billing account must also be an Alibaba Cloud account.
Service statistics
Service statistics provide information about the status of services that are deployed in all applications for the current tenant in the last 24 hours. The information includes service calls, call duration, and call errors.
Service search
Service search is used to query services provided or consumed by the destination machine.
Service monitoring
Service monitoring monitors the service calls made by your application. It can monitor queries per second (QPS), response time, and error rate of your services.
Service degradation
In contrast to service throttling, graceful service degradation pinpoints and blocks low-performance services when your application calls services. This feature ensures the stability of your application and prevents the functionality of your application from being compromised by low-performance services. EDAS allows you to configure degradation rules based on the service response time, which effectively blocks low-performance services during peak hours.
Service throttling
EDAS can apply a number of throttling rules on each application to control service traffic and ensure service stability. EDAS allows you to create QPS-aware and thread-aware throttling rules to maximize the stability of your services during peak hours.
Service topology
The service topology intuitively presents the correlation between services and relevant performance data.
HSF is a distributed service framework for enterprise-class Internet architectures. Based on a high-performance network communication framework, HSF provides a variety of features, such as service publishing and registration, service calling, service routing, service authentication, service throttling, graceful service degradation, and trace query.
Health check
The health check feature periodically checks containers and applications, and reports the results to the console. This keeps you informed of the application status in the cluster and helps you locate and troubleshoot problems.
Infrastructure monitoring
EDAS collects data from the ECS instances that run applications and collects basic metrics for CPUs, memory, workloads, networks, and disks. All monitoring data is collected and processed by application.
A cluster is a collection of cloud resources that are required to run applications.
  • In an ECS cluster, you can deploy only one application on each ECS instance.
  • Swarm clusters will soon phase out and are unavailable to first-time users. Swarm is a container management tool released by Docker. In a Swarm cluster, you can create multiple Docker instances on each ECS instance. An application can be deployed on each Docker instance.
  • Kubernetes clusters that you create in Container Service for Kubernetes (ACK) have passed the Cloud Native Computing Foundation (CNCF) standardized tests. Kubernetes clusters run stably and integrate other Alibaba Cloud services, such as SLB and Apsara File Storage NAS (NAS). After you create a Kubernetes cluster in ACK and add it to EDAS, you can log on to EDAS and deploy applications in the Kubernetes cluster.
Namespaces are used to isolate resources and services. You can use namespaces to isolate different environments, such as development, staging, and production environments. This way, you can avoid affecting service calls and configuration update pushes in different environments.
Batch operations
In the EDAS console, you can run commands on ECS instances that have EDAS Agent installed in batches. You can run commands on all ECS instances in a cluster, ECS instances that host the same application, or only specified ECS instances. This facilitates instance management and maintenance.
Light-weight configuration center
EDAS provides a light-weight configuration center that can run in an on-premises environment. This configuration center supports service discovery and configuration management.
Log collector
Log collectors collect system monitoring logs to generate monitoring data and trace information. In a VPC, instances are isolated from on-premises machines. Log collectors can connect instances to on-premises machines. You can install a log collector to collect data from all instances that are deployed in a VPC.
Real-time logs
Real-time logs are standard runtime logs for Docker containers.
Auto scaling
The auto scaling feature monitors the status of each instance in a cluster, such as CPU, response time (RT), and load. This feature can scale the cluster in or out in real time based on the instance status. This ensures service performance and improves cluster availability.
Application Real-Time Monitoring Service (ARMS)
ARMS is an application performance management (APM) service provided by Alibaba Cloud. ARMS allows you to build a monitoring system that is capable of responding within a few seconds.
Application monitoring
This feature allows you to monitor application traffic in real time and view historical data. You can locate and troubleshoot issues based on the health of applications.
Application Configuration Management (ACM)
ACM is a utility service that centrally manages and pushes application configurations in a distributed architecture environment. In EDAS, you can manage application configurations by using ACM. In addition, you can isolate and synchronize configurations between different environments by using namespaces.
Application lifecycle
Applications are the basic manageable units in EDAS. Typically, each application contains multiple instances. EDAS allows you to manage the lifecycle of applications. You can create, deploy, start, roll back, scale out or in, stop, and delete applications in EDAS.
Application instance
An application instance is an ECS instance or elastic container instance on which you can deploy applications. In an ECS cluster, you can deploy only one application instance on each ECS instance. In a Kubernetes cluster, each pod serves as an application instance.
Application instance group
You can add ECS instances that are associated with an application to different groups. This way, you can deploy different versions of the application on ECS instances in different groups. EDAS allows you to group and manage instances that are associated with an application. You can perform beta release, A/B testing, phased release, and other operations to maintain ECS instances by group. You can manage the application lifecycle, monitor resources, and handle alerts by group. This facilitates how you can manage ECS instances.
Application runtime environment
Applications run in application runtime environments. For example, HSF applications run in EDAS containers and open source Apache Tomcat runtimes.
Application diagnosis
EDAS provides detailed troubleshooting and performance analysis for applications. EDAS provides troubleshooting and analysis for the JVM heap memory and non-heap memory on individual instances that run the current application, class loader, threads, Tomcat connector statistics, and method tracing.
Cloud Service Bus (CSB)
CSB is intended for Apsara Stack and private domains. CSB helps enterprises implement cross-system and cross-protocol communication between their systems and between enterprise systems and partner systems or third-party systems. In the EDAS console, you can create an exclusive CSB instance that allows you to expose applications in the destination environment or deploy applications to the destination environment for management. In addition, it allows you to expose EDAS applications in a VPC. This way, you can perform testing and joint debugging on them from your own development environment over the Internet.
Resource group
You can divide resources in EDAS into resource groups. You can associate ECS instances, clusters, and SLB instances with resource groups. You can also use your Alibaba Cloud account to grant RAM users permissions on resource groups.