More comparisons

Azure SignalR Service vs deepstream.io

This Azure SignalR Service vs deepstream.io comparison was created based on reviews from developers and our best attempts to perform analysis by looking at documentation and other publicly available resources.

Easily build complete, trusted realtime functionality.

Take our APIs for a spin

Azure SignalR Service

deepstream.io

Getting started and developer experience

Time to "hello world"

Reviewed by 3+ independent developers

Ratings were given based on the average amount of time it takes to sign up to a new account and publish the first message.

4 / 5

5 = <30 min
4 = 30 min - 1 h
3 = 1-2 h
2 = 2-4 h
1 = 4+ h

View code example
4 / 5

5 = <30 min
4 = 30 min - 1 h
3 = 1-2 h
2 = 2-4 h
1 = 4+ h

View code example

Demos / Tutorials

A selection of online demos and tutorials so you can test and see the code in action.

Explore Ably's tutorials for our pub/sub messaging platform

Documentation

Reviewed by 3+ independent developers

Explore Ably's documentation for our pub/sub messaging platform
3.67 / 5

Getting started guides / 5

Information architecture and developer journey / 5

API reference documentation / 5

Readability, design and navigation / 5

Quality of code / 5

Breadth and quality of tutorials / 5

“The existing documentation is largely OK, most things are explained clearly, and the information is structured logically. Sometimes you may come across pages that were clearly written fast, so the quality of writing may not always be great. There are quickstarts in about seven languages, which is always nice to see, and there’s a diversity of handy how-to guides. Now, for the less positive things. First of all, the API reference seems to be a relatively short page on GitHub and nothing more. Secondly, there are only two tutorials embedded in the docs - and they both focus on building a chat room. More tutorials and more use cases, please! The UI is minimalistic. While sometimes this is a good thing, in this case, it just makes the SignalR documentation (and indeed the product) look and feel like an unimportant or second-hand solution.”

4 / 5

Getting started guides / 5

Information architecture and developer journey / 5

API reference documentation / 5

Readability, design and navigation / 5

Quality of code / 5

Breadth and quality of tutorials / 5

The deepstream.io documentation is nicely written and features several components: GitHub, guides, tutorials, docs, blog, and miscellaneous info, such as FAQs. There’s a Getting Started guide using HTTP API, JavaScript Client API, and Java Client API, with code snippets throughout. The API reference is detailed for each SDK and information architecture is pretty nice. Most sections of the documentation are well cross-linked; however, the navigation experience could be improved, especially around the API reference – some links return 404. Also, some documentation pages need to be revamped. The documentation provides tutorials for most deepstream.io features but unfortunately, there’s no demos. The documentation for the core features (data-sync, records, auth, permissions, events, rpc, etc.) is well written, but could be more detailed.

Dashboard or dev console

Reviewed by 3+ independent developers

Sign up for free and explore Ably's pub/sub messaging platform
4.33 / 5

Ease of use / 5

Stats and reports / 5

Functionality / 5

“The Azure portal is intuitive and pretty easy to use. It provides diverse functionality that allows you to view, create, and manage all the resources you need, such as SingalR instances and connection keys. Among other things, you can also control IAM roles, set up integrations and alerts, and view various logs and metrics, such as the number of connections. More contextual help such as tooltips would be nice, as it’s sometimes unclear what some settings can do."

4 / 5

Ease of use / 5

Stats and reports / 5

Functionality / 5

The deepstream.io dashboard might be a bit overwhelming to use or understand without documentation. The dashboard allows you to explore realtime data, configure permission, as well as view important metrics over time such as total messages used, active connections, cluster size, and CPU usage.

The UI design is a bit outdated and could perhaps benefit from a revamp.

SDKs

Note: Only official SDKs were taken into account.

Explore Ably's 25+ SDKs for our pub/sub messaging platform

4 SDKs

Including:

  • .NET

  • Java

  • JavaScript

3 SDKs

Including:

  • JavaScript 

  • Java

  • Android

API structure

Reviewed by 3+ independent developers

4 / 5

API consistency across SDKs / 5

Well structured / 5

Intuitive / 5

Simple / 5

“The API structure is decently consistent, intuitive, and simple across most SDKs. The Python SDK is a bit obscure, especially when you want to make use of Azure Functions. The entire API reference documentation seems to be available only on GitHub (perhaps not the best platform), and there’s not much API documentation in general. There is a Swagger spec, though, which makes testing the API easier.”

