Amazon ElastiCache is a web service that makes it easy to deploy and run Memcached or Redis protocol-compliant server nodes in the cloud. Amazon ElastiCache improves the performance of web applications by allowing you to retrieve information from a fast, managed, in-memory caching system, instead of relying entirely on slower disk-based databases. The service simplifies and offloads the management, monitoring and operation of in-memory cache environments, enabling your engineering resources to focus on developing applications. Using Amazon ElastiCache, you can not only improve load and response times to user actions and queries, but also reduce the cost associated with scaling web applications.
Amazon ElastiCache automates common administrative tasks required to operate a distributed cache environment. Using Amazon ElastiCache, you can add a caching layer to your application architecture in a matter of minutes via a few clicks of the AWS Management Console. Once a cache cluster is provisioned, Amazon ElastiCache automatically detects and replaces failed cache nodes, providing a resilient system that mitigates the risk of overloaded databases, which slow website and application load times. Through integration with Amazon CloudWatch monitoring, Amazon ElastiCache provides enhanced visibility into key performance metrics associated with your cache nodes. Amazon ElastiCache is protocol-compliant with Memcached and Redis, so code, applications, and popular tools that you use today with your existing Memcached or Redis environments will work seamlessly with the service. As with all Amazon Web Services, there are no up-front investments required, and you pay only for the resources you use.
The in-memory caching provided by Amazon ElastiCache can be used to significantly improve latency and throughput for many read-heavy application workloads (such as social networking, gaming, media sharing and Q&A portals) or compute-intensive workloads (such as a recommendation engine). In-memory caching improves application performance by storing critical pieces of data in memory for low-latency access. Cached information may include the results of I/O-intensive database queries or the results of computationally-intensive calculations.
If you are not already signed up for Amazon ElastiCache, you can click the "Sign Up Now" button on the Amazon ElastiCache detail page and complete the sign-up process. You must have an Amazon Web Services account; if you do not already have one, you will be prompted to create one when you begin the Amazon ElastiCache sign-up process. After you are signed up for ElastiCache, please refer to the Amazon ElastiCache documentation, which includes our Getting Started Guide.
Once you have familiarized yourself with Amazon ElastiCache, you can launch a Cache Cluster within minutes by using the AWS Management Console or Amazon ElastiCache APIs.
Cache Clusters are simple to create, using the AWS Management Console, Amazon ElastiCache APIs, or Command Line Tools. To launch a Cache Cluster using the AWS Management Console, click on the "Launch Cache Cluster" button on the "Amazon ElastiCache" tab. From there, all you need to specify is your Cache Cluster Identifier, Node Type, and Number of Nodes to create a Cache Cluster with the amount of memory you require. Alternatively, you can create your Cache Cluster using the CreateCacheCluster API or elasticache-create-cache-cluster command. If you do not specify an Availability Zone when creating a Cache Cluster, AWS will place it automatically based upon your memory requirements and available capacity.
Amazon ElastiCache supports Cache Nodes of the following memory types:
The following nodes are only available for Memcached
Each Node Type above lists the memory available to Memcached or Redis after taking Amazon ElastiCache System Software overhead into account. The total amount of memory in a Cache Cluster is an integer multiple of the memory available for the Cache Node Type selected. For example, a Cache Cluster consisting of ten Cache Nodes of 6 GB each will provide 60 GB of total memory.
Once your Cache Cluster is available, you can retrieve your Cache Node endpoints using the following steps on the AWS Management Console:
Alternatively, you can use the DescribeCacheClusters API to retrieve the Endpoint list.
You can then configure your Memcached or Redis client with this endpoint list and use your favorite programming language to add or delete data from your ElastiCache Nodes. In order to allow network requests to your Cache Nodes, you will need to authorize access. For a detailed explanation to get started, please refer to our Getting Started Guide.
You can think of the Amazon ElastiCache maintenance window as an opportunity to control when software patching occurs, in the event either are requested or required. If a "maintenance" event is scheduled for a given week, it will be initiated and completed at some point during the 60 minute maintenance window you identify.
Your Cache Nodes could incur some downtime during your maintenance window if software patching is scheduled. Please refer to Cache Engine Version Management for more details. Patching can be user requested - for example cache software upgrade, or determined as required (if we identify any security vulnerabilities in the system or caching software). Software patching occurs infrequently (typically once every few months) and should seldom require more than a fraction of your maintenance window. If you do not specify a preferred weekly maintenance window when creating your Cache Cluster, a 60 minute default value is assigned. If you wish to modify when maintenance is performed on your behalf, you can do so by modifying your DB Instance in the AWS Management Console or by using the ModifyCacheCluster API. Each of your Cache Clusters can have different preferred maintenance windows, if you so choose.
