Introducing Azure Managed Redis, cost-effective caching for your AI apps

By: Balan Subramanian, Partner Director of Product for Azure App Platform, Microsoft

At Microsoft’s Ignite conference, Microsoft announced Azure Managed Redis for public preview. This new offering features the latest Redis innovations, industry-leading availability¹ and a cost-effective total cost of ownership (TCO) designed for the hyperscale cloud. Azure

 Managed Redis delivers these capabilities on a trusted cloud platform, empowering businesses to scale and optimize their generative AI applications seamlessly.

Try Azure Managed Redis now! 

Latest Redis innovations

The popularity of community Redis among developers continues to rise, but a few key obstacles have become apparent. Its single-threaded architecture can hinder performance, especially at scale. And most Redis offerings on cloud platforms don't offer important new features like the vector search capabilities. Although the Enterprise tier of Azure Cache for Redis solved some of these issues, it still faced billing and tier-based feature limitations, making development and testing challenging.

To provide the best possible experience for developers and customers, we are excited to introduce the next generation of Redis on Azure: Azure Managed Redis. This new offering features four tiers—Memory Optimized, Balanced (memory + compute), Compute Optimized, and Flash Optimized—each running on the latest Redis Enterprise stack. With improvements like multi-core utilization up to 384 vCPUs, vector search for accelerating AI apps across all tiers and SKUs, Redis stack support, and active geo-replication, Azure Managed Redis allows for customers access to the most advanced Redis capabilities, including the latest version, Redis 7.4, for a truly authentic Redis experience.  

AI-ready

Azure Managed Redis brings the latest Redis innovations to support high-performance, scalable AI applications. With features like in-memory data storage, vector similarity search, and real-time processing, it enables developers to handle large datasets efficiently, accelerate machine learning, and build faster AI solutions. Its interoperability with Azure OpenAI Service enables AI workloads to be faster, scalable, and ready for mission-critical use cases, making it an ideal choice for building modern, intelligent applications.

Optimized Performance

With the Redis Enterprise stack, the single-threaded performance bottleneck of Redis can be overcome, unleashing greater performance. To leverage this, we introduced the Compute Optimized tier, which features a higher vCPU-to-memory ratio, allowing for even greater performance. Internal benchmarks show that the Compute Optimized SKU X20 delivers up to 15x higher operations per second on 20GB instances compared to the legacy Azure Cache for Redis Standard SKU C5. Additionally, the creation time of all tiers is up to three times quicker.²

Industry leading 99.999% availability SLA

Redis has evolved to support mission-critical applications beyond its initial design, necessitating high availability assurances that are challenging to achieve with a typical Redis setup. Upon GA, this latest offering can deliver up to an industry-leading 99.999% availability SLA¹. 

Cost-effective TCO

The new offer will cater to a diverse array of sizes and price points , accommodating everything from the smallest demos to the most mission-critical deployments. Customers now have the flexibility to choose their tiers according to their unique needs and use cases, with options prioritizing, memory, vCPUs (performance), or both. That means that, for the first time, you can scale up just memory or just performance based on your needs without needing to pay for memory or vCPUs you don’t need.  

Cache sizes will vary from 500 MB to 4.5 TB with plans to be available across all Azure regions, including sovereign clouds⁴. Unlike previous iterations of Azure Cache for Redis, most features will be accessible at each service tier³, providing developers with greater adaptability. You are also able to run in a non-HA mode for dev/test, lowering the price for use-cases where high availability and resiliency are not required.

Even with these new capabilities, Azure Managed Redis maintains a cost-effective TCO, when compared to our legacy Azure Cache for Redis offering. This gives developers the freedom to utilize Azure Managed Redis throughout their development processes.

Trusted cloud

Azure's reliable cloud infrastructure keeps your data and applications secure. Microsoft Entra ID provides strong identity management with multi-factor authentication and conditional access. This enhances security and user access, allowing only authorized users to reach your resources. With Azure’s global network and Microsoft’s privacy commitment, your data stays protected and accessible.

Getting started with Azure Managed Redis

We are excited to offer the public preview of Azure Managed Redis, built to drive innovation and prepare your applications for AI.

If you are attending Microsoft Ignite, please join us at these sessions and stop by the Azure Managed Redis booth for demos and to speak with an expert.

To get started with Azure Managed Redis today, please check out our product page for more information or contact our sales team.

