SOFAStack is built on a financial-grade distributed middleware system developed by Ant Group over more than ten years. It incorporates architectural practices refined in Alipay’s core financial transaction systems. All SOFAStack products and technologies are rigorously tested in Ant Group’s demanding financial scenarios to secure transactions and enable agile business development. SOFAStack also provides geo-disaster recovery and low-cost, rapid scaling, helping you overcome the challenges of migrating from a traditional centralized architecture. By enabling large-scale, high-availability distributed systems, SOFAStack supports innovation in your financial business.
SOFAStack consists of four main modules: distributed middleware, operations management (application PaaS), advanced O&M assurance, and R&D efficiency (DevOps).
Distributed middleware
Ant Group’s distributed middleware follows the principle of leading and embracing advanced industry standards while supporting smooth migration and integration of traditional financial architectures to reliably handle business upgrades. The middleware addresses key challenges in financial transaction systems, such as service and data scalability, transactional consistency, second-level disaster recovery, and elastic resource scheduling.
Dual-mode microservices
The SOFAStack Microservice Platform uses SOFAStack Microservice (SOFAStack MS) and SOFAStack Mesh to manage and govern microservices, supporting both the SOFA framework and the Service Mesh architecture.
Microservice: Provides microservices for the SOFA framework, including RPC services, dynamic configuration, and throttling and circuit breaking.
Service Mesh: Uses Service Mesh technology to support native Dubbo, Spring Cloud, and SOFA frameworks, offering non-intrusive service management and administration for Dubbo, Spring Cloud, and SOFA applications.
Service architecture
Benefits
High-performance distributed service framework
Provides high-performance, transparent remote procedure call (RPC) services with high scalability and fault tolerance.
Microservice administration center
Provides a series of service administration policies to ensure high-quality service operation and meet service level agreements (SLAs).
Highly reliable and lightweight configuration center
Provides a service for dynamically modifying configurations at runtime, featuring a graphical, centralized management interface.
Active-active data centers
Supports active-active architectures within the same city or across regions, delivering geo-disaster recovery capabilities to ensure system availability.
Scenarios
Modernizing traditional applications with microservices |
Use microservice products to break down traditional financial business systems into a modular, standardized, loosely coupled, pluggable, and extensible microservices model. This shortens the time-to-market, allows for quick product listing, and helps capture market opportunities. It not only ensures efficient customer service but also reduces operational costs. |
Rapidly scaling high-concurrency services |
Develop Internet finance services using microservice products to improve R&D efficiency, respond more flexibly to business changes, and rapidly iterate on innovative products. Scale out hot spot modules to increase processing capacity and easily handle traffic bursts. This also improves user experience and provides more small and micro-enterprises with personalized financial products and convenient, low-cost financial services. |
Active geo-redundancy with multiple data centers |
Use microservice products to quickly build a scalable, high-performance, financial-grade distributed core system with elastic scaling and active geo-redundancy capabilities. |
Task Scheduler
Task Scheduler (TS) provides a distributed task scheduling framework for distributed processing, enabling standardized, automated, visual, and centralized scheduling, monitoring, and O&M for tasks across different business systems in a financial enterprise.
Service architecture
Benefits
Support for cluster tasks
Supports sharding data by custom dimensions with no limit on the number of splitting layers, enabling full utilization of all machines in a cluster. When data volume is large, multilayer scheduling tasks can process data faster. It also supports simple, single-machine tasks.
Support for visual dependency orchestration of tasks
Orchestrates multiple tasks into a topology graph based on execution order. Tasks run automatically without manual intervention, ensuring many scheduled tasks complete quickly and efficiently.
Rich task configuration items
Provides a wide range of configuration options, including multiple routing policies, failure retry modes, and missed-trigger handling policies. It also supports online configuration of input parameters for scheduled tasks.
Flexible trigger control
Supports triggering tasks on a schedule, periodically, by an event, or manually.
Scenarios
Batch processing in the finance industry |
This scenario requires processing a specific amount of data within a set time. The traditional method processes data based on database shards, where each machine handles one shard. The drawback is that if the number of machines exceeds the number of shards, many machines remain idle, resulting in low resource utilization. Cluster tasks let you split the data yourself into index blocks, assigning each machine in the cluster a portion to process. This approach is flexible and enables task execution monitoring. |
Daily cutover in the finance industry |
Daily cutover in the finance industry is a complex operation requiring multiple tasks to execute in a specific order. Traditionally, the execution order is hard-coded into the system. When business requirements change, code modifications are necessary, resulting in poor maintainability. Use the task orchestration feature in SOFAStack Task Scheduler to easily adjust task dependencies, greatly improving maintainability and providing clear visibility into task execution. |
Message Queue
SOFAStack Message Queue is a distributed message middleware built on Apache RocketMQ. It provides asynchronous decoupling and peak-load shifting for distributed applications, supporting multiple message types and delivering highly reliable, high-throughput, and highly available asynchronous communication with transactional consistency.
