77 lines
3.3 KiB
Markdown
77 lines
3.3 KiB
Markdown
<h1> Algorand </h1>
|
|
|
|
<h2>Table of Contents</h2>
|
|
|
|
- [Introduction](#introduction)
|
|
- [Prerequisites](#prerequisites)
|
|
- [Algorand Structure](#algorand-structure)
|
|
- [Run Default Node](#run-default-node)
|
|
- [Run Relay Node](#run-relay-node)
|
|
- [Run Indexer Node](#run-indexer-node)
|
|
- [Select Capacity](#select-capacity)
|
|
|
|
***
|
|
|
|
## Introduction
|
|
|
|
[Algorand](https://www.algorand.com/) builds technology that accelerates the convergence between decentralized and traditional finance by enabling the simple creation of next-generation financial products, protocols, and exchange of value.
|
|
|
|
## Prerequisites
|
|
|
|
- Make sure you have a [wallet](wallet_connector.md)
|
|
- From the sidebar click on **Applications**
|
|
- Click on **Algorand**
|
|
|
|
## Algorand Structure
|
|
|
|
An Algorand node can be either a `Default`, `Relay` or `Indexer` node.
|
|
|
|
- Default:
|
|
- This is a non-relay node.
|
|
- It can run on Devnet, Testnet, Betanet and Mainnet.
|
|
- Relay:
|
|
- It can run only on Testnet and Mainnet.
|
|
- Indexer:
|
|
- It is a default node but with Archival Mode enabled which will make you able to query the data of the blockchain.
|
|
|
|
## Run Default Node
|
|
|
|
You can select any network you want and for the node type select Default.
|
|
|
|
If you have more than one SSH keys set, you can click on `Manage SSH keys` to select which one to use for this deployment.
|
|
|
|

|
|
|
|
After the deployment is done, SSH into the node and run `goal node status`.
|
|
|
|
Here you see your node runs on mainnet.
|
|
|
|

|
|
|
|
## Run Relay Node
|
|
|
|
Relay nodes are where other nodes connect. Therefore, a relay node must be able to support a large number of connections and handle the processing load associated with all the data flowing to and from these connections. Thus, relay nodes require significantly more power than non-relay nodes. Relay nodes are always configured in archival mode.
|
|
|
|
The relay node must be publicaly accessable, so it must have a public ip.
|
|
|
|

|
|
|
|
Once the deployment is done, SSH into the node and run `goal node status` to see the status of the node. You can also check if the right port is listening (:4161 for testnet, and :4160 for mainnet).
|
|
|
|

|
|
|
|
The next step accourding to the [docs](https://developer.algorand.org/docs/run-a-node/setup/types/#relay-node) is to register your `ip:port` on Algorand Public SRV.
|
|
|
|
## Run Indexer Node
|
|
|
|
The primary purpose of this Indexer is to provide a REST API interface of API calls to support searching the Algorand Blockchain. The Indexer REST APIs retrieve the blockchain data from a PostgreSQL compatible database that must be populated. This database is populated using the same indexer instance or a separate instance of the indexer which must connect to the algod process of a running Algorand node to read block data. This node must also be an Archival node to make searching the entire blockchain possible.
|
|
|
|

|
|
|
|
Once it's done, you can access the indexer API at port `8980` and here are the [endpoint](https://developer.algorand.org/docs/rest-apis/indexer/) you can access.
|
|
|
|
## Select Capacity
|
|
|
|
The default scenario capacity is computed based on the node (network/type) accourding to this [reference](https://howbigisalgorand.com/).
|
|
But you can still change this to higher values by selecting the option `Set Custom Capacity`.
|
|
|