You pay only for what you use and there is no minimum fee. Pricing is per Cache Node-hour consumed for each Node Type. Partial Node-hours consumed are billed as full hours. There is no charge for data transfer between Amazon EC2 and Amazon ElastiCache within the same Availability Zone. While standard Amazon EC2 Regional Data Transfer charges apply when transferring data between an Amazon EC2 instance and an Amazon ElastiCache Node in different Availability Zones of the same Region, you are only charged for the Data Transfer in or out of the Amazon EC2 instance. There is no Amazon ElastiCache Data Transfer charge for traffic in or out of the Amazon ElastiCache Node itself. For more information, please visit the pricing page.
Billing commences for a Cache Node as soon as the Cache Node is available. Billing continues until the Cache Node is terminated, which would occur upon deletion.
Node hours are billed for any time your Cache Nodes are running in an "Available" state. If you no longer wish to be charged for your Cache Node, you must terminate it to avoid being billed for additional Node hours.
Except as otherwise noted, our prices are exclusive of applicable taxes and duties, including VAT and applicable sales tax.
With Reserved Cache Nodes, you can now make a one-time, up-front payment to create a one or three year reservation to run your Cache Node in a specific Region and receive a significant discount off of the ongoing hourly usage charge. There are three Reserved Cache Node types (Light, Medium, and Heavy Utilization Reserved Cache Nodes) that enable you to balance the amount you pay upfront with your effective hourly price.
Functionally, Reserved Cache Nodes and On-Demand Cache Nodes are exactly the same. The only difference is how your Cache Node(s) are billed; with Reserved Cache Nodes, you make a one-time up-front payment and receive a lower ongoing hourly usage rate (compared with On-Demand Cache Nodes) for the duration of the term.
You can use the "Purchase Reserved Cache Nodes" option in the AWS Management Console. Alternatively, you can use the API tools to list the reservations available for purchase with the DescribeReservedCacheNodesOfferings API method and then purchase a cache node reservation by calling the PurchaseReservedCacheNodesOffering method.
Creating a Reserved Cache Node is no different than launching an On-Demand Cache Node. You simply specify the Cache Node class and Region for which you made the reservation. So long as your reservation purchase was successful, Amazon ElastiCache will apply the reduced hourly rate for which you are eligible to the new Cache Node.
Pricing changes associated with a Reserved Cache Node are activated once your request is received while the payment authorization is processed. You can follow the status of your reservation on the AWS Account Activity page or by using the DescribeReservedCacheNodes API. If the one-time payment cannot be successfully authorized by the next billing period, the discounted price will not take effect.
When your reservation term expires, your Reserved Cache Node will revert to the appropriate On-Demand hourly usage rate for your Cache Node class and Region.
When not using VPC, Amazon ElastiCache allows you to control access to your Cache Clusters through Cache Security Groups. A Cache Security Group acts like a firewall, controlling network access to your Cache Cluster. By default, network access is turned off to your Cache Clusters. If you want your applications to access your Cache Cluster, you must explicitly enable access from hosts in specific EC2 security groups. This process is called ingress.
To allow network access to your Cache Cluster, create a Cache Security Group and link the desired EC2 security groups (which in turn specify the EC2 instances allowed) to it. The Cache Security Group can be associated with your Cache Cluster at the time of creation, or using the "Modify" option on the AWS Management Console.
Please note that IP-range based access control is currently not enabled for Cache Clusters. All clients to a Cache Cluster must be within the EC2 network, and authorized via security groups as described above.
When using VPC, please see here for more information.
No. Currently, all clients to an ElastiCache Cluster must be within the Amazon EC2 network, and authorized via security groups as described here.
Yes, EC2 instances in a VPC can access Amazon ElastiCache if the ElastiCache cluster was created within the VPC. Details on how to create an Amazon ElastiCache cluster within a VPC are given here.
Amazon VPC lets you create a virtual networking environment in a private, isolated section of the Amazon Web Services (AWS) cloud, where you can exercise complete control over aspects such as private IP address ranges, subnets, routing tables and network gateways. With Amazon VPC, you can define a virtual network topology and customize the network configuration to closely resemble a traditional IP network that you might operate in your own datacenter.