Service architecture
Benefits
Out-of-the-box
Multiple message types: transactional, scheduled, and ordered messages.
Multiple subscription modes: Pub/Sub, tag filtering, clustering consumption, and broadcasting consumption.
Multiple protocols and languages: TCP/HTTP protocols and client software development kits (SDKs) for multiple languages.
Easy O&M
Manageable: Message backtracking, dead-letter queues, and monitoring and alerting.
Easy to diagnose: Multi-dimensional queries and end-to-end message traces.
Highly secure: Multi-tenant resource isolation, instance throttling, and authorization for primary accounts and RAM users.
High performance
Low latency: Millisecond-level end-to-end latency.
Resists accumulation: Can handle massive message accumulation.
High throughput: Can process tens of millions of concurrent requests.
Scalable: Supports elastic scaling based on business needs.
High availability
High availability: 99.9% service availability.
High reliability: 99.99999% data reliability.
Synchronization mechanisms, retry mechanisms, and data persistence.
Scenarios
Asynchronous decoupling |
Decouple upstream and downstream business systems using a producer-consumer model to improve business continuity. The upstream application, acting as the producer, places time-consuming operations that do not require an immediate (synchronous) response into the message queue. The downstream application, acting as the consumer, subscribes to and consumes these messages. As long as the message format remains unchanged, the sender and receiver do not need a direct connection. A failure in one system will not affect the other. |
Data consistency for distributed transactions |
After decoupling applications, you still need to ensure eventual data consistency. Use transactional messages and the reliable delivery mechanism of Message Queue to achieve system decoupling while guaranteeing eventual data consistency. |
Peak-load shifting |
During peak business hours, such as flash sales or group buying events, user requests and traffic can surge, potentially overwhelming some business systems. By introducing a message queue, you can leverage its ability to handle message accumulation. These business systems act as consumers and can process messages according to their own capacity, unaffected by the high traffic. |
API Gateway
API Gateway is an API management platform that helps enterprises uniformly manage internal and external APIs. It provides high-performance, highly secure, and highly reliable communication between network-isolated systems while ensuring internal system security. It meets enterprise needs for opening services to external partners, interconnecting hybrid clouds, and integrating heterogeneous applications within a corporate intranet. API Gateway helps customers innovate across various scenarios and businesses.
Service architecture
Benefits
Cross-cloud and cross-domain interconnection
Supports network interconnection between hybrid clouds and multiple business domains, enabling unified API management.
Weak network optimization
Optimizes data transmission and communication protocols for mobile devices in weak network conditions to improve communication quality and efficiency.
Heterogeneous integration
Supports seamless integration between business systems implemented in different languages and technology stacks.
Support for multiple client types
Supports mobile clients (iOS, Android) and non-mobile clients such as miniapps (H5, Web, backend applications).
End-to-end tracing
Supports end-to-end tracing for requests from the client to all nodes in the backend service.
Robust security
Supports multiple identity authentication and data encryption methods, such as Cross-Origin Resource Sharing (CORS) and SSL, to ensure data transmission security.
Scenarios
Frontend-backend separation scenarios |
|
Heterogeneous system integration scenarios |
|
Hybrid cloud scenarios |
|
Distributed transaction
Distributed Transaction-eXtended (DTX) is a financial-grade distributed transaction middleware that ensures eventual consistency for business activities in large-scale distributed environments. Within Ant Group, it is widely used in core fund-related links such as trading, transfers, and red envelopes, serving hundreds of millions of users. DTX integrates with service frameworks (such as SOFABoot, Spring Cloud, and Dubbo), data sources (such as Data Access Proxy ODP, RDS, MySQL, and OceanBase), and other middleware products like Message Queue to easily implement combinations such as service link-level transactions, cross-database transactions, and message transactions.
Service architecture
Benefits
Support for multiple access patterns
SOFAStack DTX has evolved to support a variety of access patterns to flexibly adapt to diverse FinTech business scenarios.
TCC pattern: Try-Confirm-Cancel (TCC) is a high-performance distributed transaction solution. This pattern offers greater flexibility and can meet the needs of nearly any transaction scenario. The TCC pattern supports custom compensation-based transactions, custom resource reservation-based transactions, and message transactions. You can intervene in the two-phase commit protocol to achieve custom optimizations and special functions for specific scenarios.
