# Stream sizing

When considering sizing for Stream it is important to understand that the traffic mix at customer sites varies widely. Some customers have traffic mixes that skew towards larger flows (think file transfers), and some will skew towards smaller flows. Performance will be lower when the traffic mix skews towards smaller flows as there will be more metadata to process. The below are guidelines for average traffic mixes and should not be considered absolute. Throughput refers to the amount of traffic observed by Sensors that forward metadata to the Vectra platform.

| **Hypervisor/Cloud** |               **VM Type**              | **Cores** | **Memory** | **Storage** | **\~ Throughput** |
| :------------------: | :------------------------------------: | :-------: | :--------: | :---------: | :---------------: |
|        VMware        |          vSphere 6.5 or later          |     2     |    8 GB    |    100 GB   |   Up to 2.5 Gbps  |
|        VMware        |          vSphere 6.5 or later          |     4     |    8 GB    |    150 GB   |   2.5 to 5 Gbps   |
|        VMware        |          vSphere 6.5 or later          |     8     |    16 GB   |    150 GB   |    5 to 10 Gbps   |
|        VMware        |          vSphere 6.5 or later          |     16    |    64 GB   |    150 GB   |   10 to 20 Gbps   |
|        Hyper-V       | Windows Server 2016 w/ HW v8 or higher |     2     |    8 GB    |    100 GB   |   Up to 2.5 Gbps  |
|        Hyper-V       | Windows Server 2016 w/ HW v8 or higher |     4     |    8 GB    |    150 GB   |   2.5 to 5 Gbps   |
|        Hyper-V       | Windows Server 2016 w/ HW v8 or higher |     8     |    16 GB   |    150 GB   |    5 to 10 Gbps   |
|        Hyper-V       | Windows Server 2016 w/ HW v8 or higher |     16    |    64 GB   |    500 GB   |   10 to 20 Gbps   |
|          KVM         |     Standard PC (Q35 + ICH9, 2009)     |     2     |    8 GB    |    100 GB   |   Up to 2.5 Gbps  |
|          KVM         |     Standard PC (Q35 + ICH9, 2009)     |     4     |    8 GB    |    150 GB   |   2.5 to 5 Gbps   |
|          KVM         |     Standard PC (Q35 + ICH9, 2009)     |     8     |    16GB    |    150 GB   |    5 to 10 Gbps   |
|          KVM         |     Standard PC (Q35 + ICH9, 2009)     |     16    |    64 GB   |    500 GB   |   10 to 20 Gbps   |
|          AWS         |                c5.xlarge               |     4     |    8 GB    |    50 GB    |    Up to 5 Gbps   |
|          AWS         |               c5.2xlarge               |     8     |    16 GB   |    50 GB    |    5 to 10 Gbps   |
|          AWS         |               c5.4xlarge               |     16    |    32 GB   |    50 GB    |   10 to 20 Gbps   |
|         Azure        |            Standard\_DS4\_v2           |     8     |    28 GB   |    50 GB    |   Up to 10 Gbps   |
|         Azure        |            Standard\_DS5\_v2           |     16    |    56 GB   |    50 GB    |   10 to 20 Gbps   |
|          GCP         |              e2-standard-4             |     4     |    16 GB   |    50 GB    |    Up to 5 Gbps   |
|          GCP         |              e2-standard-8             |     8     |    32 GB   |    50 GB    |    5 to 10 Gbps   |
|          GCP         |             e2-standard-16             |     16    |    64 GB   |    50 GB    |   10 to 20 Gbps   |

**Vectra M Series Physical Appliance**

The M series physical appliances can support up to \~75 Gbps of throughput.

{% hint style="info" %}
**Please Note:**

* Contact your account team or Vectra support for guidance on special situations, abnormal traffic mixes, or throughput needs outside of what is documented above.
* Vectra recommends that Stream VMs are configured to use storage local to the hypervisor and are not stored on a SAN. Stream VMs require extremely high throughput from their disk storage and this throughput cannot normally be sustained by SAN systems without impact to other SAN users.
* Please see [VMware vSensor specifications](broken://pages/3cJnbqaxR7C1h6axNXS6) for guidance on supported CPUs, storage/SANs, networking requirements, vMotion, Enhanced vMotion compatibility, and unsupported hypervisors. While Sensor and VMware specific, this article applies to Stream VMs as well. The general guidance in the article also applies to Hyper-V and KVM deployments.
  {% endhint %}


---

# Agent Instructions: Querying This Documentation

If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter:

```
GET https://docs.vectra.ai/deployment/stream/stream-sizing.md?ask=<question>
```

The question should be specific, self-contained, and written in natural language.
The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.