One of the scenarios where you may want to use Amazon ElastiCache in a VPC is if you want to run a public-facing web application, while still maintaining non-publicly accessible backend servers in a private subnet. You can create a public-facing subnet for your webservers that has access to the Internet, and place your backend infrastructure in a private-facing subnet with no Internet access. Your backend infrastructure could include RDS DB Instances and an Amazon ElastiCache Cluster providing the caching layer. For more information about Amazon VPC, refer to the Amazon Virtual Private Cloud User Guide.
For a walk through example of creating an Amazon ElastiCache Cluster in VPC, refer to the Amazon ElastiCache User Guide.
Following are the pre-requisites necessary to create a Cache Cluster within a VPC:
Creating an Amazon ElastiCache Cluster in an existing VPC is the same as that for a newly created VPC. Please see this for more details.
Amazon ElastiCache Nodes, deployed within a VPC, can be accessed by EC2 Instances deployed in the same VPC. If these EC2 Instances are deployed in a public subnet with associated Elastic IPs, you can access the EC2 Instances via the internet.
Amazon ElastiCache Nodes, deployed within a VPC, can never be accessed from the Internet or from EC2 Instances outside the VPC.
We strongly recommend you use the DNS Name to connect to your ElastiCache Node as the underlying IP address can change (e.g., after a cache node replacement).
A Cache Subnet Group is a collection of subnets that you must designate for your Amazon ElastiCache Cluster in a VPC. A Cache Subnet Group is created using the Amazon ElastiCache Console. Each Cache Subnet Group should have at least one subnet. Amazon ElastiCache uses the Cache Subnet Group to select a subnet. The IP Addresses from the selected subnet are then associated with the Cache Node Endpoints. Furthermore, Amazon ElastiCache creates and associates Elastic Network Interfaces to cache nodes with the previously mentioned IP addresses.
Please note that, we strongly recommend you use the DNS Names to connect to your cache nodes as the underlying IP addresses can change (e.g., after cache node replacement).
An existing Cache Subnet Group can be updated to add more subnets either for existing Availability Zones are for new Availability Zones added since the creation of the ElastiCache Cluster. However, changing the Cache Subnet Group of a deployed Cache Cluster is not currently allowed.
The basic functionality of Amazon ElastiCache remains the same whether VPC is used or not. Amazon ElastiCache manages automatic failure detection, recovery, scaling, auto discovery, and software patching whether your ElastiCache Cluster is inside or outside a VPC.
Similarly, an Amazon ElastiCache Cluster, inside or outside a VPC, is never allowed to be accessed from the Internet. Within a VPC, nodes of an ElastiCache cluster only have a private IP address (within a subnet that you define). Outside of a VPC, the access to the ElastiCache cluster can be controlled using Cache Security Groups as described here.
No, you cannot move an existing Amazon ElastiCache Cluster from outside VPC into a VPC. You will need to create a new Amazon ElastiCache Cluster inside the VPC.
Currently, direct migration of ElastiCache Cluster from inside to outside VPC is not supported. You will need to create a new Amazon ElastiCache Cluster outside VPC.
Amazon ElastiCache allows you to control access to your Cache Cluster and therefore the Cache Nodes using Cache Security Groups in non-VPC deployments. A Cache Security Group acts like a firewall controlling network access to your Cache Node. By default, network access is turned off to your Cache Nodes. If you want your applications to access your Cache Node, you can set your Cache Security Group to allow access from EC2 Instances with specific EC2 Security Group membership or IP ranges. This process is called ingress. Once ingress is configured for a Cache Security Group, the same rules apply to all Cache Nodes associated with that Cache Security Group. Cache Security Groups can be configured with the “Cache Security Groups” section of the Amazon ElastiCache Console or using the Amazon ElastiCache APIs.
In VPC deployments, access to your cache nodes is controlled using the VPC Security Group and the Cache Subnet Group. The VPC Security Group is the VPC equivalent of the Cache Security Group.
You are responsible for modifying routing tables and networking ACLs in your VPC to ensure that your ElastiCache Nodes are reachable from your client instances in the VPC. To learn more see the Amazon ElastiCache Documentation.
No, Cache Security Groups are not used when operating in a VPC. Instead they are used in the non VPC settings. When creating a cache cluster in a VPC you will need to use VPC Security Groups.
No, you can only associate VPC security groups that are part of the same VPC as your Cache Cluster.
Yes, Cache Nodes of an Amazon ElastiCache cluster can span multiple subnets as long as the subnets are part of the same Cache Subnet Group that was associated with the ElastiCache Cluster at creation time. Furthermore, all subnets must be within the same AZ.
