More Cache for Less Cash (DevLink 2014)
Download as PPTX, PDF•3 likes•8,078 views
This document summarizes different caching options available in Azure, including in-role caching for cloud services, the Azure Cache Service, and the preview Azure Redis Cache. It provides an overview of each option's features, pricing, and how to configure and monitor caches. Demos are shown of creating and using in-role caching and the Azure Cache Service. The document recommends using a mix of caching solutions based on needs and discusses common caching architectures in Azure applications.
More Cache for Less Cash (DevLink 2014)
- 1. More Cache for Less Cash Michael S. Collier @MichaelCollier DevLink 2014 – August 27, 2014
- 2. Today’s Agenda • Why Cache? • Cache Options in Azure • Demos • Monitoring and Scaling • Pricing and Features 3
- 3. Why Cache? 4 Azure Load Balancer Web Tier Business Tier Data Tier
- 4. Why Cache? 5 Azure Load Balancer Web Tier Business Tier Cache Cache Tier Data Tier
- 5. What is Azure Cache Service? A distributed, in-memory, flexible cache for all data types that can be used to speed up Azure applications and reduce database load. Basically, cache helps your app 6 become faster. Windows Azure Cache
- 6. A Cache Story Shared Cache • Cloud Services • Shared – quotas • Lacked feature parity • Multiple size options • Expensive • Performance challenged • Throttling • Deprecated August 2014 In-Role Cache • Cloud Services • Co-located (free) • Dedicated ($$) • Feature parity • Good performance • Not multi-tenant Cache Service • Cloud Services, Web Sites, or VMs • Feature parity • Managed infrastructure • Price/capacity tiers • Good performance (1ms read) 7
- 7. A Cache Story Shared Cache • Cloud Services • Shared – quotas • Lacked feature parity • Multiple size options • Expensive • Performance challenged • Throttling • Deprecated August 2014 In-Role Cache • Cloud Services • Co-located (free) • Dedicated ($$) • Feature parity • Good performance • Not multi-tenant Cache Service • Cloud Services, Web Sites, or VMs • Feature parity • Managed infrastructure • Price/capacity tiers • Good performance (1ms read) 8
- 8. A Cache Story Shared Cache • Cloud Services • Shared – quotas • Lacked feature parity • Multiple size options • Expensive • Performance challenged • Throttling • Deprecated August 2014 In-Role Cache • Cloud Services • Co-located (free) • Dedicated ($$) • Feature parity • Good performance • Not multi-tenant Cache Service • Cloud Services, Web Sites, or VMs • Feature parity • Managed infrastructure • Price/capacity tiers • Good performance (1ms read) 9
- 9. A Cache Story Shared Cache • Cloud Services • Shared – quotas • Lacked feature parity • Multiple size options • Expensive • Performance challenged • Throttling • Deprecated August 2014 In-Role Cache • Cloud Services • Co-located (free) • Dedicated ($$) • Feature parity • Good performance • Not multi-tenant Cache Service • Cloud Services, Web Sites, or VMs • Feature parity • Managed infrastructure • Price/capacity tiers • Good performance (1ms read) Redis Cache • TBD • TBD • TBD • TBD http://aka.ms/MigrateFromSharedCaching 10
- 10. In-Role Cache (Co-located) 11 Source: Windows Azure Training Kit
- 11. In-Role Cache (Dedicated) 12 Source: Windows Azure Training Kit
- 12. In-Role Cache (Dedicated) 13 Source: Windows Azure Training Kit
- 13. In-Role Cache • Based on the AppFabric Cache engine • Released w/ Oct. 2012 release of Azure SDK 1.8 • Cache part of your Cloud Service application • No quotas or throttling • Isolation, flexibility, and control • High performance • Features: named cache, regions, tagging, HA, local, notifications • Pro: You manage as part of your application • Con: You manage it; only available to Cloud Service roles in same deployment 14
- 14. Create and Configure In-Role Cache Configure a Role in the Cloud Service to host the cache 15
- 15. Configure the Cache Clients • Use NuGet to configure the cache clients • Microsoft.WindowsAzure.Caching • Settings placed in web.config or app.config • Modify to point to the cache server role <dataCacheClients> <dataCacheClient name="default"> 16 <!--To use the in-role flavor of Windows Azure Cache, set identifier to be the cache cluster role name --> <!--To use the Windows Azure Cache Service, set identifier to be the endpoint of the cache cluster --> <autoDiscover isEnabled="true" identifier="MyScores.Web" /> <!--<localCache isEnabled="true" sync="TimeoutBased" objectCount="100000" ttlValue="300" />--> </dataCacheClient> </dataCacheClients> Role Name
- 16. 17 DEMO TIME!!!