FMT pattern: To address the usability issues of the TCC pattern, DTX introduced the Framework-Managed Transactions (FMT) pattern. FMT is a non-intrusive distributed transaction solution that resolves usability challenges. Its main features are ease of use, fast integration, and no intrusion into business code.
Saga pattern: Saga is a compensation protocol. In the Saga pattern, a distributed transaction involves multiple participants, each acting as a reversal and compensation service. You must implement its forward operation and reverse rollback operation based on the business scenario.
Financial-grade quality assurance
Financial-grade disaster recovery: Provides multiple patterns and levels of disaster recovery, including same-city and geo-disaster recovery, to ensure business continuity at the highest industry standards.
Unparalleled performance: Compared to the traditional two-phase commit protocol, it reduces lock holding time and significantly improves performance. The unique Performance Bursting Mode can greatly increase throughput. It supported 544,000 transactions per second during the 2019 Double 11 sales promotions.
Simple and easy to use: Ant Group's years of practical experience have given the product fast and flexible integration capabilities, making it easy to use and maintain.
Compatibility: DTX is an abstract concept based on the service layer and is independent of the underlying transaction implementation. This means that within the scope of a distributed transaction, whether it is a relational database such as MySQL or Oracle, a key-value store such as MemCache, or a column-store database such as HBase, you can integrate it into DTX by wrapping its operations as transaction participants.
Scenarios
Payments and transfers |
Common business scenarios in the finance industry, such as payments, transfers, and accounting, have high throughput requirements. The excellent performance of SOFAStack DTX in various major sales promotions proves that performance will not be a bottleneck. |
Wealth management |
These scenarios often involve large sums of money, so they have very high requirements for product stability. SOFAStack DTX offers financial-grade quality to ensure business continuity and stability. |
Insurance and regulatory reporting |
These types of businesses are typically characterized by many participants and high complexity. SOFAStack DTX has evolved over more than a decade, making it flexible enough to handle various scenarios, meet transactional consistency requirements, and ensure perfect integration with all kinds of businesses. |
Data Access Proxy
Open Database Proxy (ODP) solves data access bottlenecks and database disaster recovery issues under massive request loads. It provides an online distributed database service with horizontal splitting, smooth scaling, and read/write splitting, offering a lightweight, low-consumption, high-performance, and high-availability solution for massive data access.
Service architecture
Benefits
Sharding
Supports sharding for RDS, OceanBase, and MySQL. After creating a distributed database, simply select a shard key. Data Access Proxy then generates sharding rules based on the shard key to achieve horizontal data splitting.
Transparent read/write splitting
Achieves read/write splitting using read-only instances of Data Access Proxy or MySQL standby instances. This helps applications handle details such as transactions, read-only or standby instance failures, and specified primary/standby access. It is non-intrusive to the application, and you can complete read/write splitting operations in the Data Access Proxy console.
Smooth scale-out of data storage
When storage capacity and access volume bottlenecks occur, Data Access Proxy supports online expansion of storage capacity. The scale-out process does not require application modification, and progress can be tracked visually.
Service upgrade or downgrade
Data Access Proxy instances can elastically scale their service capabilities by changing the number of resources.
Globally unique numeric sequence
Data Access Proxy supports distributed, globally unique, and sequentially increasing numeric sequences. This meets business needs for primary keys or unique keys in a distributed database and other specific scenarios.
Database account permission system
Data Access Proxy supports a database account and permission system similar to that of a standalone MySQL instance, ensuring operational security for accounts used by different roles.
Distributed transaction support
Data Access Proxy, combined with the middleware distributed transaction suite, supports distributed transactions to ensure data consistency in a distributed database.
Scenarios
Massive data reads and writes |
Provides a flexible data splitting mechanism with low code intrusion. You can easily implement horizontal data splitting and scaling to solve the problem of single-database bottlenecks. |
Financial-grade data disaster recovery |
Based on years of experience with financial-grade data disaster recovery scenarios within Ant Group, it provides a variety of data center-level data disaster recovery solutions for different business scenarios. This ensures data stability and business continuity. |
Database traffic allocation |
Provides a rule-based traffic distribution mechanism. It meets the needs of high-query businesses by deploying multiple database instances, such as one primary for writes and multiple secondaries for reads. |
Distributed tracing
Distributed System Tracing (DST) is a financial-grade observability solution for distributed, microservice, and cloud-native architectures. It helps users clarify complex invocation relationships between applications and quickly locate faults or slow nodes.