A Cache Parameter Group acts as a "container" for engine configuration values that can be applied to one or more Cache Clusters. If you create Cache Cluster without specifying a Cache Parameter Group, a default Cache Parameter Group is used. This default group contains engine defaults and Amazon ElastiCache system defaults optimized for the Cache Cluster you are running. However, if you want your Cache Cluster to run with your custom-specified engine configuration values, you can simply create a new Cache Parameter Group, modify the desired parameters, and modify the Cache Cluster to use the new Cache Parameter Group. Once associated, all Cache Clusters that use a particular Cache Parameter Group get all the parameter updates to that Cache Parameter Group. For more information on configuring Cache Parameter Groups, please refer to the Amazon ElastiCache User Guide.
Amazon ElastiCache by default chooses the optimal configuration parameters for your Cache Cluster taking into account the Node Type's memory/compute resource capacity. However, if you want to change them, you can do so using our configuration management APIs. Please note that changing configuration parameters from recommended values can have unintended effects, ranging from degraded performance to system crashes, and should only be attempted by advanced users who wish to assume these risks. For more information on changing parameters, please refer to the Amazon ElastiCache User Guide.
You can use the AWS Management Console, Amazon ElastiCache APIs, or Command Line Tools to see information about your Cache Parameter Groups and their corresponding parameter settings.
You can cache a variety of objects using the service, from the content in persistent data stores (such as Amazon RDS, SimpleDB, or self-managed databases hosted on EC2) to dynamically generated web pages (with Nginx for example), or transient session data that may not require a persistent backing store. You can also use it to implement high-frequency counters to deploy admission control in high volume web applications.
Yes, Amazon ElastiCache is an ideal front-end for data stores like Amazon SimpleDB and Amazon RDS, providing a high-performance middle tier for applications with extremely high request rates and/or low latency requirements.
Amazon ElastiCache is protocol-compliant with Memcached. Therefore, you can use standard Memcached operations like get, set, incr and decr in exactly the same way as you would in your existing Memcached deployments. Amazon ElastiCache supports both the text and binary protocols. It also supports most of the standard stats results, which can also be viewed as graphs via CloudWatch. As a result, you can switch to using Amazon ElastiCache without recompiling or re-linking your applications - the libraries you use will continue to work. To configure the cache servers your application accesses, all you will need to do is to update your application's Memcached config file to include the endpoints of the servers (Cache Nodes) we provision for you. You can simply use the "Copy Node Endpoints" option on the AWS Management Console or the "DescribeCacheClusters" API to get a list of the endpoints. As with any migration process, we recommend thorough testing of your new Amazon ElastiCache deployment before completing the cut over from your current solution.
Please note that Amazon ElastiCache currently allows access only from the Amazon EC2 network, so in order to use the service, you should have your application servers in Amazon EC2.
Amazon ElastiCache uses DNS entries to allow client applications to locate cache servers (Cache Nodes). The DNS name for a Cache Node remains constant, but the IP address of a Cache Node can change over time, for example, when Cache Nodes are auto replaced after a failure. See this FAQ for recommendations to deal with Cache Node failures.
Though there is no precise answer for this question, with Amazon ElastiCache, you don't need to worry about getting the number of Cache Nodes exactly right, as you can very easily add or remove Nodes later. The following two inter-related aspects could be considered for the choice of your initial configuration:
The amount of memory required is dependent upon the size of your data set and the access patterns of your application. To improve fault tolerance, once you have a rough idea of the total memory required, divide that memory into enough Cache Nodes such that your application can survive the loss of one or two Cache Nodes. For example, if your memory requirement is 14GB, you may want to use two cache.m1.large nodes instead of using one cache.m1.xlarge node. It is important that other systems such as databases will not be overloaded if the cache-hit rate is temporarily reduced during failure recovery of one or more of Cache Nodes. Please refer to the Amazon ElastiCache User Guide for more details.
Not at this time.
You can run a maximum of 20 Cache Nodes. If you need more Cache Nodes, please fill in the ElastiCache Limit Increase Request form.
The service will detect the Cache Node failure and react with the following automatic steps:
You could add more Cache Nodes to your existing Cache Cluster by using the "Add Node" option on "Nodes" tab for your Cache Cluster on the AWS Management Console or calling the ModifyCacheCluster API.
Amazon ElastiCache is ideally suited as a front-end for Amazon Web Services like Amazon RDS and Amazon SimpleDB, providing extremely low latency for high performance applications and offloading some of the request volume while these services provide long lasting data durability. The service can also be used to improve application performance in conjunction with Amazon EC2 and EMR.
