Skip to main content
For the complete documentation index for agents and LLMs, see llms.txt.

Set the Indexing Instance SizeBeta

Beta
This is coming — this page describes a feature that's still being rolled out and may not be available to you yet.

Choose how much compute the indexing pod gets. The indexing pod converts your files, splits them into chunks, and writes them to the document store.

About This Task​

If an index processes big files and indexing is slow or fails, choose a larger instance size in the index settings. You can change the size while the index is enabled or disabled.

The Instance size list shows only the sizes your organization has made available. An organization admin makes sizes available in Settings > Organization > Compute. For steps, see Managing Compute Settings.

If an admin turned off the size your index already uses, that size still appears in the list, so the index keeps showing its current size. You can keep it or switch to an enabled size. While an index has an instance size set, deployed or not, an organization admin can't turn that size off. To free the size, switch the index to another size, or reset it to the default pod size.

Sizes and Compute​

Each size sets the memory limit of the indexing pod. An index without a size uses the default of 8 GiB. Each option in Instance size shows its compute. An index gets a CPU request and a memory limit:

  • Small: 0.4+ vCPU, 4 GiB memory
  • Medium: 0.9+ vCPU, 8 GiB memory
  • Large: 1.7+ vCPU, 16 GiB memory
  • Extra large: 3.5+ vCPU, 32 GiB memory

In the API, these sizes are small, medium, large, and xl. The + means indexing can use more CPU when it's free. Memory is a hard limit. The same size name gives a service replica different compute. For example, Medium is 1 vCPU and 4 GiB memory for a service. For details, see Create a Service.

Change the Instance Size​

  1. Log in to Haystack Enterprise Platform and open Indexes.
  2. Click the index name and open the index settings.
  3. In Instance size, select the size you want. You need permission to edit indexes to change it. While the list is loading, it shows only the index's current size.

To check a size's memory limit before you save, look at the line under the Instance size list. It shows the limit of the selected size, for example, "Medium: 8 GiB memory," and updates when you select another size. If indexing runs out of memory on large files, which shows up as pod restarts or out-of-memory errors, pick a bigger size.

If the index is deployed, it restarts its indexing pods on the new size. Files that are already indexed aren't reprocessed. The size doesn't change how many credits the index uses.

Check Whether Your Index Needs a Different Size​

Open the index's Metrics tab. The CPU and memory charts show the limit of the current size, and the advice card recommends a size change when usage is close to the limit. For details, see Check Index Hardware Metrics.