Service architecture
Benefits
Distributed cross-application tracing
Traces the complete event chain for each transaction, establishes a time series of calls across different service (application) interfaces, collects performance data for each service in the chain, and traces transaction performance issues by service.
Multi-dimensional trace query
Retrieves call chains based on a Trace ID and conditions such as errors or timeouts. Analyzes trace information and queries call chain collections for various scenarios.
Multi-perspective trace display
Provides an event chain diagram, trace details, a time series chart, and a timeline to visualize system performance from all angles.
Associated query of business logs
Customize your application system’s business logs. Automatically associate business errors and summary information to quickly locate problems and track business information.
Scenarios
Problem analysis and quick localization |
In distributed scenarios, service invocations are intricate, making problem analysis and localization very difficult. A distributed tracing system can quickly locate problematic services and help resolve issues fast.
|
Optimization of application performance and architecture |
In the call relationship topology, you can view and analyze the number of calls and time consumed by each application. Identify applications with high and low loads to make reasonable use of resources.
|
Operations management
SOFAStack Cloud Application Fabric Engine (CAFE) is a PaaS platform product that provides full lifecycle management, including application management, release and deployment, O&M orchestration, monitoring and analysis, and disaster recovery and emergency response. It meets the O&M needs of both classic and cloud-native architectures in financial scenarios, supports the smooth transition of traditional architectures, and mitigates financial technology risks.
Scenarios
Unified application runtime platform |
Use the platform to solve the challenges of release, monitoring, and auditing in large-scale O&M. Integrate various cloud-native features, such as containers, Serverless, and Mesh, to improve O&M efficiency. |
Platform support for financial-grade high-availability architecture |
Provides PaaS platform support for same-city active-active, cell-based, and active geo-redundancy architectures. |
Upgrading from classic to cloud-native architecture |
Supports the path of upgrading financial infrastructure from containers to cloud-native containerization, reducing the technical risks of transitioning to new architectures and O&M models. |
Cell-based Application Service
Cell-based Application Service (LDC Hybrid Cloud, or LHC) runs on cloud-native infrastructure. In multi-data center, multi-region Kubernetes multi-cluster scenarios, it provides capabilities such as application management, release and O&M, traffic routing, and configuration synchronization. LHC is designed to evolve from a single Kubernetes cluster to an active-active federated cluster. It supports disaster recovery scenarios such as same-city active-active, two-region three-data-center, and more data center-level active-active disaster recovery. It can be combined with other SOFAStack middleware products and the OceanBase distributed database to form a cell-based active geo-redundancy architecture solution.
Service architecture
Benefits
Financial-grade releases
The release process is secure and reliable, with support for retries, canary releases, rollbacks, and traceability.
Supports hybrid releases of virtual machines and containers, providing a transition solution from virtual machines to containers.
Runtime monitoring
Use custom business dashboards to monitor business dynamics at any time.
Monitor basic application metrics in real time, such as PV, Service (invocations of application services), and SAL (calls to external services).
Comprehensively collect basic resource metrics, such as CPU, memory, and I/O traffic.
Microservice framework
Deeply integrates with Ant Group's SOFA Mesh for service registration, discovery, and communication across heterogeneous languages.
Network mode
Supports both VPC and Overlay network patterns.
Supports load balancing types of Service and Ingress.
High availability and disaster recovery
Supports same-city active-active and two-region three-data-center disaster recovery plans.
Supports upgrading to the cell-based high-availability disaster recovery plan developed by Ant Group over many years.
Scenarios
LHC operates in a cloud-native model, using a unified application PaaS platform to provide unified application and resource management along with release and O&M views. It enables multi-cluster management, cross-cluster application O&M and release, resource management, and traffic management.
Same-city active-active |
Establish multiple Kubernetes clusters in two or more zones within the same region. |
Two-region three-data-center |
|
Active geo-redundancy (Multi-region active-active) |
The data layer is sharded. Different zones can be divided into more Logic Data Centers (LDCs) to process different data shards. This design aims to ensure that the data access path from the access layer to the application layer and then to the data layer does not cross zones. With this architecture, you can achieve active-active deployment across any number of regions. |
Hybrid cloud with heterogeneous infrastructure |
Use Kubernetes to abstract away the differences in the underlying Infrastructure as a Service (IaaS). This lets you fully utilize resources on the public cloud by deploying and managing business applications on both Apsara Stack and the public cloud with unified O&M. In this scenario, you can help financial customers achieve the following goals:
|
Classic Application Service
Classic Application Service (CAS) provides a visual and automated way to manage application versions, deployment packages, and resources from an application-centric perspective. CAS offers automated and intelligent DevOps support for the entire application lifecycle, improving efficiency, reducing costs, and minimizing human error so developers can focus on business logic.
