When launching an instance, you need to choose an instance type which determines the hardware of its host computer. Each instance type provides different capabilities in terms of compute, memory and storage. Outscale provides predefined instance types grouped in families depending on their capacity, or custom instances for which you can choose the amount of vCores and memory you need. We recommend to test different instance types to better decide which most fits your needs. You can launch an instance and then modify its instance type when the instance is stopped. For more information, see Modifying an Instance Attribute. You can also get information about available predefined instance types using the The following topics are discussed: Custom instances enable you to choose the hardware configuration you need for your instances in terms of vCores and memory, in Gibibytes (GiB). This gives you a high flexibility regarding the capabilities of your instances. With the On-Demand offer, you can choose up to 78 vCores and 720 GiB of memory. Windows instances require at least 2 GiB of memory. When launching an instance using a Windows OMI, ensure that you specify an instance type (custom or predefined) that meets this requirement. Custom instances can be launched from Cockpit, or through the API with any SDK such as AWS CLI, using a tina.cXrY instance type where X corresponds to the number of vCores and Y to the amount of memory, in GiB. You can specify the type of processor you want to use for your custom instances by specifying it in the instance type. You can specify the following processor generations: v3 Intel processors: Use the - If you launch custom instances using the - If the processor generation you specify for your custom instances is not available, the instances are not launched. - For more information about the processor family corresponding to each processor generation, see the Processor Generations and Corresponding Processor Families section below. Custom instances can be either launched in the public Cloud or in a Virtual Private Cloud (VPC). Custom instances do not support ephemeral storage disks and do not have any graphic processing unit (GPU). Predefined instance types have a predefined configuration and are grouped in categories depending on their capacity and their dedicated purpose. Outscale provides both AWS-compatible and Outscale-specific instance types, that can be launched either in the public Cloud or in a VPC. They are configured with a different amount of vCores and memory, a processor family, and possibly a GPU and ephemeral storage disks. Their clock speed, corresponding to the number of instructions that can be processed by the processor in one second, ranges from 2.0 GHz to 2.6 GHz. Windows instances require at least 2 GiB of memory. When launching an instance using a Windows OMI, ensure that you specify an instance type (custom or predefined) that meets this requirement. The number used in the instance type corresponds to the generation of the processor used for the instances. For more information about the processor family corresponding to each processor generation, see the Processor Generations and Corresponding Processor Families section below. The following tables present the characteristics of each instance type in terms of: Ephemeral storage disks Ephemeral storage disks need to be activated before using them. For more information, see Defining Block Device Mappings. Processor family and the corresponding clock speed, in Gigahertz (GHz) For more information about the characteristics of processor families, see Intel documentation. General purpose instances are instance types you can choose for a classic usage. t2.nano Previous Generation Instance Types: m2.xlarge, m2.2xlarge, m2.4xlarge, m1.small, m1.medium, m1.large, m1.xlarge, t1.micro Compute-optimized instances provide high computing performances. Outscale also provide Outscale-specific instance types optimized for computing. For more information, see Outscale-specific Instances > Compute-optimized Instances. Previous Generation Instance Types: c1.medium, c1.xlarge, cc1.4xlarge, cc2.8xlarge, c3.large, c3.xlarge, c3.2xlarge, c3.4xlarge, c3.8xlarge Memory-optimized instances provide high memory performances. Outscale also provide Outscale-specific instance types optimized for memory performances. For more information, see Outscale-specific Instances > Memory-optimized Instances. Storage-optimized instances provide high-capacity ephemeral