Redpanda Cloud - Annual Commits (BYOC & Dedicated/Serverless)

Redpanda  serves two primary purposes for our organization. First, we use it as a drop-in replacement for Kafka. Second, we utilize it for streaming analytics.

We do not use Redpanda  for IoT data streaming, though it has been quoted as suitable for that use case. IoT data streaming is actually a superset of our use cases. Recently, we have started using it for AI analytics as well.

The best features of Redpanda include its lightweight nature and low resource consumption. Because we migrated from Kafka to Redpanda, this is the main selling point for our organization.

Aside from its lightweight design, Redpanda is essentially a clone of Kafka with all the good features of Kafka. The only difference is in resource requirements. Kafka needs too many resources while Redpanda is a very good, lightweight, and very fast database.

We use Kafka API compatibility with Redpanda, which helps significantly with data migration. There is no need to migrate data because it works out of the box. As a streaming solution, we never migrate the data. We simply have a clean state and replace Kafka with Redpanda. In my case, there was no need to migrate.

Redpanda’s low latency data processing is excellent. We do not experience too much latency, and it works very well for our product architecture. It is exceeding our estimations. While designing the product, we design it for a certain throughput, and Redpanda always works at that throughput.

The metrics I use to evaluate effectiveness in any real-time applications are throughput. Every application has its own throughput requirements, and so far, Redpanda is good at achieving the targeted throughput for our applications.

In Redpanda, the areas that have room for improvement are in the clustering part. Setting up clustering initially is very easy. However, if you are removing a node and attaching another node, significant improvement is required. It is not as expected, and we have had a tough time adding and removing nodes.

Other solutions lack clustering support, but Redpanda has good clustering support, though it needs further improvements for smoother cluster operations.

Aside from clustering, Redpanda is good, but it is worth noting that it is built on top of a special architecture. It works directly at the Linux kernel level. Because of that, it needs better modern hardware with a better CPU, not just a normal CPU. A server-grade CPU is required. It also needs modern memory, at least DDR5. It is not good with very old computer memory. Disk type matters as well. It is very good at NVMe  SSDs but not good with old spindle hard drives. If a customer has modern hardware, it works very well. If a customer has legacy hardware, it will not work as expected.

I have been using Redpanda for more than three years.

Regarding stability, Redpanda is good. I can rate it ten out of ten for stability.

Redpanda scales very well. I would rate it ten out of ten for scalability.

My thoughts on support for Redpanda are that it is nine out of ten.

When I compare Redpanda with other solutions, such as Confluent  Kafka and Amazon, I find that overall Redpanda is good compared to any other option. Kafka is good, but it is too resource intensive. Redpanda is good and requires fewer resources.

I find that the deployment of Redpanda is easy. It takes just less than an hour to deploy Redpanda. Operations are a bit tough, but deployment is easy.

In terms of pricing, Redpanda is free. We do not have to pay anything. It is not open source, but it is free.

I would recommend this product to other users. Overall, I would rate Redpanda as very good and would give it a ten out of ten rating.

I have worked with Redpanda  for the past two to three months. Mainly in the tech industry or software industry, there's a huge rise of streaming data.

Redpanda  serves as a very reliable and fast message broker, which lets you build applications asynchronously. The major use case is for my project specifically, we're using it for a monitoring system that we're building.

I haven't worked on enterprise-grade applications. It's a project for academic research for the college.

The industry standard for this kind of platform is Kafka. Confluent  Kafka has acquired it. Kafka is an open-source platform built by Apache. Confluent  is the commercial version of it.

The major improvement of Redpanda over Kafka is firstly, good documentation. Redpanda's documentation is very easily understandable, and they have a lot of examples. In addition to that, most of the setups include using another technology called Docker , which I am very familiar with.

Setting up technologies using Docker  is very convenient to me, and Redpanda has provided many templates for that. Redpanda has its own built-in metrics exporter, making it easier to monitor and check performance.

What makes Redpanda superior is its performance since it's written in C++. C++ is pretty much the standard for high-performance applications.

Recently, for the documentation, they've built their own AI chatbot, which is focused on giving you answers based on their documentation. While using that, I did not find it to be very good. Maybe due to the fact that it's a first prototype and was very recently released. However, from a product perspective, I do not have any problems.

I have used Redpanda for the past two to three months.

I am definitely satisfied with the stability provided by Redpanda.

In the free version, it's still working with containers itself. They've provided the template for one and three nodes for horizontal scaling.

For our project, we've set up five containers. The template is ready in the Dockerfile they provided. You just add the existing nodes and give them the respective configurations.

I have not escalated any questions to technical support.

The industry standard for this kind of platform is Kafka. Confluent Kafka has acquired it. Kafka is an open-source platform built by Apache. Confluent is the commercial version of it. I have worked with Kafka, so I pretty much know how it works. I found Redpanda's documentation and setup to be more straightforward.

Once I learned about this platform, I read about it a bit and then I looked at how they set it up on their systems. It's a simple straightforward Dockerfile. If you have Docker installed on your system, you can just spin it up quickly without any issues.

Redpanda is actually a commercial platform, but they do provide free versions as well. I've been working only with the free versions.

Anyone finishing their bachelor's degree in computer science engineering would have had some hands-on experience with Kafka and understand how it works. Shifting from Kafka to Redpanda would be very simple for them.

I'd rate the solution ten out of ten.

The main use case is a message stream. Just to process the data with window processing.

Operationally, it is good.

Two projects use artificial intelligence in combination with Redpanda  right now.

The best thing is that it takes less resources compared to Kafka. We have a bad experience with Kafka; we need to restart the services because it's built on Java and requires a huge space on servers.

Redpanda  is based on C++ and is native to the operating system. Basically, it is memory efficient. We like it.

It's easy to learn to use Redpanda for the first time because it supports the Kafka client as a developer. It supports native Kafka clients.

And from the operations perspective, it's a pretty standard installation. The online community support is also available.

The command-line tools need to be improved. To quickly check the status of the topics and all.

For example, if I wanted to check how many messages are in a topic, there is a command, but we need to use different tools to get that information. It's not really built into the native tool.

I have been using it for three years. I use version 25.3.3

I never experienced any bugs. So far, so good.

I tested it with ten-plus nodes, and it's highly scalable.

We are about 55 people, 12 people actively use it, and we use it for six products.

I never needed to contact the customer service and support because it's working as expected.

We opted for Kafka for two reasons.

• One is we used to use Kafka, and we had to restart Kafka every quarter or so due to memory issues.
• Compared to Kafka, Redpanda is stable and resource consumption is also very effective. It takes less resources compared to other tools.

It's very easy. The cluster setup is also very easy in Redpanda compared to others.

NO

Yes.

It's free. Everybody can use it, only support is paid.

We evaluated three: Kafka, Redpanda, and ActiveMQ .

I would recommend Redpanda to others because it's easy to set up, consumes less resources, and is stable compared to other tools.

Overall, I would rate it an eight out of ten because there is room for improvement. So far, it's nothing but a rewrite of Kafka. It needs improvement on some of the command-line tools. And the second thing is it doesn't have a management graphical user interface. That also needs improvement.