Service architecture
Benefits
Application-centric DevOps
Provides automated DevOps support for the entire application lifecycle, shifting the management perspective from IT resource-centric to application- and business-centric. This allows users to focus on business value while improving R&D efficiency and reducing the possibility of human error.
Customizable automated O&M
Provides customizable automated O&M through custom technology stack solutions, enhancing the flexibility of the cloud platform and its compatibility with users' existing systems. This makes it convenient for users to use their familiar, non-native SOFAStack technology frameworks on the platform.
Powerful release and deployment capabilities
Provides a variety of flexible deployment strategies, such as group release, beta release, canary release, single-data center release, and blue-green deployment. It supports visual, automated, retryable, and rollback-capable release and deployment to meet various requirements.
Flexible O&M pipeline capabilities
Provides a channel for entering and executing user-defined O&M commands and scripts, making it convenient for users to perform custom O&M operations.
Scenarios
Support for traditional O&M |
Most core services of traditional enterprises have not yet been containerized and still use traditional code packages for deployment on virtual machines or physical servers. CAS supports a smooth evolution from traditional O&M methods to containerized O&M. |
Decoupling IaaS and PaaS |
In classic O&M scenarios, CAS also supports IaaS from Alibaba Cloud and Huawei Cloud. Integration and adaptation for IaaS from other vendors are ongoing. This allows users to avoid strong dependencies on the underlying platform, truly decoupling IaaS and PaaS. |
CICD integration |
Provides comprehensive application lifecycle APIs for integration with upstream CI platforms, forming a closed-loop CICD process. |
Container Application Service
Application Kubernetes Service (AKS) fully integrates with Kubernetes, providing comprehensive platform capabilities for cluster management, authentication and authorization, container networking, and persistent volume storage. While ensuring standardized and consistent Kubernetes capabilities, it also delivers productized release and deployment capabilities for the full application lifecycle, derived from practical experience.
Service architecture
Benefits
Financial-grade releases
The release process is secure and reliable, with support for retries, canary releases, rollbacks, and traceability.
Supports hybrid releases of virtual machines and containers, providing a transition solution from virtual machines to containers.
Runtime monitoring
Use custom business dashboards to monitor business dynamics at any time.
Monitor basic application metrics in real time, such as PV, Service (invocations of application services), and SAL (calls to external services).
Comprehensively collect basic resource metrics, such as CPU, memory, and I/O traffic.
Microservice framework
Deeply integrates with Ant Group's SOFA Mesh for service registration, discovery, and communication across heterogeneous languages.
Supports native deployment of Istio to provide microservice capabilities with Service Mesh.
Network mode
Supports both VPC and classic network patterns.
High availability and disaster recovery
Supports classic same-city active-active and two-region three-data-center disaster recovery plans.
Supports the cell-based high-availability disaster recovery plan developed by Alibaba Cloud over many years.
Scenarios
Traditional R&D and O&M systems using the SOFA technology stack |
Applications in these systems are developed with SOFABoot or use SOFA Mesh directly. The relationships between systems are complex, with dependencies, and they deeply use SOFAStack products. They need seamless integration with the release and deployment capabilities of the existing PaaS:
Note In this scenario, you need to create an application service in AKS. Use the SOFABoot runtime image as the base image to build the application image. Use an in-place upgrade method for single-level deployment. This allows the application service to run on AKS, interact with services provided by virtual machines, and seamlessly integrate with existing SOFAStack products. |
Lightweight R&D and O&M systems using the SOFA technology stack |
Applications in these systems are developed with SOFABoot or use SOFA Mesh and are tightly integrated with SOFAStack products. The characteristics of these applications are:
Note In this scenario, you only need to create an application service in AKS. Use the SOFABoot runtime image as the base image to build the application image. Use an in-place upgrade method for deployment management. This allows the application service to run on AKS and seamlessly integrate with existing SOFAStack products. |
Systems using a cloud-native technology stack |
These systems typically use traditional Spring or SpringBoot technology stacks, combined with Eureka or ZooKeeper for service registration and discovery. They are self-contained with monitoring and tracing tools from the CNCF ecosystem. The characteristics of these applications are:
Note In this scenario, you only need to create an application service in AKS. Use the SOFABoot runtime image as the base image to build the application image. Use an in-place upgrade method for deployment management. This allows the application service to run on AKS and seamlessly integrate with existing SOFAStack products. |
Real-time Business Monitoring
Real-time Monitoring Service (RMS) is a financial-grade monitoring product with visual monitoring capabilities.