Memcached client libraries are available for many, if not all of the popular programming languages. See the list of available clients at the Memcached project page here. If you encounter any issues with specific Memcached clients when using Amazon ElastiCache, please engage us via the Amazon ElastiCache community forum.
Amazon ElastiCache does not require specific client libraries and works with existing Memcached client libraries without recompilation or application re-linking (Memcached 1.4.5 and later); examples include libMemcached (C) and libraries based on it (e.g. PHP, Perl, Python), spyMemcached (Java) and fauna (Ruby).
Auto Discovery is a feature that saves developers time and effort, while reducing complexity of their applications. Auto Discovery enables automatic discovery of cache nodes by clients when they are added to or removed from an Amazon ElastiCache cluster. Until now to handle cluster membership changes, developers must update the list of cache node endpoints manually. Depending on how the client application is architected, typically a client initialization, by shutting down the application and restarting it, is needed resulting in downtime. Through Auto Discovery we are eliminating this complexity. With Auto Discovery, in addition to being backwards protocol-compliant with the Memcached protocol, Amazon ElastiCache provides clients with information on cache cluster membership. A client capable of processing the additional information reconfigures itself, without any initialization, to use the most current nodes of an Amazon ElastiCache cluster.
An Amazon ElastiCache cluster can be created with nodes that are addressable via named endpoints. With Auto Discovery the Amazon ElastiCache cluster is also given a unique Configuration Endpoint which is a DNS Record that is valid for the lifetime of the cluster. This DNS Record contains the DNS Names of the nodes that belong to the cluster. Amazon ElastiCache will ensure that the Configuration Endpoint always points to at least one such “target” node. A query to the target node then returns endpoints for all the nodes of the cluster in question. After this, you can connect to the cluster nodes just as before and use the Memcached protocol commands such as get, set, incr and decr. For more details, see here. To use Auto Discovery, you will need an Auto Discovery capable client. Auto Discovery clients for Java and PHP are available for download from the Amazon ElastiCache console. Upon initialization, the client will automatically determine the current members of the Amazon ElastiCache cluster using the Configuration Endpoint. When you make changes to your cache cluster by adding or removing nodes or if a node is replaced upon failure, the Auto Discovery client automatically determines the changes and you do not need to initialize your clients manually.
To get started, download the Amazon ElastiCache Cluster Client by clicking the “Download ElastiCache Cluster Client” link on the Amazon ElastiCache console. Before you can download, you must have an Amazon ElastiCache account; if you do not already have one, you can sign up from the Amazon ElastiCache detail page. After you download the client, you can begin setting up and activating your Amazon ElastiCache cluster by visiting the Amazon ElastiCache console. More details can be found here.
No, you will not get the Auto Discovery feature with the existing Memcached clients. To use the Auto Discovery feature a client must be able to use a Configuration Endpoint and determine the cluster node endpoints. You may either use the Amazon ElastiCache Cluster Client or extend your existing Memcached client to include the Auto Discovery command set.
To take advantage of Auto Discovery, an Auto Discovery capable client must be used to connect to an Amazon ElastiCache Cluster. Amazon ElastiCache currently supports Auto Discovery capable clients for both Java and PHP. These can be downloaded from the Amazon ElastiCache console. Our customers can create clients for any other language by building upon the popular Memcached clients available.
You can take any Memcached Client Library and add support for Auto Discovery. If you would like to add or modify your own client to enable Auto Discovery, please refer to the Auto Discovery command set documentation.
Yes, we are looking at Ruby next and may add more languages after that.
Yes, Amazon ElastiCache is still Memcached protocol compliant and does not require you to change your clients. However, for taking advantage of auto-discovery feature, we had to enhance the Memcached client capabilities. If you choose to not use the Amazon ElastiCache Cluster Client, you can continue to use your own clients or modify your own client library to understand the auto-discovery command set.
Yes, the same Amazon ElastiCache cluster can be connected through an Auto Discovery capable Client and the traditional Memcached client at the same time. Amazon ElastiCache remains 100% Memcached compliant.
Yes, you can stop using Auto Discovery anytime. You can disable Auto Discovery by specifying the mode of operation during the Amazon ElastiCache Cluster client initialization. Also, since Amazon ElastiCache continues to support Memcached 100% you may use any Memcached protocol-compliant client as before.
