General Hardware Requirements for a Production System

CPU and RAM

  • CPU can be either an AMD or an Intel CPU.

  • CPU can be single or dual socket.

  • RAM (DDR) must be:

    • 128 GB or more for a production environment, depending on the configuration.

    • A minimum of 32 GB for a demo environment.

The CPU and RAM requirements depend on the number of data drives hosted per node:

Server Density

Physical CPU Cores

Memory

Up to 4LFF

12

128 GB

Up to 16LFF

24

128 GB

Up to 28LFF

32

192 GB

Up to 45LFF

40

256 GB

Up to 60LFF

56

384 GB

Up to 100LFF

64

384 GB

Storage

Operating System

The operating system requires at least 480 GB with one of the followings:

  • One Boot Optimized hardware RAID 1 module of at least 480 GB with two SSD or NVMe drives.

  • Two SATA SSD or NVMe drives in software RAID 1.

ARTESCA Service Drives

Two Read Intensive NVMe drives or Mixed Used NVMe drives, directly attached to CPU NVMe buses, and managed in Software RAID 1.

The minimum size of the service drives depends on:

  • Maximum number of data drives on the node

  • Capacity of the data drives

  • Number of nodes

  • Partner Application

For more information on the minimum size of service drives, refer to the Required Service Disk Capacity section.

Fast storage is required for ARTESCA Data Services, the ARTESCA Storage Services index, and the ARTESCA Kubernetes Platform database.

ARTESCA Data Drives

  • Standard ARTESCA configurations require a minimum of 12 data drives.

  • ARTESCA also comes with a one-node-only configuration with 4 or 8 data drives.

  • All data drives must be identical (disk technology and capacity).

  • For performance, HDDs data drives must be managed by a Hardware RAID controller, with a minimum of 2 GB of battery-backed write cache.

  • SAS (Serial-Attached SCSI) interface for data drives is highly recommended for better performance and stability. SATA (Serial Advanced Technology Attachment) is permitted only for larger drives, where no SAS option exists.

Network

Important

These ranges are reserved respectively for ARTESCA pods and services:

  • 10.233.0.0/16:

  • is the range of IP addresses allocated to the pods network;

  • is the network used by ARTESCA components;

  • has inter nodes traffic;

  • stays only in the ARTESCA cluster.

  • 10.96.0.0/12:

  • is the range of IP addresses allocated to the service network;

  • is the network for service IPs that facilitates communication between different services within a Kubernetes cluster where ARTESCA is deployed.

  • has no inter nodes traffic.

The pods network has to communicate with the service network. These communications are handled by iptables that redirects the 10.96.0.0/12 IPs to the right 10.233.0.0/16 IPs.

They are used for internal communication between the ARTESCA components and should not be used by any other element. Conflicts with your current network configuration will cause system malfunctions.

You can change the ranges of the IP addresses by setting the --pods-cidr and --services-cidr options when generating the configuration file during the installation process.

Each of these two ranges must be a /20 or larger, that is at least 4096 addresses. A smaller range is rejected during the installation.

If those forbidden CIDRs are problematic for you, contact Scality Support.

  • 2 * 1 Gb/s interfaces for the control-plane (used as an LACP bonding for resiliency).

  • 2 * 10 Gb/s interfaces for the workload-plane (used as an LACP bonding for resiliency).

Note

For the Data Network, Scality recommends using 10 Gb/s interfaces. 1 Gb/s interfaces are accepted, but Scality does not guarantee the performance of the system with these interfaces.

Important

For a 3-node cluster, Scality recommends the Data Network and the Management Network to be segregated and exposed through distinct network interfaces. Refer to Network Considerations for more information.

For more information, refer to Partner Applications.

One additional interface is required for the Hardware Management Console such as ILO admin for example.

Non-resilient, single attached, network connections are acceptable for testing environments only.

Management UI High Availability

The ARTESCA Management UI is exposed on the control-plane while the ARTESCA s3 endpoint is exposed on the workload-plane.

For the Management UI, two options are available to support different network deployments:

  1. If all your ARTESCA nodes have Layer 2 access between them (meaning that each machine receives ARP requests from others), then ARTESCA can manage a Virtual IP. The provided IP must not be used by another system. Ask your network administrator for an available IP to use.

  2. If the first option is unavailable, then ARTESCA will use the control-plane IP of the first node. When running on a single-node, this option is selected by default.

    Warning

    If you choose the second option and you lose the node hosting the IP, you can no longer reach the Management UI.

Data Load Balancer

In a multi-node context, you need to configure a load balancer in front of ARTESCA to make the system highly available and to evenly distribute traffic across the system as described in the Configure Data Load Balancer.