3 / 5

API consistency across SDKs / 5

Well structured / 5

Intuitive / 5

Simple / 5

deepstream.io tries to keep its APIs as consistent as possible across SDKs on the client side. However, the APIs are quite complex and counter-intuitive – too many methods make it not straightforward to perform operations in the same order. There’s a separate reference page for each API, which is not great. Developers need to take their time to go through the API reference to be able to use them properly. Documentation and tutorials are helpful in this regard.

Azure SignalR Service

deepstream.io

"Hello world" code example
// CREATE A HUB

public class ChatHub : Hub
{
    public Task SendMessage(string user, string message)
    {
        return Clients.All.SendAsync("ReceiveMessage", user, message);
    }
}

// SEND MESSAGES TO CLIENTS

public Task SendMessage(string user, string message)
{
    return Clients.All.SendAsync("ReceiveMessage", user, message);
}

public Task SendMessageToCaller(string user, string message)
{
    return Clients.Caller.SendAsync("ReceiveMessage", user, message);
}

public Task SendMessageToGroup(string user, string message)
{
    return Clients.Group("SignalR Users").SendAsync("ReceiveMessage", user, message);
}

// CLIENT CONNECTING TO HUB

using System;
using System.Threading.Tasks;
using System.Windows;
using Microsoft.AspNetCore.SignalR.Client;

namespace SignalRChatClient
{
    public partial class MainWindow : Window
    {
        HubConnection connection;
        public MainWindow()
        {
            InitializeComponent();

            connection = new HubConnectionBuilder()
                .WithUrl("http://localhost:53353/ChatHub")
                .Build();

            connection.Closed += async (error) =>
            {
                await Task.Delay(new Random().Next(0,5) * 1000);
                await connection.StartAsync();
            };
        }

        private async void connectButton_Click(object sender, RoutedEventArgs e)
        {
            connection.On<string, string>("ReceiveMessage", (user, message) =>
            {
                this.Dispatcher.Invoke(() =>
                {
                   var newMessage = $"{user}: {message}";
                   messagesList.Items.Add(newMessage);
                });
            });

            try
            {
                await connection.StartAsync();
                messagesList.Items.Add("Connection started");
                connectButton.IsEnabled = false;
                sendButton.IsEnabled = true;
            }
            catch (Exception ex)
            {
                messagesList.Items.Add(ex.Message);
            }
        }

        private async void sendButton_Click(object sender, RoutedEventArgs e)
        {
            try
            {
                await connection.InvokeAsync("SendMessage", 
                    userTextBox.Text, messageTextBox.Text);
            }
            catch (Exception ex)
            {                
                messagesList.Items.Add(ex.Message);                
            }
        }
    }
}

// RECEIVE MESSAGES FROM HUB

connection.On<string, string>("ReceiveMessage", (user, message) =>
{
    this.Dispatcher.Invoke(() =>
    {
       var newMessage = $"{user}: {message}";
       messagesList.Items.Add(newMessage);
    });
});
// Client A
function eventCallback (data) {
    //callback for incoming events 
}

//Subscribing to an event 
client.event.subscribe ('news/sports' , eventCallback)

//Removing specific callback to the event
client.event.unsubscribe ('news/sports' , eventCallback)

//Removing all subscriptions to the event
client.event.unsubscribe ('news/sports')

// Client B
client.event.emit ('news/sports' , 'football is happening' )

// Client B
client.event.listen ('^news /.*' , (match, response) => {
    console.log(match) // 'news/sports'
    if (/* if you want to provide */) { 
        // start publishing data via ` client.event.emit(eventName, /* data */)`
        response.accept()

        response.onStop(() => {
            // stop publishing data
        }) 
    } else {
        response.reject() // let deepstream ask another provider
    }
})

Azure SignalR Service

deepstream.io

Realtime features

Pub/Sub messaging

Pub/Sub is a design pattern that lets any number of publishers (producers) push messages to channels (also known as topics). Multiple subscribers (consumers) can subscribe to a channel to consume published messages.

