We are excited to announce the general availability of the Oracle TimesTen In-Memory Database for Kubernetes – BYOL. TimesTen is an in-memory relational database that can also be used as a cache. It delivers microsecond response and high throughput for OLTP applications. TimesTen has been successfully used across industries for a wide range of use cases.

You may have already reaped the benefits of TimesTen operating behind the scenes. Here are some examples of TimesTen in action today: 

  • TimesTen handles the billing for over a billion mobile phones worldwide as the database for customer balances.
  • Times handles 60% of the database access for one of the biggest multinational e-commerce companies as a cache in front of the Oracle Database.

Oracle TimesTen In-Memory Database (TimesTen)

TimesTen is a lightweight in-memory database that offers unrivaled performance, simple installation and management, high performance, and high availability. TimesTen supports SQL, standard APIs, complete ACID properties, and highly available replication mechanisms. A TimesTen database resides entirely in physical memory and is persistent and recoverable. You have the following options:

  • Oracle TimesTen In-Memory Database is a memory-optimized relational database that provides responsiveness and high throughput for applications. Through transactional replication, TimesTen offers high availability for the in-memory database.
  • Oracle TimesTen In-Memory Database Cache is ideal for caching performance-critical subsets of an Oracle database for improved response time in the application tier. Cache tables can be read-only or updatable. Applications automatically read and update the cache tables using standard SQL and data synchronization between the cache and the Oracle database. Cache offers applications the full generality and functionality of a relational database, the transparent maintenance of cache consistency with the Oracle database, and the high performance of an in-memory database.

Oracle TimesTen In-Memory Database for Kubernetes

TimesTen is well suited to running in containerized environments and, combined with the TimesTen Kubernetes Operator, provides a simple and robust solution for high-performance, highly available data management in Kubernetes environments. The Oracle TimesTen In-Memory Database for Kubernetes offering enables TimesTen deployment on Oracle Cloud Infrastructure (OCI) Container Engine for Kubernetes or on on-premises infrastructure. It can quickly deploy TimesTen in standalone mode or as a cache (for Oracle Database).

TimesTen In-Memory Database for Kubernetes provides the following benefits:

  • Extreme OLTP performance: Achieves extremely low latency application performance by delivering data from an in-memory database that is colocated with applications. 
  • Easy installation and deployment: Includes Helm charts and YAML manifest files that make it easy to install and deploy TimesTen databases.
  • Automatic monitoring and management: Supports automatic lifecycle management of TimesTen databases, including taking auto-corrective actions. Automatic management reduces the need for human-assisted database maintenance, allowing TimesTen production deployments to run more efficiently. 
  • Highly-available data management: Uses active standby replication topology to support high availability.

In addition to the cost of the compute resources used for your Kubernetes cluster, Oracle TimesTen In-Memory Database for Kubernetes supports Bring Your Own License (BYOL), which allows you to reuse your existing on-premises TimesTen In-Memory Database and TimesTen Application Tier Database Cache licenses.

 

Oracle TimesTen In-Memory Database for Kubernetes – BYOL

Oracle TimesTen In-Memory Database for Kubernetes – BYOL is a listing in the Oracle Cloud Infrastructure Marketplace, an online store available in the Oracle Cloud Console (Console). To find the listing:

1. Login to Oracle Cloud via https://cloud.oracle.com.

2. From the navigation menu, select Marketplace.

 

OCI Marketplace

 

3. Under Marketplace, click All Applications.

4. Enter timesten in the search box to find the Oracle TimesTen In-Memory Database for Kubernetes – BYOL listing:

TimesTen MaketPlace Listing BYOL

From this listing, you can export the TimesTen container image from Marketplace. After you have the container image, you can deploy the TimesTen Operator for Kubernetes in the Kubernetes cluster of your choice. We recommend using OKE.

  • A Do It Your Self Kubernetes cluster (DIY K8s) deployed in OCI

  • Oracle Container Engine for Kubernetes (OKE) with managed nodes

See Using Oracle Cloud Marketplace to Obtain a TimesTen Container Image (BYOL) for details about how to:

  • Export a TimesTen container image into a repository in Oracle Cloud Infrastructure Registry (Container Registry).

  • Log in to Container Registry on your development host and create a Kubernetes Secret to store the credentials for your Kubernetes cluster to access the container image from Container Registry.

See Oracle TimesTen In-Memory Database Kubernetes Operator User’s Guide for details about how to deploy the TimesTen Operator and create TimesTen databases using either YAML manifest files or Helm charts.

