Compute optimized instances
Compute optimized instances are ideal for compute-bound applications that benefit from high-performance processors.
C5 and C5n instances
These instances are well suited for the following:
-
Batch processing workloads
-
Media transcoding
-
High-performance web servers
-
High-performance computing (HPC)
-
Scientific modeling
-
Dedicated gaming servers and ad serving engines
-
Machine learning inference and other compute-intensive applications
Bare metal instances, such as c5.metal, provide your applications
with direct access to physical resources of the host server, such as processors and
memory.
For more information, see Amazon EC2 C5 Instances
Contents
Hardware specifications
The following is a summary of the hardware specifications for compute optimized instances.
| Instance type | Default vCPUs | Memory (GiB) |
|---|---|---|
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 |
c5.12xlarge |
48 | 96 |
c5.18xlarge |
72 | 144 |
c5.24xlarge |
96 | 192 |
c5.metal |
96 | 192 |
c5a.large |
2 | 4 |
c5a.xlarge |
4 | 8 |
c5a.2xlarge |
8 | 16 |
c5a.4xlarge |
16 | 32 |
c5a.8xlarge |
32 | 64 |
c5a.12xlarge |
48 | 96 |
c5a.16xlarge |
64 | 128 |
c5a.24xlarge |
96 | 192 |
c5ad.large |
2 | 4 |
c5ad.xlarge |
4 | 8 |
c5ad.2xlarge |
8 | 16 |
c5ad.4xlarge |
16 | 32 |
c5ad.8xlarge |
32 | 64 |
c5ad.12xlarge |
48 | 96 |
c5ad.16xlarge |
64 | 128 |
c5ad.24xlarge |
96 | 192 |
c5d.large |
2 | 4 |
c5d.xlarge |
4 | 8 |
c5d.2xlarge |
8 | 16 |
c5d.4xlarge |
16 | 32 |
c5d.9xlarge |
36 | 72 |
c5d.12xlarge |
48 | 96 |
c5d.18xlarge |
72 | 144 |
c5d.24xlarge |
96 | 192 |
c5d.metal |
96 | 192 |
c5n.large |
2 | 5.25 |
c5n.xlarge |
4 | 10.5 |
c5n.2xlarge |
8 | 21 |
c5n.4xlarge |
16 | 42 |
c5n.9xlarge |
36 | 96 |
c5n.18xlarge |
72 | 192 |
c5n.metal |
72 | 192 |
For more information about the hardware specifications for each Amazon EC2 instance
type,
see Amazon EC2 Instance Types
For more information about specifying CPU options, see Optimize CPU options.
Instance performance
EBS-optimized instances enable you to get consistently high performance for your EBS volumes by eliminating contention between Amazon EBS I/O and other network traffic from your instance. Some compute optimized instances are EBS-optimized by default at no additional cost. For more information, see Amazon EBS–optimized instances.
Network performance
You can enable enhanced networking on supported instance types to provide lower latencies, lower network jitter, and higher packet-per-second (PPS) performance. Most applications do not consistently need a high level of network performance, but can benefit from access to increased bandwidth when they send or receive data. For more information, see Enhanced networking on Windows.
The following is a summary of network performance for compute optimized instances that support enhanced networking.
| Instance type | Network performance | Enhanced networking |
|---|---|---|
c4.large |
Moderate | Intel 82599 VF |
c4.xlarge | c4.2xlarge | c4.4xlarge |
High | Intel 82599 VF |
c5.4xlarge and smaller |
c5d.4xlarge and smaller
|
Up to 10 Gbps † | ENA |
c4.8xlarge |
10 Gbps | Intel 82599 VF |
c5.9xlarge | c5d.9xlarge |
10 Gbps | ENA |
c5.12xlarge | c5d.12xlarge
|
12 Gbps | ENA |
c5n.4xlarge and smaller
|
Up to 25 Gbps † | ENA |
c5.18xlarge | c5.24xlarge | c5.metal |
c5d.18xlarge | c5d.24xlarge | c5d.metal
|
25 Gbps | ENA |
c5n.9xlarge |
50 Gbps | ENA |
c5n.18xlarge | c5n.metal
|
100 Gbps | ENA |
† These instances use a network I/O credit mechanism to allocate network bandwidth to instances based on average bandwidth utilization. They accrue credits when their bandwidth is below their baseline bandwidth, and can use these credits when they perform network data transfers. For more information, open a support case and ask about baseline bandwidth for the specific instance types that you are interested in.