- 17. Recap 1. Select “Enable Caching” in Azure role configuration 2. Install NuGet package (Microsoft.WindowsAzure.Caching) 3. Set role name in web.config 4. Get/Set data 18
- 18. Monitoring the Cache Diagnostic Level Data Collected 0 Critical/catastrophic server logs only 1 Data to help in assessing usage patterns, cache health, and potential errors. Default. 2 Fine grain data for all requests and important system information 3 Diagnostic data with more verboseness and system information 4 Highest verbosity logs for all requests and system information Provides a single setting for cache servers and clients Configures levels for logs, traces, performance counters & crash dumps Full details at http://aka.ms/CacheDiagnostics 19
- 19. Monitoring the Cache • Set cache diagnostic level in configuration or portal • Level (1-4) controls verbosity, perf counters, & crash dumps. • Start at 1 and increase as needed <Role name="WorkerRole1"> 20 <Instances count="1" /> <ConfigurationSettings> <!– Cache Server --> <Setting name="Microsoft.WindowsAzure.Plugins.Caching.DiagnosticLevel“ value="1" /> <!– Cache Client--> <Setting name="Microsoft.WindowsAzure.Plugins.Caching.ClientDiagnosticLevel" value="1" /> </ConfigurationSettings> </Role> Full details at http://mcollier.net/AzureCacheDiag
- 20. Monitoring the Cache public class WebRole : RoleEntryPoint { public override bool OnStart() { 21 // Enable cache diagnostics DiagnosticMonitorConfiguration dmConfig = DiagnosticMonitor.GetDefaultInitialConfiguration(); // Configure collection of cache diagnostic data. CacheDiagnostics.ConfigureDiagnostics(dmConfig); DiagnosticMonitor.Start("Microsoft.WindowsAzure.Plugins.Diagnostics.ConnectionString", dmConfig); return base.OnStart(); } }
- 21. Considerations • Unable to create DataCache in RoleEntryPoint • Max object size is 8MB post serialized • Default serializer is NetDataContractSerializer • Cost only for role instances • Only accessible from within same Cloud Service • Deployment will impact the cache (instances recycle) 22
- 22. What’s the Azure Managed Cache Service? 23 App1 VM VM App2 Shared Azure Website Stamp VM IaaS VMs Cloud Services Azure Managed Cache Service 1. SLA backed 2. Microsoft Managed 3. Scalable & Resilient
- 23. Create and Configure the Cache Service • Based on AppFabric Cache engine • Create via PowerShell only • Manage in Azure management portal • Choose an offering • Basic (128MB to 1GB in 128MB units, 1 named cache) • Standard (1GB to 10GB in 1GB units, notifications, and 10 named caches) • Premium (5GB to 150GB in 5GB units, notifications, HA, and 10 named caches) • Locate in same region as client for best perf and cost 24
- 24. Create and Configure the Cache Service PS C:> New-AzureManagedCache -Name devlinkcache ` 25 -Location "East US" ` -Sku Basic ` -Memory 128MB
- 25. 26 DEMO TIME!!!
