Smart Cache is the fastest, most scalable, distributed caching tier that runs on any operational data source, to accelerate your digital applications and fuel real-time reporting and BI. Smart Cache is optimized for rapidly changing data and multi-criteria queries, and is fully SQL compatible.
Benefits & Features:
- Native Multi-Criteria Queries for Extreme Performance - unlike Key-Value solutions, Smart Cache supports an unlimited number of advanced indexes such as ranges, compound, text and more, for optimal performance, with full SQL compliance.
- Business Driven Caching Policy for Optimal Optimization of Cost and Performance - gain control over your data and keep your most important data in RAM. With advanced business-driven policies, you’re no longer limited to LRU tiering logic.
- AIOps Autonomous Scaling - automatic triggers to scale either up or out for transactional and analytical workloads, over Kubernetes, OpenShift or GigaSpaces ElasticGrid. Scaling is performed in a rolling manner with no downtime or performance impact.
- No Code, One-Click Database Integration - with just one click, the platform scans the database and its metadata, maps its data modeling to the GigaSpaces store and defines its indexing and partitioning schema to eliminate weeks of manual work and human errors. Fast Server-Side Aggregations - Smart Cache performs the entire aggregation activity distributed in-memory on the server side, avoiding data retrieval back to the * client side. Aggregation is performed on relevant data only, regardless of total data size.
- Never Lose Writes - achieve strong data consistency. Unlike eventual consistency cache solutions, Smart Cache ensures complete data integrity, which means data is never lost. At any load, across any environment.
To learn more about GigaSpaces products, visit the website.
- Getting Started
- How to Use this Image
- Running Your First Container
- Connecting to the Client
- Running the Client with the Docker Bridge Network
- Running the Client in Another Docker Container
- Using the Host Network
- Configuring the Public Host
- Running a Production Cluster on Multiple Hosts
- Beyond the Basics
To test the Smart Cache Enterprise Docker image, run the following in your command line to display a help screen with all the available commands:
docker run gigaspaces/smart-cache-enterprise --help
For example, the version command prints version information:
docker run gigaspaces/smart-cache-enterprise version
The Smart Cache Enterprise Docker image utilizes GigaSpaces' command line interface (CLI). To learn more about the command line interface, see the CLI documentation, or use the --help option.
The Smart Cache Enterprise image requires a license key to run, which can be provided using the GS_LICENSE environment variable. You can use the tryme license, which enables you to use the full Smart Cache product for 24 hours (and then terminates the image), or you can get a longer evaluation license.
The simplest and fastest way to start working with Smart Cache Enterprise is to get a single instance up and running on your local machine. After the instance is initiated, you can start exploring the available features and capabilities.
To run a single host on your machine:
docker run --name test -it -e GS_LICENSE=tryme -p 8090:8090 -p 8099:8099 gigaspaces/smart-cache-enterprise
When running the Smart Cache Enterprise Docker image without arguments, a host is automatically started with the following components:
- Platform Manager (mapped to port
8090) Learn more - Web Management Console (mapped to port
8099) Learn more
Note: These ports are mapped to your host, so you can access them.
Docker runs containers in a bridge network by default. You can use any of the options described below to enable a client to connect to the Space.
By default, the client uses the host network interface. You can configure the client to use the Docker bridge network interface (the IP address is usually 172.17.0.x). Use the GS_NIC_ADDRESS environment variable to enable the client to contact and interact with the Space.
NOTE: This only works for clients that reside on the same host as the Space. The Docker bridge network is inaccessible to other hosts.
Docker containers that reside on the same host use the same bridge network. If the client is in a Processing Unit, you can run it via another Docker container with the pu run command.
NOTE: This only works for clients that reside on the same host as the Space. Docker containers on other hosts will use a different bridge network.
Docker can run containers on the host network using the --net=host option with the docker run command. In this case, the client can connect and interact with the Space without additional configuration.
NOTE: Docker only supports the --net=host option on Linux hosts.
By default, the GigaSpaces communication protocol (LRMI) uses the same network interface for both binding and publishing. You can modify this, using the GS_PUBLIC_HOST enviromnent variable to instruct Smart Cache Enterprise to publish itself using a different network address, for example the host's network address. In this case, you'll have to expose the ports listed in the Ports section from the Docker container to the host. For example:
docker run --name test -it -e GS_LICENSE=tryme -e GS_PUBLIC_HOST=<your-host-ip-or-name> -p 4174:4174 -p 8200-8300:8200-8300 gigaspaces/smart-cache-enterprise
When running Smart Cache Enterprise in Docker containers on multiple hosts, you need to either configure GS_PUBLIC_HOST or use the --net=host option as described above, so that containers on different hosts can interact with each other.
The GS_PUBLIC_HOST environment variable complies with common practices of Docker usage, and maintains image isolation. However, as per the Docker documentation, to get optimal performance it is recommended to use the --net=host option, which uses the host network and removes the extra network hop. The Smart Cache Docker image supports both options, so choose the one that best suits your needs.
When running Smart Cache Enterprise on multiple hosts, you can configure the GigaSpaces Manager Server IP address in your network.
GS_MANAGER_SERVERS=host1,host2,host3 by default is the local manager.
The Smart Cache Enterprise Docker image uses the ports described in the table below. You can change each port using the respective environment variable, or map it to a different port using the -p option in docker run. For example, -p 5174:4174 maps the lookup discovery port to a different port, but maintains the same port within the container.
| Environment Variable | Default Value | Description |
|---|---|---|
| GS_MULTICAST_LOOKUP_PORT | 4174 | Lookup discovery port. (learn more) |
| GS_LRMI_PORT | 8200-8300 | Network protocol port range. (learn more) |
| GS_MANAGER_REST_PORT | 8090 | REST Manager API port (learn more) |
| GS_WEBUI_PORT | 8099 | Web Managment Console port (learn more) |
| GS_WEBSTER_HTTP_PORT | 8199 | Internal web service used as part of the application deployment process. |
| GS_RMI_REGISTRY_PORT | 10098-10108 | Used to communicate with the client application. |
| GS_ZOOKEEPER_CLIENT_PORT | 2181 | Used for the Zookeeper client. |
| GS_MANAGER_ZOOKEEPER_DISCOVERY_PORT | 2888 | Used for the Zookeeper discovery ports. |
| GS_MANAGER_ZOOKEEPER_LEADER_ELECTION_PORT | 3888 | Used for the Zookeeper leader election port. |
The Smart Cache Enterprise Docker image utilizes GigaSpaces' command line interface (CLI). Any arguments following the image name are passed to the command line.
If no arguments are specified after the image, the default host run-agent --auto command will be run.
To learn more about the command line interface, refer to the CLI documentation, or use the --help option.
This Smart Cache Enterprise Docker image is based on the official openjdk image, and uses Java version 8. To use a different Java version, you have to build a new image using the JAVA_TAG build argument. For example:
docker build --build-arg JAVA_TAG=9 -t gigaspaces/smart-cache-enterprise:openjdk-9 .
If you're not sure which versions are available, refer to the supported tags page.
You can also build from a different base image, or even create your own, using the JAVA_IMAGE build argument (e.g. --build-arg JAVA_IMAGE=...)
All logs are stored in opt/gigaspaces/logs within the container. To access the logs, you can do one of the following:
- Use the
-voption in yourdocker runcommand to map this to a folder on your host. - Use the
docker cpcommand to copy the files from a Docker instance to a location on your host.
To mount the directory to get the logs, use the following command:
-v c:/gigaspaces/test/logs:/opt/gigaspaces/logs
To copy logs from a running instance, use the following command:
docker cp containerId:/opt/gigaspaces/logs .