🤔 Got questions? Schedule an office hours session.
Node RPC
Guides
Request Batching

Batching JSON-RPC requests

JSON RPC request batching allows you to send several JSON-RPC method invocations within a single HTTPS call. A batch request consists of an array of individual JSON-RPC requests. Batching JSON-RPC requests brings performance improvements, by reducing the network latency of performing several HTTP requests you would otherwise do.

You can use batching:

  • directly, by sending an HTTPS request that contains an array of individual JSON RPC calls,
  • using Ethers, by instantiating a JsonRpcBatchProvider
  • using Viem, by configuring the batching parameter when creating a client via createPublicClient.

The result of this operation is an array of responses, in the order that corresponds to each individual request from the batch.

If a particular request fails:

  • The batch request will return 200 OK HTTP status
  • The response object corresponding to the failing JSON-RPC call will be failing.

Request batching is supported over HTTPS only. There is no batching support over WSS currently.

Example: direct batching

This example shows low-level request batching by issuing an array of JSON RPC methods calls.

batch-call.sh
curl https://mainnet.gateway.tenderly.co/$TENDERLY_NODE_ACCESS_KEY \
-X POST \
-H "Content-Type: application/json" \
-d '[
  {"jsonrpc": "2.0", "id": 1, "method": "eth_blockNumber", "params": []},
  {"jsonrpc": "2.0", "id": 2, "method": "eth_accounts", "params": []},
  {"jsonrpc":"2.0","id":3,"method":"eth_getBalance","params":["0xd8da6bf26964af9d7eed9e03e53415d37aa96045","latest"]},
  {"jsonrpc":"2.0","id":4,"method":"eth_getStorageAt","params":["0xc02aaa39b223fe8d0a0e5c4f27ead9083c756cc2","0x3a988d762a24303c37d08f1543db6143453b579691d5c20fed39629ff1334cca","latest"]},
  {"jsonrpc":"2.0","id":0,"method":"eth_gasPrice","params":[]},
  {"jsonrpc": "2.0", "id": 3, "method": "tenderly_simulateTransaction", "params": []}
]'

Example: Viem

Viem makes batching possible at the level of the client, configured to perform batching.

node/samples/viem/0-batch-calls.ts
import { createPublicClient, http } from "viem";
import { mainnet } from "viem/chains";
 
const client = createPublicClient({
  chain: mainnet,
  transport: http(
    `https://mainnet.gateway.tenderly.co/${process.env.TENDERLY_NODE_ACCESS_KEY}`,
    {
      batch: true,
    }
  ),
});
 
const [blockNumber, balance, ensName] = await Promise.all([
  client.getBlockNumber(),
  client.getBalance({ address: "0xd2135CfB216b74109775236E36d4b433F1DF507B" }),
  client.getEnsName({ address: "0xd2135CfB216b74109775236E36d4b433F1DF507B" }),
  client.getChainId(),
]);
console.log("Results are in");
console.log({ blockNumber, balance, ensName });
 
//  So will this
const bnPromise = client.getBlockNumber();
 
const balancePromise = client.getBalance({
  address: "0xd2135CfB216b74109775236E36d4b433F1DF507B",
});
 
console.log(await bnPromise);
console.log(await balancePromise);

Limits and pricing

The total usage footprint of a batch is expressed in TU (Tenderly Units), and it’s the sum of TU of all requests within the batch, according to the pricing.

The batch-request usage is subject to the monthly quota and rate limiting.

WebSocketsAll Integration Guides