From 26344af1fb0ed7ea4e776a1f1b857821346c9e43 Mon Sep 17 00:00:00 2001 From: J-Loudet Date: Thu, 1 Feb 2024 10:34:31 +0000 Subject: [PATCH] deploy: 733a1323c4cebaa3105615085c4bd097c297b92a --- blog/2024-01-31-zenoh-flow-getting-started/index.html | 2 +- 1 file changed, 1 insertion(+), 1 deletion(-) diff --git a/blog/2024-01-31-zenoh-flow-getting-started/index.html b/blog/2024-01-31-zenoh-flow-getting-started/index.html index e763f626..2cb191af 100644 --- a/blog/2024-01-31-zenoh-flow-getting-started/index.html +++ b/blog/2024-01-31-zenoh-flow-getting-started/index.html @@ -1,6 +1,6 @@ Zenoh-Flow 0.6.0-rc: Getting Started · Zenoh - pub/sub, geo distributed storage, query
-

Zenoh-Flow 0.6.0-rc: Getting Started

31 January 2024 -- Paris.

At ZettaScale, we are developing a next-generation middleware called Zenoh. Zenoh gained a lot of traction with very positive feedback from diverse industries, including Robotics, Industrial Automation, Automotive and more. If you haven’t tried it yet and you think it could help you, don’t hesitate, we guarantee it’s easy and worth your time!

This blog post will focus on another exciting project we are building: Zenoh-Flow. As its name indicates, Zenoh-Flow brings together the control of data flow programming and the power of Zenoh. Think of distributed applications where you don’t have to know the exact location of your computing units, but simply agree on the name (and type) of the resources they are going to exchange, how they are connected and, if needed, the properties of the host(s) on which they are going to run. Add to this list a validation step, uniquely named resources, great performance and you would start to have a good idea of what Zenoh-Flow has to offer!

If you want more details and an introduction to its core concepts, we have already written a blog post earlier that we invite you to read — we might skip explaining few concepts that we introduced before.

Let us dive in!

Data Flow descriptor

Zenoh-Flow 0.6.0-rc: Getting Started

💡️
All the code examples given below work with the tagged version blog-post-31-01-2024. We will update this blog post once we have released Zenoh-Flow 0.6.0-rc. To clone this version you can use the following command:
git clone https://github.com/eclipse-zenoh/zenoh-flow --branch blog-post-31-01-2024 --depth 1

To facilitate the discovery of Zenoh-Flow we have included in the repository an example data flow called “getting started”. Similar to the very first program you write in every programming language, its purpose is to print a sentence akin to “Hello, world!”. However, as we are building a data flow and using Zenoh, we have to introduce slightly more logic and we have to make our flow interact via Zenoh.

Hence, our getting started will subscribe to the resource zf/getting-started/hello — a Source — in the hope of receiving names, then it will forward this name to a node that will generate a greeting — an Operator — and finally forward the generated greeting to be both published on Zenoh, on the resource zf/getting-started/greeting, and written to a file — two Sinks.

This flow is already written in the file: zenoh-flow/examples/flows/getting-started.yaml.

Data Flow Descriptor

name: getting-started
+
Zenoh-Flow 0.6.0-rc: Getting Started
31 January 2024 -- Paris.
At ZettaScale, we are developing a next-generation middleware called Zenoh. Zenoh gained a lot of traction with very positive feedback from diverse industries, including Robotics, Industrial Automation, Automotive and more. If you haven’t tried it yet and you think it could help you, don’t hesitate, we guarantee it’s easy and worth your time!
This blog post will focus on another exciting project we are building: Zenoh-Flow. As its name indicates, Zenoh-Flow brings together the control of data flow programming and the power of Zenoh. Think of distributed applications where you don’t have to know the exact location of your computing units, but simply agree on the name (and type) of the resources they are going to exchange, how they are connected and, if needed, the properties of the host(s) on which they are going to run. Add to this list a validation step, uniquely named resources, great performance and you would start to have a good idea of what Zenoh-Flow has to offer!
💡️
If you want an introduction to the core concepts of Zenoh-Flow and motivating use-cases, we have written a blog post earlier that we invite you to read — we might skip explaining few concepts that we introduced before.
Let us dive in!
Data Flow descriptor
Zenoh-Flow 0.6.0-rc: Getting Started
💡️
All the code examples given below work with the tagged version blog-post-31-01-2024. We will update this blog post once we have released Zenoh-Flow 0.6.0-rc. To clone this version you can use the following command:
git clone https://github.com/eclipse-zenoh/zenoh-flow --branch blog-post-31-01-2024 --depth 1
To facilitate the discovery of Zenoh-Flow we have included in the repository an example data flow called “getting started”. Similar to the very first program you write in every programming language, its purpose is to print a sentence akin to “Hello, world!”. However, as we are building a data flow and using Zenoh, we have to introduce slightly more logic and we have to make our flow interact via Zenoh.
Hence, our getting started will subscribe to the resource zf/getting-started/hello — a Source — in the hope of receiving names, then it will forward this name to a node that will generate a greeting — an Operator — and finally forward the generated greeting to be both published on Zenoh, on the resource zf/getting-started/greeting, and written to a file — two Sinks.
This flow is already written in the file: zenoh-flow/examples/flows/getting-started.yaml.
Data Flow Descriptor
name: getting-started
 
[Required] We start with the name of the data flow. It is a human-readable summary of what your data flow will do. Zenoh-Flow will use it when displaying information regarding its status.
vars:
   TARGET_DIR: "/home/zettascale/zenoh-flow/target"
   BUILD: "debug"