1. Home
  2. Getting Started
  3. Hardware Performance Guide for the OmniCenter Virtual Appliance
  1. Home
  2. OmniCenter Reference
  3. Hardware Performance Guide for the OmniCenter Virtual Appliance

Hardware Performance Guide for the OmniCenter Virtual Appliance

Deployment

Netreo’s OmniCenter product is deployed as a virtual appliance (VA), typically hosted within the customer’s virtualization environment. All of OmniCenter functions and features are fully supported when deployed as a VA.

These performance guidelines will assist customers in allocating the appropriate amount of resources to an OmniCenter VA (OCVA). This is not a comprehensive guide, and is intended only to give a general idea of the performance levels that can be expected based on the amount of resources allocated. For additional guidance when setting up your OCVA, please feel free to contact Netreo technical support for customized advice.

OmniCenter is installed in virtual environments by deploying a machine image. Do not provision a new guest before deploying the image.

OmniCenter uses a Linux-based core for basic hardware support and OS functions, and is designed to make maximum use of the available hardware resources. For this reason, Netreo does not recommend deploying an OCVA in virtual environments that are heavily oversubscribed, as performance may be adversely affected.

OmniCenter can be configured with many optional features such as application response time (ART) monitoring, email and exchange monitoring, traffic flow analysis, and IP Telephony management. OmniCenter deployments using these features (especially traffic flow and IP Telephony) will generally require more resources for a given size of environment. OmniCenter can be configured in excess of these guidelines up to the limit of your software device license.

You should also consider carefully the merits of monitoring your VM environment from a guest inside the environment, as a virtual environment outage could disable OmniCenter and prevent you from being alerted. If your VM environment is not highly redundant and stable, Netreo recommends the use of dedicated hardware.

General Requirements

  • VMware vSphere 5.0 or later, Microsoft Hyper-V or Scale HC3 hypervisor.
  • Intel or AMD processors >= 2.0 GHz with hardware virtualization support.
  • 8 GB RAM minimum.
  • Disk storage requirements may vary by environment, the values listed below should cover 95% of implementations.
  • SSD storage is highly preferred, especially for environments over 1000 devices. For environments over 2,500 devices, or environments over 300 devices where traffic flow technologies (NetFlow, sFlow, IPFIX) are in use, SSD-based storage is mandatory.
  • Datastore latency is a significant factor. For optimal performance, Netreo recommends environments with < 5 millisecond write latency and < 10 millisecond total latency. NFS-based datastores are not supported.

Due to constant database updates during data collection, most environments will have a 90%-write, 10%-read disk I/O profile—which may require special configuration of your datastore for optimal performance. OmniCenter is extremely disk write intensive and is not tolerant of high latency or unstable storage environments. If you have concerns about your storage performance or stability, dedicated hardware is recommended.

Scalability Guidelines

The following recommendations are based on Netreo best practices and an average customer deployment. Customers with specialized environments that require large amounts of monitoring, rapid monitoring intervals, large numbers of simultaneous users, or large volumes of IPTel call data or traffic flow data will require more resources.

These recommendations are also based on hundreds of hours of testing and real-world experience with diverse customer environments across many industries. While provisioning fewer resources may work adequately for some customers, our experience has shown that under-provisioning can lead to unsatisfactory performance and reduced reliability. The advantage of a VM environment is that even if over-provisioned for a specific environment, virtual hypervisors will happily share any unused resources (including disk, which we recommend be thin-provisioned so it is only used if and when needed) with other VMs. We therefore do not recommend provisioning less than the prescribed amount of CPUs, memory and disk space.

Number of Devices Recommended Configuration
Less than 100 4 vCPU16
8 GB RAM17
100 GB Disk space18
Up to 500 8 vCPU19
8 GB RAM20
100 GB Disk space21
Up to 1,000 16 vCPU22
32 GB RAM23
200 GB Disk space24 (SSD strongly recommended)
Up to 3,000 32 vCPU25
64 GB RAM26
600 GB Disk space27 (SSD required)
Use of a service engine is recommended for networks with more than 1,000 devices.
Up to 5,000 64 vCPU28
128 GB RAM29
1 TB Disk space30 (SSD required)
More than 5,000 Please contact Netreo sales for a detailed discussion of your design requirements.

The above specifications are general guidelines only. Traffic flow can have a drastic effect on performance depending on flow volume. If more than 10% of your network devices will be sending traffic flow information, please contact Netreo for performance scaling options.

Updated on November 25, 2019

Was this article helpful?

Need Support?
Can’t find the answer you’re looking for? Don’t worry we’re here to help!
Contact Support

Leave a Reply