Cache management FAQs

• Use redis-benchmark.exe to load test your Redis server. Use redis-benchmark.exe to get a feel for possible throughput before writing your own performance tests.
• Use redis-cli to monitor the cache using the INFO command. For instructions on downloading the Redis tools, see How can I run Redis commands?
• If you use Transport Layer Security/Secure Socket Layer (TLS/SSL) on your cache instance, add the --tls parameter to your Redis tools commands, or use a proxy like stunnel to enable TLS/SSL.
• Redis-benchmark uses port 6379 by default. Use the -p parameter to override this setting if your cache uses the SSL/TLS port 6380 or the Enterprise tier port 10000.
• If necessary, you can enable the non-TLS port through the Azure portal before you run the load test.
• Make sure the client virtual machine (VM) you use for testing is in the same region as your Azure Cache for Redis instance.
• Ensure that your client VM has at least as much computing and bandwidth capability as the cache you're testing. For best results, use D-series and E-series VMs for your clients.
• If you're on Windows, enable Virtual Receive-side Scaling (VRSS) on the client machine. For more information, see Virtual Receive-side Scaling in Windows Server 2012 R2.
• Enable cache diagnostics so you can monitor the health of your cache. You can view the metrics in the Azure portal, and you can also download and review your metrics using the tools of your choice.
• If your load is causing high memory fragmentation, scale up to a larger cache size.

The following examples show how to use redis-benchmark.exe. Run these commands from a VM in the same region as your cache for accurate results.

First, test pipelined SET requests using a 1k payload:

redis-benchmark.exe -h <yourcache>.redis.cache.windows.net -a <yourAccesskey> -t SET -n 1000000 -d 1024 -P 50

After you run the SET test, run pipelined GET requests using a 1k payload:

redis-benchmark.exe -h <yourcache>.redis.cache.windows.net -a <yourAccesskey> -t GET -n 1000000 -d 1024 -P 50

Enabling server garbage collection (GC) can optimize the client and provide better performance and throughput when you use StackExchange.Redis. For more information on server GC and how to enable it, see the following articles:

• The <gcServer> element
• Fundamentals of garbage collection
• Garbage collection and performance

Redis server doesn't natively support Transport Layer Security (TLS), but Azure Cache for Redis does support TLS. If you connect to Azure Cache for Redis with a client like StackExchange.Redis that supports TLS, then use TLS.

If the cache uses TLS, you must enable TLS by using the --tls option for Redis tools like redis-cli. You can also use a utility such as stunnel to securely connect the tools to the TLS port by following the directions in the Announcing ASP.NET Session State Provider for Redis Preview Release blog post.

For instructions on downloading the Redis tools, see How can I run Redis commands?

• Avoid using certain Redis commands that take a long time to complete, unless you fully understand the result of these commands. For example, don't run the KEYS command in production. Depending on the number of keys, it could take a long time to return. Redis is a single-threaded server that processes commands one at a time. If you issue the KEYS command, Redis doesn't process subsequent commands until it finishes processing the KEYS command.

  The redis.io site has time complexity details for each operation it supports. Select each command to see the complexity for each operation.

• What size keys to use depends on the scenario. If your scenario requires larger keys, you can adjust the ConnectionTimeout, then retry values and adjust your retry logic. From a Redis server perspective, smaller key values give better performance.

• These considerations don't mean that you can't store larger values in Redis, but latencies are higher. If you have one set of data that's larger than another, you can use multiple ConnectionMultiplexer instances, each configured with a different set of timeout and retry values. For more information, see What do the StackExchange.Redis configuration options do?

Each Azure Cache for Redis pricing tier has different limits for client connections, memory, and bandwidth. While each size of cache allows up to some number of connections, each connection to Redis involves associated overhead. An example of such overhead is CPU and memory usage because of TLS/SSL encryption.

The maximum connection limit for a given cache size assumes a lightly loaded cache. If load from connection overhead plus load from client operations exceeds capacity for the system, the cache can experience capacity issues even if you don't exceed the connection limit for the current cache size.

For more information about the connection limits for each tier, see Azure Cache for Redis pricing. For more information about connections and other default configurations, see Default Redis server configuration.

• Use Standard or Premium tier for Production systems. The Basic Tier is a single node system with no data replication and no service-level agreement (SLA). Also, use at least a C1 cache for Production. C0 caches are typically used for simple dev/test scenarios.
• Be aware that Redis is an in-memory data store, and data loss can occur in some scenarios. For more information, see Troubleshoot data loss in Azure Cache for Redis.
• Develop your system so it can handle connection blips caused by patching and failover.
• Use Premium-tier Azure Redis instances for better network latency and throughput, because they have better hardware for both CPU and network.