Explore Ably's pub/sub messaging implementation

Limited

You can use Azure SignalR in combination with backplanes such as Redis or Event Grid, which support the pub/sub model. Azure SignalR service is essentially a proxy for client-server communication and does not provide pub/sub capabilities by itself.

Message queues

A message queue is a form of asynchronous service-to-service communication. Messages are stored on a queue until they are processed. Note that each message is only consumed by one subscriber (consumer).

Explore Ably's message queues implementation

Presence

Presence enables you to track the online and offline status of devices and end-users in real time and to store their state. Essential for chat apps and multiplayer games.

Explore Ably's presence implementation

Message history

Message history provides a means to retrieve previously published messages. For this to be possible, message data must be stored (persisted) somewhere.

Explore Ably's message history implementation

Connection state recovery (stream resume)

In the case of unreliable network conditions, clients may suddenly disconnect.Connection state recovery ensures that when they reconnect, the data stream resumes exactly where it left off.

Explore Ably' s connection state recovery implementation

Guaranteed message ordering

Ordering ensures that messages are delivered to consumers in the same order that producers publish them.

Explore Ably' s guaranteed message ordering implementation

Exactly-once semantics

Exactly-once is a system-wide data integrity guarantee that ensures each message is delivered to consumers exactly-once.

Explore Ably' s idempotent publishing implementation

Message delta compression

Message delta compression enables you to only send the changes from the previous message to subscribers each time there’s an update, instead of the entire message. Useful for use cases where there is a significant degree of similarity between successive messages.

Explore Ably' s message delta compression implementation

Native push notifications

Native push notifications can be used to deliver messages even when clients are offline. Useful for geolocation updates or news alerts.

Explore Ably's push notifications implementation

Webhooks

Webhooks provide a mechanism to get messages and other types of events (such as clients entering or leaving channels) pushed to your servers over HTTP.

Explore Ably's webhooks implementation

Limited

You can send webhooks to Azure SignalR Service via bindings for Azure Functions. However, SignalR Service can only consume webhooks, but it cannot send webhooks to other systems.

Limited

deepstream.io only uses webhooks for user authentication.

Serverless functions

A serverless function is essentially an isolated, single-purpose piece of code that is only executed when it’ triggered by an event. For example, you can use serverless functions to send a welcome message to clients when they become present on chat channels.

Note that serverless functions are usually fully managed by cloud vendors.

Explore Ably's serverless functions implementation

Limited

Can only trigger Azure Functions. Azure SignalR Service doesn’t have in-built support for other serverless platforms, such as AWS Lambda or Google Cloud Functions.

Built-in integrations

Which popular services & systems are Azure SignalR Service and deepstream.io integrated with?

Explore Ably's library of integrations

Webhooks

  • Custom
  • IFTTT
  • Zapier

Serverless functions

  • AWS Lambda
  • Cloudflare Workers
  • Google Cloud Functions
  • Microsoft Azure Functions

Streaming

  • Apache Kafka
  • Apache Pulsar
  • Amazon Kinesis
  • Amazon SQS
  • RabbitMQ
  • AMQP

Webhooks

  • Custom
  • IFTTT
  • Zapier

Serverless functions

  • AWS Lambda
  • Cloudflare Workers
  • Google Cloud Functions
  • Microsoft Azure Functions

Streaming

  • Apache Kafka
  • Apache Pulsar
  • Amazon Kinesis
  • Amazon SQS
  • RabbitMQ
  • AMQP

Known limits and constraints

Find out practical limits, such as the maximum message size, or the maximum number of concurrent connections.

Explore the practical limits of the Ably pub/sub messaging platform

Throughput 

Unknown

Maximum message size

No size limit. Large messages are split into smaller chunks (max 2 KB), transmitted separately, and reassembled on the consumer side.   