Amazon ElastiCache allows you to control if and when the Memcached protocol-compliant software powering your Cache Cluster is upgraded to new versions supported by Amazon ElastiCache. This provides you with the flexibility to maintain compatibility with specific Memcached versions, test new versions with your application before deploying in production, and perform version upgrades on your own terms and timelines. Unless you specify otherwise, your Cache Cluster will automatically be upgraded to new Memcached minor versions as they are supported by Amazon ElastiCache. This patching will occur during your scheduled maintenance window, and will be announced on the Amazon ElastiCache Forum in advance. If you wish to turn off automatic version upgrades, you can do so by selecting the "Auto Minor Version Upgrade" for your Cache Cluster to "No". Since major version upgrades involve some compatibility risk, they will not occur automatically and must be initiated by you. This approach to cache software patching puts you in the driver's seat of version upgrades, but still offloads the work of patch application to Amazon ElastiCache. You can learn more about version management by reading the FAQs that follow. Alternatively, you can refer to the Amazon ElastiCache User Guide. While Cache Engine Version Management functionality is intended to give you as much control as possible over how patching occurs, we may patch your Cache Cluster on your behalf if we determine there is any security vulnerability in the system or cache software.
You can specify any currently supported version (minor and/or major) when creating a new Cache Cluster. If you have opted out of automatically scheduled upgrades by selecting the "Auto Minor Version Upgrade" option to "No" but wish to manually initiate an upgrade to a supported minor version or major version release, you can do so using the "Modify" option for your Cache Cluster. Simply specify the version you wish to upgrade to via the "Cache Engine Version" field. The upgrade will then be applied on your behalf either immediately (if the "Applied Immediately" option is checked) or during the next scheduled maintenance window for your Cache Cluster.
Yes. You can do so by creating a new Cache Cluster with the new Cache Engine Version. You can point your development/staging application to this Cache Cluster, test it and decide whether or not to upgrade your original Cache Cluster.
In the context of Memached, version numbers are organized as follows: Memcached version = X.Y.Z X = Major version, Y = Release level, Z = Version number within release series. From the Amazon ElastiCache standpoint, a version change would be considered major if either major version or release level is being changed. Example: going from 1.4.X -> 1.5.X. A version change would be considered minor if the version number within the release is being changed. Example: going from 1.4.5 -> 1.4.6. As of today, Amazon ElastiCache supports the Memached major version 1.4. We plan to support additional Memcached major versions in the future.
Over time, we plan to support additional Memcached versions for Amazon ElastiCache, both major and minor. The number of new version releases supported in a given year will vary based on the frequency and content of the Memcached version releases and the outcome of a thorough vetting of the release by our engineering team. However, as a general guidance, we aim to support new Memcached versions within 3-5 months of their General Availability release.
In terms of deprecation policy:
Amazon ElastiCache supports the Memcached text and binary protocol as of version 1.4.5 of Memcached.
Amazon ElastiCache for Redis is a web service that makes it easy to deploy and run Redis protocol-compliant server nodes in the cloud. The service enables the management, monitoring and operation of a Redis node; creation, deletion and modification of the node can be carried out through the ElastiCache console, the command line interface or the web service APIs. Amazon ElastiCache for Redis supports Redis Master / Slave replication.
Yes, Amazon ElastiCache for Redis is protocol-compliant with open source Redis. Code, applications, drivers and tools a customer uses today with their existing standalone Redis data store will continue to work with ElastiCache for Redis and no code changes will be required for existing Redis deployments migrating to ElastiCache for Redis unless noted. At launch, we will support Redis 2.6.13.
An ElastiCache for Redis node is the smallest building block of an Amazon ElastiCache for Redis deployment. Each ElastiCache for Redis node supports the Redis protocol and has its own DNS name and port. Multiple types of ElastiCache for Redis nodes are supported, each with varying amount of CPU capability, and associated memory. An ElastiCache for Redis node may take on a primary or a read replica role. A primary node can be replicated to multiple read replica nodes. An ElastiCache for Redis cluster is a collection of one or more ElastiCache for Redis nodes of the same role; the primary node will be in the primary cluster and the read replica node will be in a read replica cluster. At this time a cluster can only have one node. In the future, we will increase this limit. A cluster manages a logical key space, where each node is responsible for a part of the key space. Most of your management operations will be performed at the cluster level. An ElastiCache for Redis replication group encapsulates the primary and read replica clusters for a Redis installation. A replication group will have only one primary cluster and zero or many read replica clusters. All nodes within a replication group (and consequently cluster) will be of the same node type, and have the same parameter and security group settings.
Yes, we support persistence to local storage of the cache node today using Redis AOF append logs. Also, there are two other ways in which you may achieve persistence: 1. You can attach a Redis Slave running on an EC2 instance to an ElastiCache for Redis primary node. The Redis Slave can be used to generate RDB snapshots and / or AOF append logs as needed, and you may transfer these files to S3 for durability. See here for how to set this up. 2. You can call the Redis SYNC command on an ElastiCache for Redis node to retrieve the node’s contents. We will be adding the ability to create RDB snapshots and store them in Amazon S3 soon.