StackExchange.Redis best practices

• Set AbortConnect to false, then let the ConnectionMultiplexer reconnect automatically.
• Use a single, long-lived ConnectionMultiplexer instance rather than creating a new connection for each request.
• Redis works best with smaller values, so consider chopping up bigger data into multiple keys. For example, in What is the ideal value size range for redis?, 100 kb is considered large. For more information, see Consider more keys and smaller values.
• Configure your ThreadPool settings to avoid timeouts.
• Use at least the default connectTimeout of 5 seconds. This interval gives StackExchange.Redis sufficient time to reestablish the connection if there's a network blip.
• Be aware of the performance costs associated with different operations you run. For instance, the KEYS command is an O(n) operation and should be avoided. The redis.io site has details about the time complexity of each operation it supports. Select each command to see the complexity for each operation.

The Common Language Runtime (CLR) ThreadPool has two types of threads, Worker and I/O Completion Port (IOCP).

• WORKER threads are used for things like processing the Task.Run(…), or ThreadPool.QueueUserWorkItem(…) methods. Various components in the CLR also use these threads when work needs to happen on a background thread.
• IOCP threads are used for asynchronous I/O, such as when reading from the network.

The thread pool provides new worker threads or I/O completion threads on demand without any throttling until it reaches the minimum setting for each type of thread. By default, the minimum number of threads is set to the number of processors on a system.

Once the number of existing busy threads hits the minimum threads number, the ThreadPool throttles the rate at which it injects new threads to one thread per 500 milliseconds.

Typically, if your system gets a burst of work that needs an IOCP thread, it processes that work quickly. However, if the burst is more than the configured minimum setting, there's some delay in processing some of the work as the ThreadPool waits for one of two possibilities:

• An existing thread becomes free to process the work.
• No existing thread becomes free for 500 ms so a new thread is created.

Basically, when the number of Busy threads is greater than Min threads, you experience a 500-ms delay before the application processes the network traffic. Also, an existing thread that stays idle for longer than 15 seconds is cleaned up, and this cycle of growth and shrinkage can repeat.

Error messages from StackExchange.Redis build 1.0.450 or later print ThreadPool statistics, as shown in the following example.

System.TimeoutException: Timeout performing GET MyKey, inst: 2, mgr: Inactive,
queue: 6, qu: 0, qs: 6, qc: 0, wr: 0, wq: 0, in: 0, ar: 0,
IOCP: (Busy=6,Free=994,Min=4,Max=1000),
WORKER: (Busy=3,Free=997,Min=4,Max=1000)

The example shows that for the IOCP thread, there are six busy threads and the system is configured to allow four minimum threads. In this case, the client is likely to see two 500-ms delays, because 6 > 4.

It's best to set the minimum configuration value for IOCP and WORKER threads to something larger than the default value. There's no one-size-fits-all guidance on this value, because the right value for one application is likely too high or low for another application. This setting can also affect the performance of other parts of complicated applications. You need to fine-tune this setting to your specific needs. A good starting place is 200 or 300. Then test and tweak as needed.

Configure the minimum threads setting

You can change this setting programmatically by using the ThreadPool.SetMinThreads (...) method.

For example, in NET Framework, you set this value in Global.asax.cs in the Application_Start method:

    private readonly int minThreads = 200;
    void Application_Start(object sender, EventArgs e)
    {
        // Code that runs on application startup
        AreaRegistration.RegisterAllAreas();
        RouteConfig.RegisterRoutes(RouteTable.Routes);
        BundleConfig.RegisterBundles(BundleTable.Bundles);
        ThreadPool.SetMinThreads(minThreads, minThreads);
    }

If you use .NET Core, you set the value in Program.cs just before the call to WebApplication.CreateBuilder():

    const int minThreads = 200
                  
    ThreadPool.SetMinThreads(minThreads, minThreads);
    
    var builder = WebApplication.CreateBuilder(args);
    // rest of application setup

It's also possible to specify the minimum threads setting by using the minIoThreads or minWorkerThreads configuration setting under the <processModel> configuration element in Machine.config. Machine.config is typically located at %SystemRoot%\Microsoft.NET\Framework\<versionNumber>\CONFIG\.

Setting the number of minimum threads in this way isn't recommended because it's a system-wide setting. If you do set minimum threads this way, you must restart the application pool.

Related content

• See other Azure Cache for Redis FAQs.