Based on multi-dimensional aggregation of massive data from logs, metrics, and traces, it provides users with visual monitoring functions from multiple perspectives, including business monitoring, application monitoring, cloud-native monitoring, basic resource monitoring, log query and analysis, and distributed tracing. It features rich visual dashboards and provides an alert subscription function.
This service helps O&M, R&D, and Site Reliability Engineers (SREs) quickly discover, locate, analyze, and solve problems, effectively ensuring the availability of online systems.
Refined through Ant Group's scenarios, it provides comprehensive observability and insight analysis capabilities for large-scale and complex business scenarios.
Service architecture
Benefits
Comprehensive real-time monitoring
Provides monitoring capabilities from various perspectives, including business, application, basic resources, and cloud-native. It achieves second-level monitoring for key metrics and minute-level monitoring for regular metrics, featuring high reliability, high timeliness, and low latency.
Flexible alert rules
Set alert rules based on dimensions such as business characteristics, time periods, and importance levels to avoid false positives and false negatives.
Convenient custom configuration
Offers rich custom product configuration functions for convenient and efficient product and alert configuration.
Open technology stack configuration
Achieves deploy-and-monitor for Kubernetes and SOFA technology stack applications. Simply configure the technology stack to access and monitor non-standard business applications.
Visual dashboards
Rich visual dashboards help you customize personalized monitoring dashboards.
Distributed tracing
Provides application topology and trace query functions to observe complex call relationships between applications and services, performance metrics, error information, and associated logs. This enables root cause analysis of faults, service administration, application development and debugging, performance management, performance tuning, architecture control, and fault attribution for DevOps work.
Log query and association
Provides log query and log association functions. Users can not only query logs but also perform historical and contextual queries. You can also view error logs associated with Error metrics and business logs associated with traces, making problem analysis and localization more convenient and efficient.
Low resource consumption
Ensures extremely low consumption of host CPU, memory, and other resources while reliably transmitting large amounts of monitoring data.
High availability
Provides minute-level monitoring deployment capability for tens of thousands of devices, with automatic fault recovery and cluster scalability.
Stable and efficient time series and data storage
Continuously aggregates data online to ensure controllable data capacity. Provides intelligent tiered storage and placement policies.
Scenarios
Multi-dimensional O&M |
Deeply integrates with application services of technology stacks like Kubernetes and SOFA. It provides a one-stop solution for collecting data from infrastructure, middleware, application runtimes, and businesses. Through features like metric monitoring, log analysis, tracing, and alert subscription, it offers multi-dimensional O&M analysis for application performance, running status, and resource usage. This helps to promptly discover and locate problems with applications, resources, and the platform.
|
Problem analysis and quick localization |
In distributed scenarios, service invocations are intricate, making problem analysis and localization very difficult. The distributed tracing system can quickly locate problematic services and help resolve issues at faulty nodes.
|
Application performance optimization |
In the call relationship topology, you can analyze the number of calls and the time consumed for each application. This analysis helps you identify applications with high and low loads to optimize resource utilization.
|
Advanced O&M
The TRaaS Technology Risk Control Platform is a platform product based on the long-term practical methodologies and internal tools of Ant Group's SREs. It solves O&M problems for users during their cloud adoption and distributed transformation processes, including observability, fault emergency response, disaster recovery, chaos engineering, fund security, and stress testing.
High availability management platform
High Availability Service (HAS) is a high-availability management and control platform focused on disaster recovery. It provides end-to-end capabilities for disaster recovery plans, including switchover and recovery for the entire stack from customer business to middleware, PaaS, and IaaS. It also offers capabilities for disaster recovery planning and simulated drills, along with monitoring for the overall status of data centers and disaster recovery, a disaster recovery dashboard, environment inspections, and risk emergency response.
HAS provides a disaster recovery service view, plan orchestration capabilities, and switchover and recovery capabilities. It supports one-click data center-level disaster recovery switchover and recovery in a multi-data center deployment architecture.
Service architecture
Benefits
Complete implementation of Ant Group's technology risk management methodology and platform tools
The High Availability Service platform tool exports the technology risk control content accumulated by Ant Group over many years. On this basis, it helps users build a technology risk control system tailored to their actual situation, improving their overall technology risk control level.
Effective improvement of technology risk control efficiency
Use the technology risk control capabilities of the High Availability Service to automate, standardize, and routinize daily O&M. This reduces operational complexity, makes O&M results clearly visible, and achieves closed-loop management of risk events.