Oracle Container Engine for Kubernetes – OKE

Oracle TimesTen In-Memory Database for Kubernetes uses Oracle Container Engine for Kubernetes (OKE) for container management and orchestration. Kubernetes is an open-source system for automating containerized applications’ deployment, scaling, and management across a cluster of hosts. A Kubernetes cluster contains a controller node and one or more nodes. The nodes create and manage containers. Kubernetes groups the containers that make up an application into logical units called pods for easy management and discovery.

OKE is an OCI service that enables you to easily create, manage, and deploy applications to Kubernetes clusters. The nodes in a Kubernetes cluster are OCI compute instances.

When you create the node pools, you assign a shape (type of VM) to each compute instance. Those shapes can be:

  • X86
  • Arm
  • Flex
  • GPU

A node pool is a subset of nodes within a cluster with the same configuration, especially the shape. This configuration enables you to optimize their workload, adapt to changes, and reduce costs, and it allows Oracle to provide finer granularity and reduce license costs. 

As an example, let’s look at an E3 VM Flex instance, which is designed with 16 GB per OCPU by default to meet the workload requirements of most of our customers. At this default ratio, the E3 VM is significantly cheaper than comparable instances offered by our competitors, but we wanted to do better. You can optimize your costs further by choosing the exact shape you need to match your workload.

  • Memory Intensive: For applications like in-memory databases or big data processing engines, you can configure an E3 VM instance with a high core-to-memory ratio. With the ability to select up to 64 GB per OCPU, you can create an E3 flexible VM with 16 OCPU and 1024 GB, with savings of 38% over an E3 VM with the default memory-to-core ratio (64 OCPU and 1024 GB).
  • Compute Intensive: For applications like media processing or machine learning, customers can configure the VM to maximize the compute processing power by choosing a low core-to-memory ratio. Customers can configure an E3 flexible VM with 64 OCPU and as little as 64 GB (a 1:1 core-to-memory ratio). This configuration reduces your cost by up to 46% compared to an E3 VM with the default memory-to-core ratio. Note: For ML workloads, customers often use GPU shapes instead.
  • Low Cost: For application workloads like microservices or development environments, customers can build a VM without paying for any OCPU or memory that they don’t need. The smallest E3 flexible VM is now a 1-OCPU, 1-GB VM. This option enables customers to deploy lightweight applications to the cloud in the most cost-effective manner possible. Note: For those workloads, customers often use Arm shapes instead of X86/AMD shapes.

By adopting the Memory-intensive model, you can optimize the cost of computing resources and reduce the number of licenses and costs associated with TimesTen. In this example, 16 OCPUs are used instead of 64 OCPUs for the same 1024 GB of memory.

You can also use other pool nodes for AI and ML Workloads (HPC or GPU) and create another pool for Low-Cost application Workloads inside the same OKE cluster. This allows you to group resources for better management, chargeback, etc. 

In the real world, application workloads are not static. They change as your business grows or according to seasonality. Flexible VMs are ideal for this change. You can modify flexible VMs anytime and as often as you want. You can reconfigure your VM with a few clicks while preserving key instance properties, including private and ephemeral public IP addresses. This functionality lets you scale up for increased performance or scale down to optimize costs. Flexible VMs ensure that the cloud adapts to your application workload rather than vice versa.

 

Oracle TimesTen In-Memory Database in a Nutshell

Oracle embraces the in-memory world, providing extremely low latency and high throughput for your applications. Oracle TimesTen In-Memory Database is the world’s fastest OLTP database. It can be deployed as a standalone or for high availability. You can also offload historical aggregated data that is no longer needed online in memory to an Oracle Database for long-term storage and analysis. Once moved into a disk-based database by an internal mechanism, the aggregated and underlying raw data are purged from the TimesTen database.

The TimesTen cache feature is also ideal for caching a performance-critical subset of an Oracle database in the application tier for improved response time. Applications perform read/write operations on the cache tables using SQL and PL/SQL with automatic persistence, transactional consistency, and data synchronization with the Oracle database.

Oracle TimesTen In-Memory Database is well suited to running in containerized environments and, combined with the TimesTen Kubernetes Operator, provides a simple and robust solution for high-performance, highly available data management in Kubernetes environments.

Those technologies working together allow you to benefit from the best in-memory databases for OLTP and Analytics workloads. TimesTen is well-suited for high-throughput and velocity workloads in real-time data management for telecom and networking, capital markets, defense and intelligence, travel and reservations, call centers, self-service portals, smart metering, gaming, business intelligence, and fraud detection.