Instance Types
DNAnexus supports several instance types (virtual computer configurations on which jobs can be run) on both AWS and Azure clouds.
The names of DNAnexus instance types are generally derived via the following scheme:
Cloud provider prefix | | Memory infix | | Storage infix | | Version infix | | Core suffix |
∅ (AWS) azure: (Azure) | + | mem1_ (<=2GB/core) mem2_ (~4GB/core) mem3_ (>=7GB/core) mem4_ (~14GB/core) mem5_ (~28GB/core) | + | ssd1_ (<=20GB/core) ssd2_ (~32GB-128GB/core) ssd3_ (>600GB/core) hdd2_ (>100GB/core) | + | ∅ (version1) v2_ (version2) | + | x1 x2 x4 x8 x16 x20 x32 x36 x48 x64 x96 x128 |
Not all name combinations are available as instance types. Please, see the full lists of instance types below.
Cloud provider prefix: Denotes the cloud provider.
Memory infix: Denotes the memory capacity (per core).
Storage infix: Denotes the local storage technology and capacity (per core). ssd represents a solid-state drive, whereas hdd represents a regular hard disk drive.
Version infix (optional): Denotes the version of the instance type.
Core suffix: Denotes the number of cores.
For example, the mem1_ssd1_v2_x8 is an instance on the AWS cloud provider at its 2nd version in mem1_ssd1 class, has 8 cores, 16GB of memory (2GB/core) and 160GB of solid-state drive storage (20GB/core). Similarly, the azure:mem1_ssd1_x8 is an instance on the Azure cloud provider, has 8 cores, 15.7GB of memory (~1.9GB/core) and 128GB of solid-state drive storage (16GB/core).
For each combination of memory and storage from the table above, not all core suffixes are available. The following table summarizes the available instance types.
Each row represents an instance type prefix (cloud provider, memory, storage, and version), and each column represents a core suffix (number of cores).
The value in each cell is the amount of total memory in GB.
Current Version
cloud provider | instance class | version | _x1 | _x2 | _x4 | _x8 | _x12 | _x16 | _x20 | _x24 | _x32 | _x36 | _x48 | _x64 | _x72 | _x96 | _x128 | |
AWS | mem1_ssd1 | v2 | | 4 | 8 | 16 | | 32 | | | | 72 | | | 144 | | | |
AWS | mem1_ssd2 | v2 | | 4 | 8 | 16 | | 32 | | | | 72 | | | 144 | | | |
AWS | mem1_hdd2 | | 1.7 | | | 7 | | | | | 60.5 | | | | | | | |
AWS | mem2_ssd1 | v2 | | 8 | 16 | 32 | | 64 | | | 128 | | 192 | 256 | | 384 | | |
AWS | mem2_hdd2 | v2 | | 8 | 16 | | | | | | | | | | | | | |
AWS | mem3_ssd1 | v2 | | 16 | 32 | 64 | | 128 | | | 256 | | 384 | 512 | | 768 | | |
AWS | mem3_ssd2 | v2 | | 15.25 | 30.5 | 61 | | 122 | | | 244 | | | 488 | | | | |
AWS | mem3_ssd3 | | | 16 | 32 | 64 | 96 | | | 192 | | | 384 | | | 768 | | |
AWS | mem3_hdd2 | v2 | | 16 | 32 | 64 | | | | | | | | | | | | |
AWS | mem4_ssd1 | | | | | | | | | | | | | | | | 1,952 | |
Azure | azure:mem1_ssd1 | | | 3.9 | 7.8 | 15.7 | | 31.4 | | | | | | | | | | |
Azure | azure:mem2_ssd1 | | | 3.5 | 7 | 14 | | 28 | | | 56 | | | | | | | |
Azure | azure:mem3_ssd1 | | | 14 | 29 | 56 | | 112 | 140 | | | | | | | | | |
Azure | azure:mem4_ssd1 | | | 28 | 56 | 112 | | 224 | | | 448 | | | | | | | |
Azure | azure:mem5_ssd2 | | | | | | | | | | | | | 1,792 | | | 3,892 | |
Previous versions
cloud provider | instance class | version | _x1 | _x2 | _x4 | _x8 | _x12 | _x16 | _x20 | _x24 | _x32 | _x36 | _x48 | _x64 | _x72 | _x96 | _x128 |
AWS | mem1_ssd1 | | | 3.8 | 7.5 | 15 | | 30 | | | 60 | | | | | | |
AWS | mem1_ssd2 | | | 3.8 | 7.5 | 15 | | 30 | | | 60 | | | | | | |
AWS | mem2_ssd1 | | | 7.5 | 15 | 30 | | | | | | | | | | | |
AWS | mem2_hdd2 | | 3.8 | 7.5 | 15 | | | | | | | | | | | | |
AWS | mem3_ssd1 | | | 15 | 30.5 | 61 | | 122 | | | 244 | | | | | | |
AWS | mem3_ssd2 | | | | 30.5 | 61 | | 122 | | | 244 | | | | | | |
AWS | mem3_ssd3 | | | 16 | 32 | 64 | 96 | | | 192 | | | 384 | | | 768 | |
AWS | mem3_hdd2 | | | 17.1 | 34.2 | 68.4 | | | | | | | | | | | |
NOTE:
The amount of memory actually available for jobs is slightly less, as DNAnexus may use a small amount for operational purposes.
