---
title: "ETL Multithreading"
slug: "etl-multithreading"
updated: 2023-02-10T20:45:05Z
published: 2023-02-10T20:45:05Z
canonical: "docs.interprose.com/etl-multithreading"
---

> ## Documentation Index
> Fetch the complete documentation index at: https://docs.interprose.com/llms.txt
> Use this file to discover all available pages before exploring further.

# ETL Multithreading

#### *Introduced with the 01-15-2023 ACE release and updated with the 02-12-2023 ACE release.*

## **How does it benefit me?**

You can load data into ACE twice as fast.

## **What is it?**

Multithreading is the ability to execute multiple threads concurrently. In ACE, this allows a single file to be loaded in multiple threads at the same time, thereby speeding up the load process. Multithreading is available for the following ETL load methods.

1. Clients
2. Insurance Companies
3. Account Flags
4. Bankruptcy
5. Notes (Account)
6. Deceased
7. Disputes
8. Email Logs
9. Conversion Transactions (Transactions Load Method with Conversion = true)
10. Text Message Logs
11. Phone Logs
12. Letters
13. Legal Cases
14. Itemization Transactions
15. Accounts (Bundling must be disabled)

## **What do I have to do?**

### Set Maximum ETL Threads

From the Setup -> System page, set the maximum number of ETL threads to use.

![](https://cdn2.hubspot.net/hub/2513684/hubfs/Multithreading2-1.jpg?upscale=true&width=1106&upscale=true&name=Multithreading2-1.jpg)

**Maximum ETL Threads**The optimal number to set for Maximum ETL Threads is 10 to 30. Actual results may vary depending on file specifications and concurrent activity in your ACE application and database.

### 

### Set Multithreading to True

Set Multithreading to true for an ETL profile with one of the ETL load methods listed above. Run a job to load the data using the ETL profile.

![](https://cdn2.hubspot.net/hub/2513684/hubfs/Multithreading1.jpg?upscale=true&width=1064&upscale=true&name=Multithreading1.jpg)

When Not to UseThe order of the data being loaded using this option is not explicit. Therefore, if data being loaded is dependent on previous data being loaded from the same file, multithreading should not be used.