Resources

• Azure Managed Redis product page
• Azure Managed Redis pricing page 
• Azure Cache for Redis product page

Featured Ignite 2024 sessions

• Azure Managed Redis: Fully Managed Redis for the Hyperscale Cloud
• Azure Managed Redis: Designed to supercharge your GenAI Apps
• Transforming Retail with Azure Managed Redis and GenAI

FAQs

• Q: What is the difference between the new tiers?
  • A: We have three tiers as a part of the new offering: Memory Optimized, Balanced (Memory + Compute), Compute Optimized, and Flash Optimized. Each tier will offer the same features and memory sizes, but different levels of compute performance.
    • Memory Optimized Tier: Ideal for memory-intensive use cases that require a high memory-to-vCPU ratio (8:1) but do not need the highest throughput performance. It provides a lower price point for scenarios where less processing power or throughput is necessary, making it an excellent choice for development and testing environments.
    • Balanced (Memory + Compute) Tier: Offers a balanced memory-to-vCPU (4:1) ratio, making it ideal for standard workloads. Positioned as a middle-ground option, it provides a healthy balance of memory and compute resources.
    • Compute-Optimized Tier: Designed for performance-intensive workloads requiring maximum throughput, with a low memory-to-vCPU (2:1) ratio. It is ideal for business-critical applications that demand the highest performance.
    • Flash-Optimized Tier: Enables Redis clusters to automatically move less frequently accessed data from memory (RAM) to NVMe storage, allowing for the cost-effective scaling of caches with large datasets.
  • These tier options are designed to make it easier for you to optimize your price/performance configuration.

• Q: Why is the performance of the new offering higher?
  • A: There are many reasons! To begin with, Azure Managed Redis is designed to utilize multiple CPU cores, which improves both throughput and latency. Additionally, it operates on the most recent Azure hardware, with optimized CPU, memory, and networking.
    
    
• Q: Is this a marketplace offering?
  • A: No. This is a first party offering that is not available on the Azure marketplace. It can be provisioned just like any other first-party Azure service, and you can apply your current spending commitment and enterprise agreement to this offering as well.
• Q: Will this be offered in sovereign clouds?
  • A: Yes. While not yet available during preview, support for all Azure regions and sovereign clouds is planned.
    
    
• Q: How does support work?
  • A: Support is handled by Microsoft. Additionally, experts from Redis Inc. will be brought in at an early stage to assist during the support process. But support will all be handled through the Microsoft support systems.
    
    
• Q: Is this sold by Microsoft or Redis Inc.?
  • A: This offer is transacted by Microsoft.
    
    
• Q: How is this different than the current Enterprise and Enterprise Flash tiers?
  • A: This new offer is an advancement of the existing Enterprise and Enterprise Flash tiers, operating on a battle-tested but upgraded software stack. The primary changes include: 
    
    • Latest Redis software: Azure Managed Redis delivers the latest Redis versions (v7.4) built for the hyperscale cloud. 
    • Expanded vCPU and memory options: Choose from a much broader range of configurations, from 500 MB to 4.5 TB, with the option to select up to 384 vCPUs to further boost performance and optimize diverse workloads. 
    • Enterprise-grade features: The latest Redis innovations, including vector search, a multimodal data model, active geo-replication, and more, are available across all tiers to accelerate GenAI apps.³ 
    • Cost-savings: Overall prices have been reduced, providing more value and enabling developers to do more with their budgets. 
    • First party service: Transactions now bypass the Azure Marketplace, for a simpler and faster buying experience. 
    • Non-high availability offering: Optimize costs with non-HA setups for dev testing and test cases. 
    • Enhanced security: Support for Microsoft Entra ID offers robust identity management with multi-factor authentication and conditional access, ensuring secure, authorized user access. 
    • Regional availability: Future compatibility across all Azure regions and sovereign clouds allows for greater flexibility in deployment and scalability.⁴
      
      
• Q: How can I migrate to the new offering?
  • A: To get started is to create a new instance and migrate data from your old cache using the export feature. In the future, we plan on offering streamlined tools to make migration seamless.
    
    
• Q: How does high availability work?
  • A: Each instance by default includes built-in zone redundant data replication. If the primary instance fails for any reason, like a power outage or patching event, it will automatically switch to the replica. Unlike earlier versions of Azure Cache for Redis, you can run any instance in non-HA mode, though this is recommended only for demo or dev/test purposes. Availability SLA can reach up to 99.999% when using active geo-replication.¹ Refer to the SLA page for the most recent details.
    
    
• Q: Can I use Azure Managed Redis for vector search on AI applications?
  • A: Yes. All tiers³ of the new offering include Redis Stack support, which includes rich vector storage and search features.
    
    
• Q: Are the new SKUs compatible with community Redis?
  • A: Yes, the new offering leverages the Redis Enterprise stack, which is compatible with community Redis. Since Azure Managed Redis operates as a managed service, certain commands are unsupported, and these exceptions are documented here.
    
    
• Q: What is the price of the new tiers?
  • A: For the full details on pricing, please see the pricing page.

¹Up to 99.999% uptime SLA is planned for the General Availability of Azure Managed Redis

²Source: Azure Managed Redis delivers up to 15x greater performance (operations per second) compared to the existing Azure Cache for Redis offering. This claim is based on internal benchmarking study conducted by Microsoft, demonstrating that users can experience a 15x increase in performance when comparing x20 to C5 SKUs on 20GB instances. Actual performance may vary.

³B0, B1 SKU options, and Flash Optimized tier, may not have access to all features an capabilities 

⁴All public regions are planned for general availability. Public Preview launch will have 21 regions. All regions in sovereign and government clouds are planned for post general availability.