What is our primary use case?
Our main use case for
Ably revolves around the incredibly difficult build, real-time data broadcasting, and live collaboration throughout the organization for our clients. We also use it for its power features that require instant messaging and updates without refreshing their browser. It works through WebSocket, giving instant updates without refreshing the browser, just as in the stock market where all updates come at once. Similarly, it provides live updates without refreshing the browser and utilizes multiple endpoints with hundreds of network presence globally to ensure incredible low latency, regardless of geography. It works really well for the organization and follows compliances such as HIPAA and GDPR.
I can share a specific example where we are working for a client making a system for filing taxes that is integrated through multiple APIs, Kafka, and various message receivers. Here, we have a notification service, email service, and messages service. We integrated Ably to ensure that everything is delivered to the customer without any delay, achieving 100% message delivery and survivability, accelerating the time to market for our client, focusing on developer productivity and system reliability. It has helped us ensure that our projects are delivered on time, with customers receiving proper calls without using call servers, as it automatically manages reconnections and delivers messages in the exact order they were sent.
What is most valuable?
I love Ably for its advanced pub/sub messaging capabilities, which help us with messaging Delta Compression, allowing us to send only changes from previous messages instead of entire payloads every time. For example, if I have to say 'hi', instead of sending the whole payload, it only sends the change in the current payload, reducing bandwidth consumption and enabling high-frequency data streaming. This major change has made our clients really happy as they receive notifications and messages without latency. Additionally, if a client goes offline, Ably stores messages in history for up to 72 hours, allowing clients to query history and catch up on missed events, ensuring complete stream continuity. It also supports multi-protocol, using WebSockets and gracefully falling back to server-sent events and long polling as needed, including support for MQTT and IoT devices along with pub/sub control for easy migration.
I appreciate the serverless integration that has the integrated native web hook, instantly triggering serverless functions such as AWS Lambda and Azure Function when specific real-time events occur. It also allows Kafka Connector for seamless ingestion of messages from Kafka topics to stream them out to millions of users in milliseconds.
What needs improvement?
In this AI era, Ably can improve by adding more AI integrations to help reduce manual rework through AI agents or bots.
For how long have I used the solution?
I have been working in my current field for almost 15 years.
What do I think about the stability of the solution?
Ably is very stable and we have not experienced any outages or reliability issues.
What do I think about the scalability of the solution?
Ably's scalability is excellent, as we have scaled up to 250,000 users without seeing glitches or downtimes.
How are customer service and support?
Customer support is a 10 out of 10 because when we reached out, they resolved our query within 30 to 45 minutes. I rate customer support a 10 on a scale of 1 to 10.
Which solution did I use previously and why did I switch?
Before choosing Ably, we evaluated other options such as Wowza
Stream,
PubNub, and Pusher, considering their services and ROI, but ultimately, Ably was the best suited for us.
How was the initial setup?
I faced no challenges with the configuration process as it was done well, and all environments were easily configured, whether local, test, or system integration environments.
What about the implementation team?
Procuring the solution was easy within our organization and not at all tough.
What was our ROI?
Ably has positively impacted our organization by providing a good return on investment by avoiding the labor cost of building and maintaining a custom WebSocket server. We have saved at least 60 to 70% in cost and 80 to 90% in manual labor associated with creating and maintaining those servers. This serverless infrastructure focuses on building a scalable custom real-time infrastructure in-house, saving nearly $200,000 for the organization. There is zero-DevOps overhead, eliminating the need for a dedicated Site Reliability Engineering team to manage node scaling, load balancing, and connection requests. As a result, our senior engineers can focus on core revenue-generating product features. Overall, we have saved time, money, resources, and the return on investment is significant. If we paid ₹2 or $2, we got $10 back from Ably, so it's a 5X ROI for us.
What's my experience with pricing, setup cost, and licensing?
The metering and billing experience through
AWS Marketplace is really well managed, with licensing, metering, and billing on point, and we received maximum discounts. We use cost as a basis for the number of licenses and it is cost-effective.
The pricing, setup cost, and licensing experience are well managed through AWS Marketplace, providing a genuine price. We use a consumption-based model, meaning we pay-as-you-go based on actual usage instead of full server fees. The clear real-time billing structure shows our consumption, and it offers various models, but we use the pay-as-you-go model which suits us best.
What other advice do I have?
My advice to others looking at Ably is that anyone needing a dedicated serverless message deliverable to reduce time to market, improve cost, latency, and message deliverability should definitely go for it.
Deploying Ably in our environment is really easy. The deployment is seamless and integrates easily into our workloads and systems, taking no more than a day.
Ably integrates very well with other AWS services we use, with a specific example being the CLI, which is well integrated with AWS Private Link. I rate this product a 5 out of 5.
Which deployment model are you using for this solution?
Public Cloud
If public cloud, private cloud, or hybrid cloud, which cloud provider do you use?
Amazon Web Services (AWS)