Not all instance types are accessible to all customers.
DNAnexus adds new instance types when AWS or Azure offers new generations of instances with better hardware. The new instance types will have version infix in their names, e.g. mem1_ssd1_v2_x8. The new instance types usually have better CPU, higher memory, and/or higher local storage. Within an instance class (such as mem1_ssd1) you can access both old instance types (such as mem1_ssd1_x8) and new ones (such as mem1_ssd1_v2_x8). We recommend that you always use the latest version available, for better performance.
Supported Ubuntu Linux releases are 20.04 and 16.04. All releases are supported on all instance types in all regions.
The following table shows all the instance types available on AWS.
Instance Type | Cores | Memory (GB) | Storage (GB) | |
| | | | |
mem1_ssd1_x2 | 2 | 3.8 | 40 | |
mem1_ssd1_x4 | 4 | 7.5 | 80 | |
mem1_ssd1_x8 | 8 | 15 | 160 | |
mem1_ssd1_x16 | 16 | 30 | 320 | |
mem1_ssd1_x32 | 32 | 60 | 640 | |
mem1_ssd1_x36 | 36 | 72 | 900 | |
mem1_ssd1_v2_x2 | 2 | 4 | 50 | |
mem1_ssd1_v2_x4 | 4 | 8 | 100 | |
mem1_ssd1_v2_x8 | 8 | 16 | 200 | |
mem1_ssd1_v2_x16 | 16 | 32 | 400 | |
mem1_ssd1_v2_x36 | 36 | 72 | 900 | |
mem1_ssd1_v2_x72 | 72 | 144 | 1,800 | |
| | | | |
mem1_ssd2_x2 | 2 | 3.8 | 160 | |
mem1_ssd2_x4 | 4 | 7.5 | 320 | |
mem1_ssd2_x8 | 8 | 15 | 640 | |
mem1_ssd2_x16 | 16 | 30 | 1,280 | |
mem1_ssd2_x36 | 36 | 60 | 2,880 | |
mem1_ssd2_v2_x2 | 2 | 4 | 160 | |
mem1_ssd2_v2_x4 | 4 | 8 | 320 | |
mem1_ssd2_v2_x8 | 8 | 16 | 640 | |
mem1_ssd2_v2_x16 | 16 | 32 | 1,280 | |
mem1_ssd2_v2_x36 | 36 | 72 | 2,880 | |
mem1_ssd2_v2_x72 | 72 | 144 | 5,760 | |
| | | | |
mem1_hdd2_x8 | 8 | 7 | 1,680 | |
mem1_hdd2_x32 | 32 | 60.5 | 3,360 | |
| | | | |
mem2_ssd1_x2 | 2 | 7.5 | 40 | |
mem2_ssd1_x4 | 4 | 15 | 80 | |
mem2_ssd1_x8 | 8 | 30 | 160 | |
mem2_ssd1_v2_x2 | 2 | 8 | 75 | |
mem2_ssd1_v2_x4 | 4 | 16 | 150 | |
mem2_ssd1_v2_x8 | 8 | 32 | 300 | |
mem2_ssd1_v2_x16 | 16 | 64 | 600 | |
mem2_ssd1_v2_x32 | 32 | 128 | 1,200 | |
mem2_ssd1_v2_x48 | 48 | 144 | 1,800 | |
mem2_ssd1_v2_x64 | 64 | 256 | 2,400 | |
mem2_ssd1_v2_x96 | 96 | 384 | 3,600 | |
| | | | |
mem2_hdd2_x1 | 1 | 3.8 | 410 | |
mem2_hdd2_x2 | 2 | 7.5 | 840 | |
mem2_hdd2_x4 | 4 | 15 | 1,680 | |
mem2_hdd2_v2_x2 | 2 | 8 | 1,000 | |
mem2_hdd2_v2_x4 | 4 | 16 | 2,000 | |
| | | | |
mem3_ssd1_x2 | 2 | 15 | 40 | |
mem3_ssd1_x4 | 4 | 30.5 | 80 | |
mem3_ssd1_x8 | 8 | 61 | 160 | |
mem3_ssd1_x16 | 16 | 122 | 320 | |
mem3_ssd1_x32 | 32 | 244 | 640 | |
mem3_ssd1_v2_x2 | 2 | 16 | 75 | |
mem3_ssd1_v2_x4 | 4 | 32 | 150 | |
mem3_ssd1_v2_x8 | 8 | 64 | 300 | |
mem3_ssd1_v2_x16 | 16 | 128 | 600 | |
mem3_ssd1_v2_x32 | 32 | 256 | 1,200 | |
mem3_ssd1_v2_x48 | 48 | 384 | 1,800 | |
mem3_ssd1_v2_x64 | 64 | 512 | 3,200 | |
mem3_ssd1_v2_x96 | 96 | 768 | 3,600 | |
| | | | |
mem3_ssd2_x4 | 4 | 30.5 | 800 | |
mem3_ssd2_x8 | 8 | 61 | 1,600 | |
mem3_ssd2_x16 | 16 | 122 | 3,200 | |