Proactively perceive business operation risks through daily inspections and handle them promptly before they affect the business.
Automated fault diagnosis and standardized emergency plans quickly locate and recover from faults, reducing service downtime caused by failures.
Fault drills proactively test the high-availability capabilities of applications.
Supports dual-center disaster recovery switchover for Ant Group products to meet regulatory compliance requirements.
Rapid updates to the technology risk control content library
The technology risk teams of Alibaba Cloud and Ant Group jointly build a content library for daily inspections, fault diagnosis, and emergency plans based on their experience in controlling technology risks both internally and externally. Users can enjoy the latest technology risk control content.
Financial-grade disaster recovery
The disaster recovery level can reach up to Level 5.
Rich disaster recovery capabilities, including a disaster recovery dashboard for monitoring and alerting, simulated disaster recovery drills, and disaster recovery inspections.
Validated at scale by Alipay and MYbank.
End-to-end disaster recovery
Supports end-to-end disaster recovery capabilities from the customer's application to the central state, achieving overall disaster recovery, end-to-end monitoring, and O&M without the need for multi-platform integration. It includes multi-layer disaster recovery for the entire link, including user applications, middleware, PaaS, and IaaS.
Multi-scenario disaster recovery
Supports all disaster recovery scenarios in the finance industry:
Same-city active-active
Geo-disaster recovery with primary/secondary setup
Two-region three-data-center
LDC cell-based architecture
Scenarios
Daily risk control |
In daily O&M scenarios, use the interaction of multi-functional modules to automatically execute daily O&M scripts, achieving regular and controllable daily inspection O&M. At the same time, continuously update and optimize the content of daily inspections, fault diagnosis, and emergency plans to enrich and improve the application technology risk control system and simplify daily application O&M operations. |
Fault drills |
To continuously improve product high availability, use the fault drill module of the High Availability Service to design and plan drill plans and recovery solutions. Then, continuously discover and solve problems with the disaster recovery plan during the drill process. This reduces the probability of failures during product use, improves fault recovery efficiency, and thus effectively enhances product high availability. |
Data center-level disaster recovery |
|
End-to-end stress testing
Loadcenter provides enterprises with a one-stop stress testing service covering performance stress testing, stress test report output, and stress test risk control. Combined with Ant Group's years of experience in online end-to-end stress testing, it delivers a high-fidelity, low-cost, and high-threat-identification online stress testing experience.
Service architecture
Benefits
Complex scenario modeling capability
Supports multiple traffic models and allows for quick import and configuration of traffic.
Supports templated scripts and custom development scripts to meet the stress testing needs of business scenarios with varying complexity.
Powerful report analysis capability
Archive and centrally manage stress test records.
Combined with the real-time monitoring service, the stress test results record standard application monitoring data and business custom monitoring data. This lets you quickly identify bottlenecked applications and related performance metrics.
Compare and analyze reports to track the evolution of application performance baselines.
Stable pressure output capability
Horizontally scale out pressure machines and dynamically adjust pressure in seconds. This can meet the stress testing needs of tens of millions of TPS.
Support for internal network interface stress testing
The pressure machine resource pool supports both public and private tenant patterns. In public pattern, you do not need to prepare your own pressure machines and can execute stress tests at any time. In private pattern, using your own pressure machines can save bandwidth costs and reduce network latency. It also supports internal interface-level testing without exposing them to the public network, which is more secure.
Support for multiple protocols
Supports standard HTTP/HTTPS and SOFARPC protocols.
Provides a custom development script mode based on the Java language, allowing you to extend proprietary protocols yourself.
Reliable production stress testing risk control capability
Integrates with multiple FinTech products, combining application monitoring, business monitoring, and O&M capabilities. It can automatically stop stress tests when risks are discovered.
Combined with SOFA middleware products, it uses a shadow end-to-end stress testing solution to isolate stress test traffic from normal traffic. This lets you conduct stress tests in a production environment.
Scenarios
End-to-end stress testing is suitable for any application scenario that requires stress testing or simulated traffic.
New system online testing |
Before a new system goes online, conduct stress and load testing on the system based on the expected business model. Test whether the system has performance issues and whether the expected capacity can handle the business pressure after going online. |
Original system baseline regression |
Periodically perform performance regression on the online system in a constant scenario. Observe whether the system's performance has changed and promptly discover system performance attenuation caused by iterations and technology upgrades. |
System capacity assessment |
Before launching operational activities, conduct stress testing to assess the system capacity. Continuously increase the pressure based on the business scenario to evaluate the system's capacity level. This allows for optimization and scale-out in advance. If there are throttling measures, they can also be verified through stress testing. |
System fault drills |
Use continuous stress test traffic to verify whether the business will be affected when the system is abnormal. You can use stress test traffic in conjunction with fault injection drills and data center disaster recovery drills to observe the degree of business impact and recovery capability. |
Fund security monitoring
The fund security monitoring platform is a real-time reconciliation platform that uses a bypass method to provide real-time analysis and alerting on the flow of funds in business processes. It is designed to technically ensure that funds are not lost as they flow through business systems.