We currently do not support automatically migrating from Memcached to Redis or vice versa. You may, however, use a Memcached client to read from a Memcached cluster and use a Redis client to write to a Redis cluster. Similarly, you may read from a Redis cluster using a Redis client and use a Memcached client to write to a Memcached cluster. Make sure to consider the differences in data format, and cluster configuration between the two engines.
Yes, with Amazon ElastiCache for Redis you can create a read replica in another AWS Availability Zone. Upon a failure of the primary node, we will provision a new primary node. In scenarios where the primary node cannot be provisioned, you can decide which read replica to promote to be the new primary. For more details on how to handle node failures see here.
Amazon ElastiCache for Redis will repair the node by acquiring new service resources, and will then redirect the node's existing DNS name to point to the new service resources. Thus, the DNS name for a Redis node remains constant, but the IP address of a Redis node can change over time. When Redis replication option is selected, and the primary Redis node fails and cannot be restarted, you will be given the option to initiate a failover to a read replica node. The failover target can be in the same zone or another zone. To failback to the original zone, promote the read replica in the original zone to be the primary. If you associate an SNS topic with the ElastiCache for Redis cluster, when the replacement Redis node is configured and ready to be used, an SNS notification is sent to inform you that the node recovery has occurred. You may choose to architect your application to force the Redis client library to reconnect to the repaired Redis server node. This can help as some Redis libraries will stop using a server indefinitely when they encounter communication errors or timeouts.
Amazon ElastiCache monitors the primary node, and in case the node becomes unavailable or unresponsive, the primary node is healed using the approach described here. However, if the primary node cannot be healed, you will have the choice to promote one of the read replicas to be the new primary. See here for how to select a new primary. The DNS record of the primary’s endpoint will be updated to point to the promoted read replica node. A read replica node in the original primary’s AZ will then be created to be a read replica in the replication group and will follow the new primary.
Yes, during a primary node failure, the read replicas continue to service requests. After the primary node is restored, either as a healed node or as a promoted read replica, there is a brief period during which the read replicas will not serve any requests as they sync the cache information from the primary.
You can configure your Redis installation using a cache parameter group, which must be specified for a Redis cluster. All read replica clusters use the parameter group of their primary cluster. A Redis parameter group acts as a “container” for Redis configuration values that can be applied to one or more Redis primary clusters. If you create a Redis primary cluster without specifying a cache parameter group, a default parameter group is used. This default group contains defaults for the node type you plan to run. However, if you want your Redis primary cluster to run with specified configuration values, you can simply create a new cache parameter group, modify the desired parameters, and modify the primary Redis cluster to use the new parameter group.
Yes, Redis appears as an Engine option in the ElastiCache console. You can create a new Redis cache cluster with the Launch Wizard by choosing the Redis engine. You can also modify or delete an existing Redis cluster using the ElastiCache console.
Yes, just as you can create Memcached clusters within a VPC, you can create Redis clusters within a VPC as well. If your account is a VPC by default account, your Redis clusters will be created within the default VPC associated with your account. Using the ElastiCache console, you can specify a different VPC when you create your cluster.
No, Amazon ElastiCache for Redis does not support Redis passwords. This is because of the inherent limitations of passwords stored in a configuration file. Instead of relying on Redis passwords, ElastiCache for Redis clusters are associated with an EC2 security group, and only clients within this security group have access to the Redis server.
Read Replicas serve two purposes in Redis:
When you run a Cache Node with a Read Replica, the “primary” serves both writes and reads. The Read Replica acts as a “standby” which is “promoted” in failover scenarios. After failover, the standby becomes the primary and accepts your cache operations. Read Replicas also make it easy to elastically scale out beyond the capacity constraints of a single Cache Node for read-heavy cache workloads.
There are a variety of scenarios where deploying one or more read replicas for a given primary node may make sense. Common reasons for deploying a read replica include:
You can create a read replica in minutes using a CreateReplicationGroup API or a few clicks of the Amazon ElastiCache Management Console. When creating a replication group, you specify the MasterCacheClusterIdentifier. The MasterCacheClusterIdentifier is the cache cluster Identifier of the “primary” cache cluster from which you wish to replicate. You then create the read replica cluster within the replication group by calling the CreateCacheCluster API specifying the ReplicationGroupIdentifier and the CacheClusterIdentifier of the master cluster. As with a standard cache cluster, you can also specify the Availability Zone. When you initiate the creation of a read replica, Amazon ElastiCache takes a snapshot of your primary cache cluster and begins replication. As a result, you will experience a brief I/O suspension on your primary cache cluster as the snapshot occurs. The I/O suspension typically lasts on the order of one minute.