- 26. Recap 1. Create Managed Cache via PowerShell 2. Install NuGet package (Microsoft.WindowsAzure.Caching) 3. Set cache service endpoint & key in web.config 4. Get/Set data 27
- 27. Monitoring the Cache 28
- 28. Scaling the Cache 29 Dynamically scale without loosing any existing data
- 29. High Availability 30 Client Primary Secondary Read / Write Write Premium only Doubles required memory Increased latency & decreased throughput
- 30. Pricing Details 31 Basic Standard Premium Price Per Unit (Preview) $25/month (prorated hourly) $100/month (prorated hourly) $400/month (prorated hourly) Cache Size 128 MB 1 GB 5 GB Scale Up to 8 units Up to 10 units Up to 30 units Named Caches 1 10 10 High Availability Notifications See http://mcollier.net/AzureCachePricing
- 31. Benefits of the Azure Managed Cache Service • Use from any app types (VM, Web Site, Mobile Service, Cloud Service) • Each instance deployed within dedicated VMs (fast, predictable performance) • No quotas or throttling • Store up to 150GB per cache instance • Avg. retrieval time of 1ms. Insert in about 1.2ms • Highly available / distributed across multiple servers • Managed service – focus on apps not infrastructure • Same .NET cache API used with in-role cache for Cloud Services • Support for ASP.NET Session State and Page Output Caching • Unique cache service instance for each app, or share across apps 32
- 32. Azure Redis Cache (Preview) • Secure, dedicated Redis cache • Hosting and monitoring by Microsoft • 250MB – 26GB (Preview) • Tiers • Basic – single node; dev/test and non-critical workloads • Standard – provides master/slave replication 33
- 33. Azure Redis Cache (Preview) 34
- 34. Monitoring and Alerts • Set metrics and timeframe to monitor • Pin to dashboard • Alert when threshold reached 35
- 35. 36 DEMO TIME!!!
- 36. Recap 1. Create Redis Cache via portal 2. Add Redis C# client (e.g. StackExchange.Redis) 3. Obtain Redis endpoint and password from portal 4. Determine formatter for .NET objects 5. Get/Set data 37
- 37. Common Architecture 38 Web Roles Storage Table 3. Write 1. Read Windows Azure Cache 2. Read
- 38. Common Architecture 39 Windows Azure Load Balancer Web Roles Worker Roles SQL Database (Azure) Read Azure Cache Read Preload the cache Carefully choose expiration and eviction settings
- 39. Common Architecture 40 Azure Load Balancer Web Roles Storage Queue Worker Roles Storage Table SQL Database (Azure) Azure Cache Write Read Write Read Write
- 40. Cloud Service Fundamentals - Caching • Factory implementation • Custom BinarySerializer class • protobuf-net • Fast and small http://mcollier.net/AzureCSF 41
- 41. What Cache? • Azure Shared Cache – out of service on September 1, 2014 42 New Development Need GA and SLA
- 42. Where to Cache? 43 Mix cache solutions for price / manageability / features Azure Load Balancer Web Roles Azure Cache Read • Data shared by other services/apps • General application data In-role cache • Session / Page Output Cache • FREE!
- 43. Summary • In-Role Cache • Cloud Services only • Co-located (free) or Dedicated ($$$) • Isolated cache-related workload • Cache Service • Microsoft managed service, dedicated tenant, SLA backed • Cloud Services, Web Sites, and VMs • Full parity with in-role cache (SDK) • Redis Cache • Microsoft managed Redis Cache • Master/Slave replication in Standard tier • Normal Redis features/functionality 44
- 44. More Resources • Azure Cache Options Overview • http://msdn.microsoft.com/en-us/library/azure/gg278356.aspx • Which Azure Cache Offering is right for me? • http://msdn.microsoft.com/en-us/library/azure/dn766201.aspx • Cloud Service Fundamentals – Caching Basics • http://azure.microsoft.com/blog/2013/10/03/cloud-service-fundamentals-caching-basics/ • Capacity Planning Spreadsheet (In-Role Cache) • http://msdn.microsoft.com/en-us/library/hh914129 • Lap around Azure Redis Cache • http://azure.microsoft.com/blog/2014/06/04/lap-around-azure-redis-cache-preview/ 45
Editor's Notes
- #5: SQL DB can often be a limiting factor in WA applications. Can get slow and throttle as pressure increases
- #8: Similar feature parity with Windows Server AppFabric Cache (on-premises)
- #9: Similar feature parity with Windows Server AppFabric Cache (on-premises)
- #10: Similar feature parity with Windows Server AppFabric Cache (on-premises)
- #11: Similar feature parity with Windows Server AppFabric Cache (on-premises)
- #15: A region is a subgroup for cached items. Regions also support the annotation of cached items with additional descriptive strings called tags. Regions support the ability to perform search operations on any tagged items in that region.