Maximum number of units

100 SignalR Service units (sub-instances) per instance (applies to the Standard tier).

Maximum number of concurrent connections

1000/unit (applies to the Standard tier)

Publisher throughput 

160.000-200.000 updates per second for Single Node

4,000,000,000 messages per hour for Cluster

Maximum message size

1MB 

Maximum number of topics

Unknown

Maximum number of connections

250 for Single Node

750 for Cluster

Supported development platforms, languages, open protocols and cloud models

Development platforms & operating systems

Which popular development platforms and operating systems do Azure SignalR Service and deepstream.io support via official SDKs?

Explore the development platforms supported by Ably
  • Android
  • Java / JVM
  • iOS
  • macOS
  • iPadOS
  • tvOS
  • watchOS
  • Mono
  • .NET
  • Android
  • Java / JVM
  • iOS
  • macOS
  • iPadOS
  • tvOS
  • watchOS
  • Mono
  • .NET

Languages

Which popular programming languages do Azure SignalR Service and deepstream.io support via offical SDKs?

Explore the programming languages supported by Ably
  • JavaScript
  • Node.js
  • TypeScript
  • Java
  • Objective-C
  • Swift
  • Go
  • PHP
  • Python
  • Ruby
  • Flutter
  • Clojure
  • Scala
  • JavaScript
  • Node.js
  • TypeScript
  • Java
  • Objective-C
  • Swift
  • Go
  • PHP
  • Python
  • Ruby
  • Flutter
  • Clojure
  • Scala

Open protocols

Which popular open protocols do Azure SignalR Service and deepstream.io support?

Explore the open protocols supported by Ably
  • WebSocket
  • HTTP
  • AMQP
  • MQTT
  • STOMP
  • SSE
  • Webhooks
  • WebSocket
  • HTTP
  • AMQP
  • MQTT
  • STOMP
  • SSE
  • Webhooks

Cloud models

Which popular cloud models do Azure SignalR Service and deepstream.io support?

  • Self-hosted
  • Cloud-managed
  • Serverless
  • Globally-distributed
  • Self-hosted
  • Cloud-managed
  • Serverless
  • Globally-distributed
Global and reliable edge service

Edge messaging network with latency-based routing

Latency-based routing ensures that clients are always routed to the nearest datacenter and point of presence.

Explore Ably's routing mechanism that mitigates network and DNS issues

Multi-region data replication (message durability)

Multi-region data replication (storage) protects against single points of failure and ensures message data durability.

Learn how Ably ensures message durability

Limited

deepstream.io stores data in a combination of storage and cache layers for distribution across multiple nodes. However this platform also relies on shared devices such as a Redis for a message bus which itself (Redis layer) might become a point of failure and congestion.

Uptime SLAs

Here’s what the most common SLAs amount to in terms of downtime over a calendar year:

99.999% SLA = 5m 15s downtime per year

99.99% SLA = 52m 35s downtime per year

99.95% SLA = 4h 22m 58s downtime per year

99.9% SLA = 8h 45m 56s downtime per year

99% SLA = 3d 15h 39m 29s downtime per year

Source: https://uptime.is/

99.95% for premium accounts

99.0% for non-premium accounts

No specific uptime SLA

Quality of Service

What QoS guarantees do Azure SignalR Service and deepstream.io provide natively?

Explore Ably's availability and uptime guarantees for our pub/sub messaging platform
  • Multi-region data replication (storage)
  • Exactly-once semantics
  • Guaranteed message ordering
  • Connection state recovery (stream resume)
  • Multi-region data replication (storage)
  • Exactly-once semantics
  • Guaranteed message ordering
  • Connection state recovery (stream resume)
Security

API key authentication

The simplest way to authenticate. Involves using private API keys that you can usually create and edit via a dashboard. Recommended to be used server-side, as private API keys shouldn’t be shared with untrusted parties.

Explore Ably's implementation of API key authentication

Token-based authentication

Which popular token-based authentication mechanisms do Azure SignalR Service and deepstream.io support?

Note that token-based authentication is usually the recommended strategy on the client-side as it provides more fine-grained access control and limits the risk of credentials being compromised.

Explore Ably's implementation of token-based authentication
  • Ephemeral tokens
  • JWTs
  • Ephemeral tokens
  • JWTs

Configurable rules and permissions

Which types of configurable rules and permissions do Azure SignalR Service and deepstream.io support?