The read replicas are as easy to delete as they are to create; simply use the Amazon ElastiCache Management Console or call the DeleteCacheCluster API (specifying the CacheClusterIdentifier for the read replica you wish to delete).
You can connect to a read replica just as you would connect to a primary cache node, using the DescribeCacheClusters API or AWS Management Console to retrieve the endpoint(s) for you read replica(s). If you have multiple read replicas, it is up to your application to determine how read traffic will be distributed amongst them.
At this time, Amazon ElastiCache allows you to create up to five (5) read replicas for a given primary cache node.
In the event of a failover, any associated and available read replicas should automatically resume replication once failover has completed (acquiring updates from the newly promoted read replica).
Creating a read replica of another read replica is not supported.
No. This is not supported. Instead, you may attach a Redis “Slave”, running in your EC2 instance, to your ElastiCache for Redis master node and take a RDB snapshot. You may then create an ElastiCache for Redis standalone primary cache node from that snapshot. Refer to the Amazon ElastiCache User Guide for more details.
Updates to a primary cache node will automatically be replicated to any associated read replicas. However, with Redis’s asynchronous replication technology, a read replica can fall behind its primary cache node for a variety of reasons. Typical reasons include:
Read replicas are subject to the strengths and weaknesses of Redis replication. If you are using read replicas, you should be aware of the potential for lag between a read replica and its primary cache node, or “inconsistency”. Click here for guidance on how to find out the “inconsistency” of your read replica.
You can use the standard DescribeCacheClusters API to return a list of all the cache clusters you have deployed (including read replicas), or simply click on the "Cache Node" tab of the Amazon ElastiCache Management Console.
Amazon ElastiCache monitors the replication status of your read replicas and updates the Replication State field to Error if replication stops for any reason. You can review the details of the associated error thrown by the Redis engine by viewing the Replication Error field and take an appropriate action to recover from it. You can learn more about troubleshooting replication issues in the Troubleshooting a Read Replica problem section of the Amazon ElastiCache User Guide. If a replication error is fixed, the Replication State changes to Replicating.
Amazon ElastiCache allows you to gain visibility into how far a read replica has fallen behind its primary through the Amazon CloudWatch metric ("Replica Lag") available via the AWS Management Console or Amazon CloudWatch APIs.
As discussed in the previous questions, “inconsistency” or lag between a read replica and its primary cache node is common with Redis asynchronous replication. If an existing read replica has fallen too far behind to meet your requirements, you can reboot it. Keep in mind that replica lag may naturally grow and shrink over time, depending on your primary cache node’s steady-state usage pattern.
You can easily delete a read replica with a few clicks of the AWS Management Console or by passing its cache cluster identifier to the DeleteCacheCluster API. If you want to delete the read replica in addition to the primary cache node, you must use the DeleteReplicationGroup API or AWS Management Console.
A read replica is billed as a standard cache node and at the same rates. Just like a standard cache node, the rate per “Cache Node hour” for a read replica is determined by the cache node class of the read replica – please see Amazon ElastiCache detail page for up-to-date pricing. You are not charged for the data transfer incurred in replicating data between your primary cache node and read replica. Billing for a read replica begins as soon as the read replica has been successfully created (i.e. when status is listed as “active”). The read replica will continue being billed at standard Amazon ElastiCache cache node hour rates until you issue a command to delete it.
Amazon ElastiCache detects and automatically recovers from the most common failure scenarios such as node failures so that you can resume cache operations as quickly as possible without administrative intervention. Amazon ElastiCache recommends a failover in the event of any of the following:
Initiated failover is supported by Amazon ElastiCache so that you can resume cache operations as quickly as possible. When failing over, Amazon ElastiCache simply flips the DNS record for your cache node to point at the read replica, which is in turn promoted to become the new primary. We encourage you to follow best practices and implement cache node connection retry at the application layer. Start-to-finish, failover typically completes within three to six minutes.
No. Your read replica may only be provisioned in the same or different Availability Zone of the same Region as your cache node primary.
Yes, you can gain visibility into the location of the current primary by using the AWS Management Console or DescribeCacheClusters API.
After failover, my primary is now located in a different Availability Zone than my other AWS resources (e.g. EC2 instances).