1. Best practices framework for Oracle Cloud Infrastructure
  2. Decide Your Storage Solution

Decide Your Storage Solution

Oracle Cloud Infrastructure provides different types of storage system based on the kind of access method (block, file, object), required throughput and I/O, frequency of access (standard, archive), and availability and durability domains.
  • Match your data with most efficient storage solution

    Profile and categorize your data so the correct decision for storage can be made for the best performing and most cost effective solution. This includes understanding actual access requirements, frequency, and format.

  • Collect metrics on Storage, not just compute

    Collecting usage metrics for data driven decisions are just as crucial for data as compute is for providing the best performance and cost optimization. Latency, bandwidth, and even persistence requirements fluctuate and all contribute to deciding on the best state of you data.

  • Apply least privilege principal to data solutions

    In addition to securing your data from malicious actors, limited data access to only resources that need it will ensure that non-malicious actors do not add unnecessary load to your storage solution, increasing performance and reducing cost.

Identify the Application Storage Characteristics and Requirements

Cloud Architect, Application Architect

Identify the storage performance metrics and different characteristics (for example, the file size, access patterns, latency, throughput, and persistence of data) that matter for your workload, and implement a data-driven approach, using benchmarking or load testing.
Use this data to identify where your storage solution is constrained and examine configuration options that could address this.

Evaluate Available Storage Options

Cloud Architect, Infrastructure Architect

Evaluate the type of storage and access methods required. If your application requires block volume access, you'll typically evaluate throughput, IOPS, latency, data growth, cost and durability and look at either network attached Block Volumes or locally attached NVMe Disks.
Both have benefits and drawbacks and you will need to evaluate the benefits and drawbacks of each based on your application requirements.

Object Storage is great for storing very large amounts of data in a durable way or storing data for long term archival at a lower cost then Block volumes. However it requires that the application is capable of interacting with the Object Storage interfaces.

Understand the different storage options and configurations available. The available storage options are described below.

  • Oracle Cloud Infrastructure Local NVMe disks

    Some instance shapes in Oracle Cloud Infrastructure include locally attached NVMe devices. These devices provide extremely low latency, high performance block storage. These devices are not protected in any way by Oracle Cloud Infrastructure; they are individual devices locally installed on your instance. It is your responsibility to protect and manage the durability the data on these devices.

  • Oracle Cloud Infrastructure Block Volumes
    This service lets you dynamically provision and manage block storage volumes. All volumes have built-in durability and run on redundant hardware within a single AD. It provides integrated features to back up your data to Object Storage. The backups can be used for business continuity and disaster recovery.

    Note:

    Block volumes are network attached and their network bandwidth usage counts towards the overall provisioned bandwidth limit of the selected compute instance type.
  • Oracle Cloud Infrastructure File Storage

    This service is a shared file system that provides a durable, scalable, secure, enterprise-grade network file system. Data is replicated for durability within each availability domain.

  • Oracle Cloud Infrastructure Object Storage

    This service provides a highly durable and available (across multiple ADs in a multi-AD region and across multiple FDs in a one-AD region), internet scale and high performance storage for your unstructured data. There are two tiers available based upon the frequency of accessing data (Standard and archive data tiers).

Change or Adjust Your Storage Solution If Your Workload Changes

Cloud Architect, Infrastructure Architect

Over time your workload demand might change, impacting on your solution's performance efficiency or cost optimization . Use metrics to evaluate if adjusting storage solution settings might positively impact your workload. Often smaller actions like changing Block Volume performance characteristic can have a big impact when done at scale.

Collect Metrics and Make Storage Decisions Based On Them

Cloud Architect, Infrastructure Architect, DevOps Architect

You should apply a data-driven approach and use existing metrics or run benchmarks against existing workloads to gather valuable data used when deciding on the cloud based storage approach and also identify potential scaling(performance and data growth) bottlenecks.

Learn More