- #18: Add In-Role Cache to an existing Cloud Service Show adding package via NuGet Show pre-deployed version in management portal. Show metrics. Install-package Microsoft.WindowsAzure.Caching
- #20: http://msdn.microsoft.com/en-us/library/windowsazure/hh914135.aspx
- #21: Server – added by Visual Studio Client – added by NuGet package http://msdn.microsoft.com/en-us/library/windowsazure/jj136940.aspx http://msdn.microsoft.com/en-us/library/windowsazure/hh914135.aspx
- #23: Unable to create DataCache in RoleEntryPoint “Server collection cannot be empty” WA Cache settings not automatically available in the web.cong in context of RoleEntryPoint Programmatic only http://msdn.microsoft.com/en-us/library/windowsazure/jj852128.aspx
- #24: Scale Units for Cloud Services
- #27: Change site to use new cache service In cloud – have one Web Site and one Cloud Service using the same cache
- #30: Scale cache units w/o losing data. Changing offerings will cause cache data loss.
- #33: Ability to use the Cache Service from any app type (VM, Web Site, Mobile Service, Cloud Service) Each Cache Service instance is deployed within dedicated VMs that are separated/isolated from other customers – which means you get fast, predictable performance. There are no quotas or throttling behaviors with the Cache Service – you can access your dedicated Cache Service instances as much or as hard as you want. Each Cache Service instance you create can store (as of today’s preview) up to 150GB of in-memory data objects or content. You can dynamically increase or shrink the memory used by a Cache Service instance without having to redeploy your apps. Web Sites, VMs and Cloud Service can retrieve objects from the Cache Service on average in about 1ms end-to-end (including the network round-trip to the cache service and back). Items can be inserted into the cache in about ~1.2ms end-to-end (meaning the Web Site/VM/Cloud Service can persist the object in the remote Cache Service and gets the ACK back in 1.2ms end-to-end). Each Cache Service instance is run as a highly available service that is distributed across multiple servers. This means that your Cache Service will remain up and available even if a server on which it is running crashes or if one of the VM instances needs to be upgraded for patching. The VMs that the cache service instances run within are managed as a service by Windows Azure – which means we handle patching and service lifetime of the instances. This allows you to focus on building great apps without having to worry about managing infrastructure details. The new Cache Service supports the same .NET Cache API that we use today with the in-role cache option that we support with Cloud Services. So code you’ve already written against that is compatible with the new managed Cache Service. The new Cache Service comes with built-in provider support for ASP.NET Session State and ASP.NET Output Page Caching. This enables you to easily scale-out your ASP.NET applications across multiple web servers and still share session state and/or cached page output regardless of which customer hit which server. The new Cache Service supports the ability to either use a separate Cache Service instance for each of your apps, or instead share a single Cache Service instance across multiple apps at once (which enables easy data sharing as well as app partitioning). This can be very useful for scenarios where you want to partition your app up across several deployment units.
- #39: Basic – cache-aside pattern
- #40: Pre-load the cache
- #41: Architecture diagram – read from cache; other process updates cache data Reference - http://blog.cynapta.com/2013/11/building-cynapta-services-part-ii-architecture-drilldown/
- #42: http://blogs.msdn.com/b/windowsazure/archive/2013/10/03/cloud-service-fundamentals-caching-basics.aspx
- #43: 1 Medium Cloud Service instance (3.5 GB RAM): $119/month 2 GB Standard cache (2, 1GB units): $100/month 3 A7 dedicated instances (56 GB RAM each): $3,147/month 150GB Premium cache (30, 5GB units): $6,000/month
- #46: Cache (Preview) Features - http://msdn.microsoft.com/en-us/library/windowsazure/dn386143.aspx
- #52: 1 Medium Cloud Service instance (3.5 GB RAM): $119/month 2 GB Standard cache (2, 1GB units): $100/month 3 A7 dedicated instances (56 GB RAM each): $3,147/month 150GB Premium cache (30, 5GB units): $6,000/month