mem3_ssd2_x32 | 32 | 244 | 6,400 | |
mem3_ssd2_v2_x2 | 2 | 15.25 | 475 | |
mem3_ssd2_v2_x4 | 4 | 30.5 | 950 | |
mem3_ssd2_v2_x8 | 8 | 61 | 1,900 | |
mem3_ssd2_v2_x16 | 16 | 122 | 3,800 | |
mem3_ssd2_v2_x32 | 32 | 244 | 7,600 | |
mem3_ssd2_v2_x64 | 64 | 488 | 15,200 | |
| | | | |
mem3_ssd3_x2 | 2 | 16 | 1,250 | |
mem3_ssd3_x4 | 4 | 32 | 2,500 | |
mem3_ssd3_x8 | 8 | 64 | 5,000 | |
mem3_ssd3_x12 | 12 | 96 | 7,500 | |
mem3_ssd3_x24 | 24 | 192 | 15,000 | |
mem3_ssd3_x48 | 48 | 384 | 30,000 | |
mem3_ssd3_x96 | 96 | 768 | 60,000 | |
| | | | |
mem3_hdd2_x2 | 2 | 17.1 | 420 | |
mem3_hdd2_x4 | 4 | 34.2 | 850 | |
mem3_hdd2_x8 | 8 | 68.4 | 1,680 | |
mem3_hdd2_v2_x2 | 2 | 16 | 500 | |
mem3_hdd2_v2_x4 | 4 | 32 | 1,000 | |
mem3_hdd2_v2_x8 | 8 | 64 | 2,000 | |
| | | | |
mem4_ssd1_x128 | 128 | 1,952 | 3,840 | |
| | | | |
The following table shows all the instance types available on Azure.
Instance Type | Cores | Memory (GB) | Storage (GB) |
| | | |
azure:mem1_ssd1_x2 | 2 | 3.9 | 32 |
azure:mem1_ssd1_x4 | 4 | 7.8 | 64 |
azure:mem1_ssd1_x8 | 8 | 15.7 | 128 |
azure:mem1_ssd1_x16 | 16 | 31.4 | 254 |
| | | |
azure:mem2_ssd1_x1 | 1 | 3.5 | 128 |
azure:mem2_ssd1_x2 | 2 | 7 | 128 |
azure:mem2_ssd1_x4 | 4 | 14 | 128 |
azure:mem2_ssd1_x8 | 8 | 28 | 256 |
azure:mem2_ssd1_x16 | 16 | 56 | 512 |
| | | |
azure:mem3_ssd1_x2 | 2 | 14 | 128 |
azure:mem3_ssd1_x4 | 4 | 28 | 128 |
azure:mem3_ssd1_x8 | 8 | 56 | 256 |
azure:mem3_ssd1_x16 | 16 | 112 | 512 |
azure:mem3_ssd1_x20 | 20 | 140 | 640 |
| | | |
azure:mem4_ssd1_x2 | 2 | 28 | 128 |
azure:mem4_ssd1_x4 | 4 | 56 | 128 |
azure:mem4_ssd1_x8 | 8 | 112 | 256 |
azure:mem4_ssd1_x16 | 16 | 224 | 512 |
azure:mem4_ssd1_x32 | 32 | 448 | 1024 |
| | | |
azure:mem5_ssd2_x64* | 64 | 1,792 | 8,192 |
azure:mem5_ssd2_x128* | 128 | 3,892 | 16,384 |
NOTE: Some of the resources on a worker instance will be shared with DNAnexus processes that support your job and provide API services. Your job's virtual file system will have approximately 95% of the available "Storage" to use as local scratch space.
* These high memory instance types are expensive, please use with caution.
DNAnexus now supports GPU instance types on AWS cloud to all users.
The following table shows all the GPU instance types available on AWS and Azure.
Instance Type | GPU | Cores | Memory (GB) | Storage (GB) | |
| | | | | |
mem1_ssd1_gpu2_x8 | 2 | 8 | 15 | 60 | |
mem1_ssd1_gpu2_x32 | 4 | 32 | 60 | 240 | |
| | | | | |
mem2_ssd1_gpu_x16 | 1 | 16 | 64 | 225 | |
mem2_ssd1_gpu_x32 | 1 | 32 | 128 | 900 | |
mem2_ssd1_gpu_x48 | 4 | 48 | 192 | 900 | |
mem2_ssd1_gpu_x64 | 1 | 64 | 256 | 900 | |
| | | | | |
mem3_ssd1_gpu_x8 | 1 | 8 | 61 | 160 | |
mem3_ssd1_gpu_x32 * | 4 | 32 | 244 | 640 | |
mem3_ssd1_gpu_x64 * | 8 | 64 | 488 | 1,280 | |
| | | | | |
azure:mem3_ssd2_gpu4_x64 * | 4 | 64 | 488 | 2,048 | |
* These GPU instance types are expensive, please use with caution.
