Snowpark Container Services: Azure Instance Families

Instance families are grouped as current generation or previous generation. Azure instance families are available in three types:

  • General Compute (GEN): Best price-performance for general-purpose containerized workloads.

  • High Memory (MEM): High memory-to-vCPU ratio for applications that require large amounts of RAM, such as CPU-based model serving, large-scale in-memory data processing, and vector index serving.

  • GPU Accelerated (GPU): For machine learning training, inference, and AI workloads requiring GPU acceleration.

For pricing information, see the Snowflake Service Consumption Table.

Note

Region availability is subject to change. To retrieve current availability and instance family specifications programmatically, use SHOW COMPUTE POOL INSTANCE FAMILIES.

Current Generation Instance Families

General Compute Instance Families (Current Generation)

Current generation x86 instances offering the best price-performance for general-purpose workloads.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

Node limit

Region Availability

GEN_X64_G2_2

1

6

100

12.5

150

Available everywhere

GEN_X64_G2_4

3

13

100

12.5

150

Available everywhere

GEN_X64_G2_8

6

28

100

12.5

150

Available everywhere

GEN_X64_G2_16

14

58

100

12.5

150

Available everywhere

GEN_X64_G2_32

28

116

100

16.0

150

Available everywhere

High Memory Instance Families (Current Generation)

Current generation x86 instances optimized for memory-intensive workloads.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

Node limit

Region Availability

MEM_X64_G2_8

6

58

100

12.5

150

Available everywhere

MEM_X64_G2_32

28

240

100

12.5

150

Available everywhere

MEM_X64_G2_64

60

492

100

16.0

150

Available everywhere

MEM_X64_G2_96

92

652

100

16.0

150

Available everywhere

GPU Accelerated Instance Families (Current Generation)

Azure GPU instance families feature three NVIDIA GPU architectures, each suited to different AI and ML workloads.

NVIDIA T4

Turing GPU for cost-effective inference and light ML workloads.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

GPU

GPU Memory per GPU (GB)

Node limit

Region Availability

GPU_NV_XS

3

26

100

8.0

1 NVIDIA T4

16

10

Not available in Switzerland North, UAE North, Central US, and UK South regions

NVIDIA A10

Mid-range Ampere GPU for ML model development and inference on small to medium models.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

GPU

GPU Memory per GPU (GB)

Node limit

Region Availability

GPU_NV_SM

32

424

100

40.0

1 NVIDIA A10

24

10

Not available in Central US

GPU_NV_2M

68

858

100

80.0

2 NVIDIA A10

48

5

Not available in Central US

NVIDIA A100

High-throughput Ampere GPU for large-scale model training and large dataset processing.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

GPU

GPU Memory per GPU (GB)

Node limit

Region Availability

GPU_NV_3M

44

424

100

40.0

2 NVIDIA A100

160

On Request

Not available in Central US, North Europe, and UAE North

GPU_NV_SL

92

858

100

80.0

4 NVIDIA A100

320

On Request

Not available in Central US, North Europe, and UAE North

Previous Generation Instance Families

General Compute Instance Families (Previous Generation)

Previous generation x86 instances. For new workloads, use GEN_X64_G2 instances instead.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

Node limit

Region Availability

CPU_X64_XS

1

6

100

12.5

500

Available everywhere

CPU_X64_S

3

13

100

12.5

500

Available everywhere

CPU_X64_M

6

28

100

12.5

500

Available everywhere

CPU_X64_SL

14

58

100

12.5

500

Available everywhere

CPU_X64_L

28

116

100

16.0

500

Available everywhere

High Memory Instance Families (Previous Generation)

Previous generation x86 instances optimized for memory. For new workloads, use MEM_X64_G2 instances instead.

Instance Family

vCPU

Memory (GiB)

Storage (GB)

Bandwidth limit (Gbps)

Node limit

Region Availability

HIGHMEM_X64_S

6

58

100

8.0

150

Available everywhere

HIGHMEM_X64_M

28

240

100

16.0

150

Available everywhere

HIGHMEM_X64_SL

60

492

100

32.0

150

Available everywhere