SSD I/O performance
If you use all the SSD-based instance store volumes available to your instance, you get the IOPS (4,096 byte block size) performance listed in the following table (at queue depth saturation). Otherwise, you get lower IOPS performance.
| Instance Size | 100% Random Read IOPS | Write IOPS |
|---|---|---|
c5ad.large |
16,283 | 7,105 |
c5ad.xlarge |
32,566 | 14,211 |
c5ad.2xlarge |
65,132 | 28,421 |
c5ad.4xlarge |
130,263 | 56,842 |
c5ad.8xlarge |
260,526 | 113,684 |
c5ad.12xlarge |
412,500 | 180,000 |
c5ad.16xlarge |
521,053 | 227,368 |
c5ad.24xlarge |
825,000 | 360,000 |
c5d.large *
|
20,000 | 9,000 |
c5d.xlarge *
|
40,000 | 18,000 |
c5d.2xlarge *
|
80,000 | 37,000 |
c5d.4xlarge *
|
175,000 | 75,000 |
c5d.9xlarge |
350,000 | 170,000 |
c5d.12xlarge |
700,000 | 340,000 |
c5d.18xlarge |
700,000 | 340,000 |
c5d.24xlarge |
1,400,000 | 680,000 |
c5d.metal |
1,400,000 | 680,000 |
* For these instances, you can get up to the specified performance.
As you fill the SSD-based instance store volumes for your instance, the number of write IOPS that you can achieve decreases. This is due to the extra work the SSD controller must do to find available space, rewrite existing data, and erase unused space so that it can be rewritten. This process of garbage collection results in internal write amplification to the SSD, expressed as the ratio of SSD write operations to user write operations. This decrease in performance is even larger if the write operations are not in multiples of 4,096 bytes or not aligned to a 4,096-byte boundary. If you write a smaller amount of bytes or bytes that are not aligned, the SSD controller must read the surrounding data and store the result in a new location. This pattern results in significantly increased write amplification, increased latency, and dramatically reduced I/O performance.
SSD controllers can use several strategies to reduce the impact of write amplification. One such strategy is to reserve space in the SSD instance storage so that the controller can more efficiently manage the space available for write operations. This is called over-provisioning. The SSD-based instance store volumes provided to an instance don't have any space reserved for over-provisioning. To reduce write amplification, we recommend that you leave 10% of the volume unpartitioned so that the SSD controller can use it for over-provisioning. This decreases the storage that you can use, but increases performance even if the disk is close to full capacity.
For instance store volumes that support TRIM, you can use the TRIM command to notify the SSD controller whenever you no longer need data that you've written. This provides the controller with more free space, which can reduce write amplification and increase performance. For more information, see Instance store volume TRIM support.
Instance features
The following is a summary of features for compute optimized instances:
| EBS only | NVMe EBS | Instance store | Placement group | |
|---|---|---|---|---|
|
C4 |
Yes |
No |
No |
Yes |
|
C5 |
Yes |
Yes |
No |
Yes |
|
C5a |
Yes |
Yes |
No |
Yes |
|
C5ad |
No |
Yes |
NVMe * |
Yes |
|
C5d |
No |
Yes |
NVMe * |
Yes |
|
C5n |
Yes |
Yes |
No |
Yes |
* The root device volume must be an Amazon EBS volume.
For more information, see the following:
Release notes
-
C5 and C5d instances feature a 3.1 GHz Intel Xeon Platinum 8000 series processor from either the first generation (Skylake-SP) or second generation (Cascade Lake).
-
C5a and C5ad instances feature a second-generation AMD EPYC processor (Rome) running at frequencies as high as 3.3. GHz.
-
C4 instances and instances built on the Nitro System require 64-bit EBS-backed HVM AMIs. They have high-memory and require a 64-bit operating system to take advantage of that capacity. HVM AMIs provide superior performance in comparison to paravirtual (PV) AMIs on high-memory instance types. In addition, you must use an HVM AMI to take advantage of enhanced networking.
-
Instances built on the Nitro System have the following requirements:
-
NVMe drivers must be installed
-
Elastic Network Adapter (ENA) drivers must be installed
The current AWS Windows AMIs meet these requirements.
-
-
Instances built on the Nitro System instances support a maximum of 28 attachments, including network interfaces, EBS volumes, and NVMe instance store volumes. For more information, see Nitro System volume limits.
-
Launching a bare metal instance boots the underlying server, which includes verifying all hardware and firmware components. This means that it can take 20 minutes from the time the instance enters the running state until it becomes available over the network.
-
To attach or detach EBS volumes or secondary network interfaces from a bare metal instance requires PCIe native hotplug support.
-
Bare metal instances use a PCI-based serial device rather than an I/O port-based serial device. The upstream Linux kernel and the latest Amazon Linux AMIs support this device. Bare metal instances also provide an ACPI SPCR table to enable the system to automatically use the PCI-based serial device. The latest Windows AMIs automatically use the PCI-based serial device.
-
There is a limit on the total number of instances that you can launch in a Region, and there are additional limits on some instance types. For more information, see How many instances can I run in Amazon EC2?
in the Amazon EC2 FAQ.
