MM Ep-46 – The Curious Case of HarperDB, a NoSQL Database!

In the ever-evolving landscape of modern applications, data needs hold supremacy.

The choice of a database system can make or break the success of an application.

Industry used to love SQL databases till data sizes were manageable.

While the industry now leans towards NoSQL databases for their flexibility, lower cost and economic scaling of data.

HarperDB in this context emerges as a disruptor, offering a unique approach to transforming the data needs of modern applications.

In this issue of Monday Muse, let me elaborate the basic considerations behind selecting HarperDB as a NoSQL database which will be relevant for similar NoSQL database choices.

HarperDB = NoSQL with a Cause

NoSQL databases have gained popularity for their ability to handle unstructured data and scale horizontally, meeting the demands of modern, dynamic applications.

However, data engineers argue that the lack of a standardized query language and data consistency issues can create chaos in application development.

HarperDB ticks off the box by providing a SQL-based interface while retaining the scalability advantages.

This approach addresses the concerns of productivity in developers who find comfort in the familiarity of SQL queries. The result is a bridge between tradition SQL and innovation in NOSQL.

Consistency = Data Reliability

One of the primary contentions against NoSQL databases is the eventual consistency model, which sacrifices immediate consistency for availability and partition tolerance.

HarperDB, on the contrary, adopts a strong consistency model, ensuring that data remains reliable and coherent at all times.

HarperDB – like PostgreSQL – have enterprise grade ACID SQL transactions, meaning the validity of data is quite reliable.

In the realm of modern applications, especially those needing real-time interactions, data consistency is non-negotiable.

HarperDB’s commitment to maintaining data integrity aligns with the view that consistency is a fundamental requirement for any robust modern application.

Data as a Service = Fusion of Service with Database

In the age of microservices and APIs, exposing data as a service is paramount for fostering interconnectivity between applications.

HarperDB embraces this trend by offering a RESTful API. It enables developers to effortlessly expose data to other applications while maintaining control over access.

There is also a frequent argument that NoSQL databases often lack the seamless integration required for modern application architectures.

HarperDB’s approach challenges this perception by providing a user-friendly API that facilitates the effortless sharing of data, aligning with the contrarian belief that data should be a fluid and accessible resource.

As I was playing around with HarperDB recently, I thought of adding below some commands to show how easy it is to set up and use HarperDB:

//Download docker image

docker pull harperdb/harperdb

//Run docker image

docker run -d \

-v /home/samit_das/harperdb/hdb \

-e HDB_ADMIN_USERNAME=$HDB_ADMIN \

-e HDB_ADMIN_PASSWORD=$PASS \

-e THREADS=4 \

-p 9925:9925 \

-p 9926:9926 \

harperdb/harperdb

//Generate token

curl –location –request POST ‘http://localhost:9925’ \

–header ‘Content-Type: application/json’ \

–data-raw ‘{

“operation”: “create_authentication_tokens”,

“username”: “username”,

“password”: “password”

}’

//Create a new schema

curl –location –request POST ‘http://localhost:9925’ \

–header ‘Content-Type: application/json’ \

–header ‘Authorization: Bearer eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9…….’ \

–data-raw ‘{

“operation”: “describe_database”,

“database”: “test-data”

}’

//Create a new table

curl –location –request POST ‘http://localhost:9925’ \

–header ‘Content-Type: application/json’ \

–header ‘Authorization: Bearer eyJhbGciOiJSUzI1NiIsInR5cCI6IkpXVCJ9……..’ \

–data-raw ‘{

“operation”: “create_table”,

“database”: “test-data”,

“table”: “names”,

“primary_key”: “id”

}’

API reference: https://docs.harperdb.io/docs/developers/rest

Reporting = Raw Data to Actionable Insights

NoSQL databases are becoming more popular for storing and processing large volumes of unstructured or semi-structured data. However, they pose some challenges for data analysis and reporting.

But our application requirement demands that the data stored should not merely be a repository but a source of actionable insights.

HarperDB facilitates this by providing robust reporting capabilities through its SQL interface.

By enabling developers to harness the power of SQL queries, HarperDB empowers them to extract meaningful information from the data, facilitating informed decision-making.

This approach challenges the notion that NoSQL databases are sufficient for the modern application’s reporting needs.

Data Security

The importance of data security can not be overstated.

HarperDB adopts a proactive stance on security, providing features such as encryption at rest and in transit. I have seen that database queries happen over https as well, MTLS (mutual TLS stands can be enforced). The JWT tokens can be refreshed at specific interval.

HarperDB prioritizes safeguarding of sensitive information, challenging the assumption that NoSQL databases inherently sacrifice security for flexibility.

In Conclusion

In the sea of database options, NoSQL database like HarperDB challenges the widely accepted norms of SQL/NoSQL.

By fusing the familiarity of SQL with the scalability of NoSQL, embracing strong consistency, prioritizing data security, and facilitating seamless data exposure – HarperDB presents a compelling case for those who seek a nuanced, approach to data management.

Hope, this was useful – please have a look at this to learn about the features of a NoSQL database.

That’s all for this week.

Till next week!