solutions

UPM for Enterprise

Core Capabilities

Comprehensive support for mainstream open source data systems, including databases, caching, messaging and searching systems. Optimized and featuring advanced architecture for efficient management, significantly reducing TCO.


Overview

UPM Capabilities Diagram

UPM Capabilities Overview

Below are the core capabilities of UPM for Enterprise:

Unified Workload Management

  • UPM for enterprise provides a comprehensive suite of common Workloads, Units, and UnitSets definitions, supporting various types of data services and data system middleware services
  • Template-based workload definition enables scalable service extensions, allowing customers to customize configurations for specific service types and versions, including sidecar container definitions, automated monitoring agent injection, and security container integration
  • Advanced configuration management through template-based definition and expansion

Scalable Enterprise Operations Management

  • Enterprise grade databases replication management
  • Integrated databases synchronization Management
  • Advanced Redis Replication Management
  • Comprehensive Redis Cluster Management

Unified Operations Workflow and Interface

  • Centralized Resource Management Platform
  • Streamlined Database Cluster Workflow Management
  • Integrated Service Ticket Management System
  • Unified Administrative Interface

Why Flexible Architecure is Crucial for Stateful Services

Kubernetes is leading the next generation of application infrastructure evolution. Compared to traditional virtualization technologies such as VMWare/KVM/XEN, Kubernetes offers more flexible and advanced capabilities in resource elastic scheduling and management. While conserving physical resources, it provides unparalleled efficiency in deployment, operation, and daily maintenance management.

However, fundamental infrastructure systems such as databases, caches, messaging systems, search engines, and even AI inference services are all Stateful Service types, presenting significant technical challenges in Kubernetes and containerized environments. Unlike standard applications (such as Java, Python, or Node.JS based applications), these infrastructure data services essentially handle frequent processing of various states - including business data, state data, log data, cache data, and search index data. Additionally, given the criticality of these data service systems and their security requirements, they need support for flexible deployment, operation, scaling, and maintenance across different cluster topologies, high-availability configurations, and scalable and elastic topologies. This fundamental difference requires any system running on the Kubernetes platform to have capabilities supporting these infrastructure-category data service systems. H Furthermore, from the perspective of individual data systems like PostgreSQL databases or Redis cache systems, the technology vendors and communities behind them only provide simple Kubernetes Operators to address basic deployment and control issues. Unlike traditional physical server or virtualization environments, users building next-generation data center infrastructure through Kubernetes and containers in real data centers need comprehensive, flexible, and centralized management capabilities. An integrated, highly flexible, and unified cloud-native platform is the key to truly solving this challenge.

The UPM for Enterprise platform emerged in response to these technical challenges. Taking a centralized management perspective, it leverages highly abstract design patterns, microservice architecture, deep understanding of mainstream data service systems, and mastery of high availability and service continuity elastic management technology and architecture. Through Kubernetes’ core resource definition abstract modeling, layered abstract management logic design, microservice-based flexible expansion, integrated scheduling and control implementation, modular implementation, and time-tested mature detailed design and implementation for handling data service system complexity, it ultimately provides a feature-rich, secure, reliable, and flexibly scalable next-generation cloud-native data service system platform at the data center level.


Support Matrix

UPM for Enterpirse Supported Deployment Environment :

TypeNameEnvironments
DatacenterVPCPrivate, Public, Hybrid
Host TypeServerPhysical, Virtualized, Containerized
SoftwareStacksSupported Minimal Version
Host OSRHEL
8.6
9.4
RockyLinux
8.6
9.4
Container RuntimeKubernetes containerd
1.7
OpenShift runC
1.7
Container PlatformKubernetes
1.28
Openshift
4.14
4.16

  • All the supported versions will be update along with the latest release.

UPM for Enterprise Supported Data systems :

Data SystemsSupported VersionSupported Topology*HA and Auto Failvoer
RDMS Services
MySQL
8.0.26
8.0.27
8.0.28
8.0.30
8.0.31
8.0.32
8.0.33
8.0.34
8.0.35
Standalone, Semisync Replication, Async ReplicationHA: Yes, Auto_Failover: No
PostgreSQL
15.7
15.8
16.3
Standalone, Sync Replication, Async ReplicationHA: Yes, Auto_Failover: No
PostgreSQL Cluster
16.3
16.4
PostgreSQL Cluster ( Connection pooling included )HA: Yes, Auto_Failover: Yes
Caching Services
Redis
6.2.14
6.2.16
7.0.14
7.0.15
7.2.5
Standalone, ReplicationHA: Yes, Auto_Failover: No
Redis
6.2.14
6.2.16
7.0.14
7.0.15
7.2.5
Replication with RedisSentinel, Redis ClusterHA: Yes, Auto_Failover: Yes
Redis Cluster
6.2.14
6.2.16
7.0.14
7.0.15
7.2.5
Redis ClusterHA: Yes, Auto_Failover: Yes
Messaging Services
Kafka
2.8.2
3.5.2
3.6.2
3 nodes zookeeper cluster with multi nodes KafkaHA: Yes, Auto_Failover: Yes
Zookeeper
3.8.4
3.9.2
3 nodes zookeeper clusterHA: Yes, Auto_Failover: Yes
Searching Services
Elasticsearch
7.17.14
7.17.15
Standalone, ClusterHA: Yes, Auto_Failover: Yes

Note: Please check "High Availability" section for further information. All the supported versions will be update along with the latest release. UPM for Enterprise has powerful extension architecture that support future's new or customized data systems.