Service architecture
Benefits
Non-intrusive to the production system. Collects reconciliation data through a bypass.
Rules are configurable without coding. Rules can be added or modified at any time to meet various reconciliation needs.
Supports multiple reconciliation timeliness options, such as real-time, near real-time, T+1, and T+H, to meet your different needs for monitoring the timeliness of fund loss risks.
Rich management functions, providing a reconciliation dashboard, coverage measurement, and other control capabilities.
Supports notification channels such as text message, email, and DingTalk. It can provide immediate monitoring and emergency support for core businesses.
Provides a measurement function for the coverage of fund loss risk monitoring and is equipped with expert consulting services to bring years of accumulated fund loss prevention experience to cloud users.
Scenarios
Business escort |
Helps you periodically or regularly review core businesses involving fund links. By configuring reconciliation rules, you can cross-check various types of data or perform logical checks on data content to ensure that core businesses are running correctly. |
Change risk inspection |
Before a change is released, add reconciliation rules for the changed business table and associated tables, or data check rules for the changed business table. This ensures that there are no blind spots in fund loss risk monitoring after the change goes online. |
Historical data sweep |
Batch check whether historical data of existing businesses already has discrepancies. Promptly analyze the reasons for the discrepancies, fix vulnerabilities, and recover fund losses. |
Data Quality monitoring |
Missing or incomplete data can also indirectly lead to fund losses. You can configure reconciliation rules to check data integrity, monitor data quality, and promptly discover faults. |
R&D efficiency
Originating from Ant Group's deep background in R&D and engineering practices in the Internet finance field, this provides a financial-grade one-stop intelligent R&D platform for customers in the industry who want to undergo digital transformation. It offers R&D delivery capabilities that balance agile delivery with stable innovation, risk control, and quality assurance. It also endows financial products with high-availability capabilities and promotes the continuous improvement of R&D efficiency.
The one-stop R&D efficiency platform is suitable for the R&D efficiency needs of small and medium-sized enterprises, Internet companies, and financial institutions that require multi-team collaboration, multi-module collaborative development, rapid iteration, and quality improvement.
Service architecture
Benefits
Secure and stable
Based on the Ant Group architecture, the R&D efficiency platform uses a high-availability, dynamically scalable service framework system to build a stable distributed code hosting service. It achieves multi-center, storage-agnostic, and reliable O&M monitoring, along with comprehensive automated quality assurance, ensuring a smooth R&D process.
Intelligent R&D system
The R&D efficiency platform integrates multiple tool platforms. It not only lets you experience the speed of writing code and the stability of code execution but also provides real-time monitoring and intelligent analysis of code. This helps R&D teams conduct a comprehensive and multi-dimensional analysis of code to proactively avoid R&D risks.
Diversified management
The R&D efficiency platform has flexible and elastic configuration capabilities. It can provide diversified R&D processes and quality standards suitable for the different needs of various enterprises based on their unique characteristics.
One-stop service
The R&D efficiency platform provides end-to-end tool services covering the entire R&D lifecycle. The platform's functions are more comprehensive, greatly simplifying the application deployment and release process. It makes it easy to quickly move applications to the cloud and simplifies product development.
Hot-pluggable components
An orchestratable engine easily supports the insertion of customized continuous integration (CI) and continuous delivery (CD) components. Based on configurable rules, it adapts to the context of R&D nodes and automates the Inspection Pipeline of the process.
Data-driven
The data R&D process is systematized, data quality inspection methods are tool-based and diversified, and the data environment is stable and diverse. It monitors and measures service data from the six stages of planning, design, implementation, deployment, maintenance, and unpublishing to achieve more efficient and stable architecture administration.
Quickly build applications
The scaffolding module provides quick creation and download of SOFABoot projects with integrated middleware dependencies. Developers only need to focus on business development, which simplifies the difficulty of R&D, improves R&D efficiency, and helps developers quickly build and manage distributed applications.
Scenarios
The R&D efficiency platform can provide solutions for various types of users in different scenarios. Common application scenarios are as follows:
Small and medium-sized enterprises |
|
Internet companies |
|
Internet finance companies |
|