Catalyst NoSQL

Introduction

Catalyst NoSQL is a fully-managed, powerful database that provides you with a non-relational, non-SQL means of data storage. With NoSQL, you can utilize a distributed database built on a highly scalable, proprietary infrastructure for the storage requirements of all your unstructured, semi-structured, and disparate data in Catalyst. You can create and operate NoSQL databases without the burden of performing any administrative tasks of provisioning, configuring, or scaling the backend setup.

Catalyst offers NoSQL in addition to its conventional relational database, the Data Store in the CloudScale service. Depending on your business requirements and various other factors, such as the possibility of a fixed schema in your database or the need for a read-heavy system, you can determine the right data storage option for your use case. Refer to this comparison section between Data Store and NoSQL for more help.

Catalyst supports the key-value pair based document-type data storage in NoSQL, with provisions to store your data in a Custom JSON format that supports numerous data types. With the ability to partition the data across clusters, NoSQL offers an high-volume storage option, as well as enables multi-level scalability.

If you choose to build an application with Catalyst NoSQL, you can easily create tables in your project, configure them, add data, or query data from the Catalyst console. You can also easily migrate your existing NoSQL databases from third-party sources into Catalyst in a few simple steps.

Catalyst provides server-side SDKs to perform various NoSQL CRUD operations, such as adding or querying data, in the following programming environments:

• Java
• Node.js
• Python

Refer to the linked SDK documentations to learn about the SDK methods available. You can also avail NoSQL APIs in Catalyst for these operations. Refer to the API documentation for details.

Catalyst Data Store vs NoSQL

Catalyst provides both Data Store and Catalyst NoSQL as two different data storage options, where the former is a SQL and relational database and the latter, a non-relational database. You can opt for the database suited to your needs based on various factors such as your business logic, your application’s data structure, or your usage.

Storing unstructured data in a relational database poses problems in schema organization and often leads to data redundancy. You must therefore choose the right storage platform suitable to your use case.

Here’s a handy guide that will help you decide on the best database option for your requirements.

Catalyst Data Store

Choose Catalyst Data Store if the following conditions are met-

• Architecture: Your database’s architecture is relational, i.e., the data points are related to one another.

• Data Structur: Your data is well-structured and can be tightly organized into rows and columns to collectively form a table.

• Schema: The schema of your database is uniform, known in advance, and can be designed and finalized statically ahead of data operations.

• Query Language: You require using SQL to query, update, and maintain the database. SQL goes well with relational, structured databases.

• Priorities: Your priority for your database is ACID compliance (Atomicity, Consistency, Isolation, and Durability), and not horizontal scalability.

• Read/Write Throughput: You require fast, error-free, efficient querying with minimal lag and overhead for a read-heavy database.

Catalyst NoSQL

Choose Catalyst NoSQL if the following conditions are met-

• Architecture: There are no inter-dependencies or relationships between the data points in your database, and they exist as independent entities.

• Data Structure: Your database contains semi-structured or unstructured disparate data that cannot be stored in the conventional tabular format, and is loosely organized.

• Schema: You require the schema of your database to be highly flexible where all the items need not adhere to the same structure, easy to set up, or if it cannot be designed in advance.

• Data Format: Your data will primarily be in the JSON format, and you require the support of multiple data types.

• Priorities: Your priority is high scalability, both vertically and horizontally, and you require storage across multiple database nodes with dynamic resource allocation.

• Read/Write Throughput: You are building a write-heavy system where partitioning storage across distributed clusters with peer-to-peer replication can be beneficial.