Explore Ably's configurable rules and permissions
  • API keys rules and permissions
  • Operation rules and permissions
  • Admin rights
  • API keys rules and permissions
  • Operation rules and permissions
  • Admin rights

Message encryption

Which types of message encryption do Azure SignalR Service and deepstream.io support?

Explore Ably's message encryption mechanisms
  • Encrypted at rest
  • Encrypted in transit
  • Message payload encryption
  • Encrypted at rest
  • Encrypted in transit
  • Message payload encryption

Formal certifications

Which formal certifications are Azure SignalR Service and deepstream.io compliant with?

Explore Ably's security and compliance for our pub/sub messaging platform
  • SOC 2 TYPE I
  • SOC 2 Type II
  • HIPAA
  • EU GDPR
  • SOC 2 TYPE I
  • SOC 2 Type II
  • HIPAA
  • EU GDPR
Pricing & Support

Free package

What do the free packages offered by Azure SignalR Service and deepstream.io consist of?

Explore Ably's free package for our pub/sub messaging platform

The Free plan includes one free Azure SignalR Service unit, which has a cap of 20.000 messages per day and can sustain 20 concurrent connections.

N/A

Pricing model

How are the Azure SignalR Service and deepstream.io pricing models calculated?

Explore Ably's pricing model for our pub/sub messaging platform

In addition to the Free plan mentioned above, Azure SignalR Service comes with a Standard plan that has the following costs:

- $ 1.61 price per unit per day

- $1 per million messages (first million free)

deepstream.io pricing is usage-based and calculated per hour.

$0.052 per hour for a six-instance cluster on AWS EC2  t2.medium

$0.008 per hour for a single node on cache.t2.medium

It’s unclear how pricing is calculated for the Enterprise package.

Enterprise package

What benefits do the Azure SignalR Service and deepstream.io enterprise packages offer?

Explore Ably's enterprise package for our pub/sub messaging platform

There is no Azure SignalR Service-specific enterprise package.

No Enterprise package is explicitly specified. However, it can be obtained by contacting the vendor directly.

Community

Reviewed by 3+ independent developers

Explore Ably's community support channel for our pub/sub messaging platform
3.50 / 5

Presence on multiple channels / 5

Size and activity / 5

“Azure SignalR Service has community engagement on various different channels, such as Stack Overflow, Gitter, Microsoft & ASP .NET Forums, or GitHub. The community doesn’t seem to be very large. As an example, if you look at the ASP.NET SignalR forum, you will notice there aren’t all that many conversation threads. However, as a plus, most questions do have at least a reply. GitHub also seems to have some activity around troubleshooting issues.”

5 / 5

Presence on multiple channels / 5

Size and activity / 5

deepstream.io has a strong community presence on various channels, such as Stack Overflow, Slack, GitHub, and Twitter. Their community is active, approachable, and helpful, and generates a lot of content: 110 repositories, 6.8K stars, and 383 forks on GitHub. Similarly, there are over 100 questions on Stack Overflow, 690 users on Slack and more than 1000 followers on Twitter. There are no events, and no community presence on Discourse, and Gitter.

Support

What types of support options and response times do Azure SignalR Service and deepstream.io offer?

Explore Ably's support options for our pub/sub messaging platform

General support options 

Email, support ticket, phone, technical documentation, community support (e.g. forums or StackOverflow). 

Enterprise support

The Professional Direct (ProDirect) plan includes 24/7 support, a Support API (to create & manage support tickets programmatically), and operational & architectural guidance from delivery managers.

Response time

Within 8 business hours for Developer customers

Between 1 and 8 hours initial response time (depending on severity) for Standard and ProDirect customers

General Support Options

Email, technical documentation, community support.

Response time

< 24 hours for non-technical queries

< 1 hour for production system down incident (only applies to Enterprise support package)

Disclaimer: The information presented for Azure SignalR Service was last updated on 5 November 2020 and on 28 February 2021 for deepstream.io. It is possible that some details may now be out of date. If you think that’s the case, please let us know so we can update them. In any case, you should not rely solely on the information presented here and must check with each provider before deciding to integrate or buy any of these two solutions.


View more comparisons