storage disks and memory. Previous Generation Instance Types: i2.xlarge, i2.2xlarge, i2.4xlarge, i2.8xlarge GPU Commodity instances provide a GPU and can be chosen for general graphic applications. MV3 instance types are general purpose instances that also provide a dedicated NVIDIA K2 GPU in passthrough mode. Previous Generation Instance Types: oc1.4xlarge, oc2.8xlarge, os1.2xlarge, os1.4xlarge, os1.8xlarge, os3.2xlarge Remote visualization instances provide high graphic computing performances thanks to a dedicated NVIDIA K2 GPU in passthrough mode. Previous generation instance types: nv1.small, nv1.medium, nv1.large, nv1.xlarge, nv2.xlarge, nv2.2xlarge, nv2.4xlarge, og3.2xlarge These instances are in Beta version. They are launched using previous generation processors and you may have an insufficient capacity error. These instances are in Beta version. They are launched using previous generation processors and you may have an insufficient capacity error. These instances are in Beta version. They are launched using previous generation processors and you may have an insufficient capacity error. The following table presents the processor family used for each processor generation available for Outscale instances:DescribeInstanceTypes API method. For more information, see DescribeInstanceTypes.Custom Instances
tinav2.cXrY instance type.tinav3.cXrY instance type.tinav4.cXrY instance type.tinav5.cXrY instance type.tina.cXrY instance type, these instances are launched with the last processor generation available for custom instances, if possible. If not, they are launched with the previous processor generation. Predefined Instance Types
AWS-compatible Instances
General Purpose Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU 1 0,5 / / t2.micro 1 1 / / t2.small 1 2 / / t2.medium 2 4 / / t2.large 2 8 / / m3.medium 1 3.75 1 x 4 / m3.large 2 7.5 1 x 32 / m3.xlarge 4 15 2 x 40 / m3.2xlarge 8 30 2 x 80 / m4.large 2 8 / / m4.xlarge 4 16 / / m4.2xlarge 8 32 / / m4.4xlarge 16 64 / / m4.10xlarge 40 160 / / m5.large 2 8 / / m5.xlarge 4 16 / / m5.2xlarge 8 32 / / m5.4xlarge 16 64 / / m5.12xlarge 48 192 / / m5.24xlarge 96 384 / / Compute-optimized Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU c4.large 2 3.75 / / c4.xlarge 4 7.5 / / c4.2xlarge 8 15 / / c4.4xlarge 16 30 / / c4.8xlarge 36 60 / / c5.large 2 4 / / c5.xlarge 4 8 / / c5.2xlarge 8 16 / / c5.4xlarge 16 32 / / c5.9xlarge 36 72 / / c5.18xlarge 72 144 / / Memory-optimized Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU r4.large 2 15.25 / / r4.xlarge 4 30.5 / / r4.2xlarge 8 61 / / r4.4xlarge 16 122 / / r4.8xlarge 32 244 / / r4.16xlarge 64 488 / / r3.large 2 15 1 x 32 / r3.xlarge 4 30.5 1 x 80 / r3.2xlarge 8 61 1 x 160 / r3.4xlarge 16 122 1 x 320 / r3.8xlarge 32 244 2 x 320 / Storage-optimized Instances
GPU Commodity Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU g3.4xlarge 16 122 / 1 g3.8xlarge 32 244 / 2 g3.16xlarge 64 488 / 4 g2.2xlarge 8 15 1 x 60 1 g2.8xlarge 32 60 / 1 Outscale-specific Instances
General Purpose Instances with GPU
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU mv3.large 2 7.5 1 x 32 1 mv3.xlarge 4 15 2 x 40 1 mv3.2xlarge 8 30 2 x 80 1 Remote Visualization Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU og4.xlarge 8 60 1 x 128 1 og4.2xlarge 16 120 1 x 256 2 og4.4xlarge 24 180 1 x 384 3 og4.8xlarge 32 240 1 x 512 4 Memory-optimized Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU om5.2xlarge 6 160 / / om5.4xlarge 12 320 / / om5.6xlarge 18 480 / / om5.8xlarge 24 640 / / om5.12xlarge 36 960 / / om5.18xlarge 54 1440 / / Compute-optimized Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU oc5.2xlarge 6 80 / / oc5.4xlarge 12 160 / / oc5.6xlarge 18 240 / / oc5.8xlarge 24 320 / / oc5.12xlarge 36 480 / / oc5.18xlarge 54 720 / / Big Data Usage Instances
Instance Type vCores Memory (in GiB) Ephemeral Storage Disks (in GiB) GPU io5.2xlarge 6 80 1 x 640 / io5.4xlarge 12 160 2 x 640 / io5.6xlarge 18 240 3 x 640 / io5.8xlarge 24 320 4 x 640 / io5.12xlarge 36 480 6 x 640 / io5.18xlarge 54 720 9 x 640 / Processor Generations and Corresponding Processor Families
Intel Processor Generation Corresponding Processor Family v2 Intel Sandy Bridge v3 Intel Haswell v4 Intel Broadwell v5 Intel Skylake
Windows® is a registered mark of Microsoft Corporation in the United States and/or other countries.
AWS™ and Amazon Web Services™ are trademarks of Amazon Technologies, Inc or its affiliates in the United States and/or other countries.
See Legal Mentions.
