Instance Types

Instance types determine the hardware characteristics of your instances.

You must specify an instance type when you launch an instance, but you can also modify the instance type later on. For more information, see Creating / Launching Instances and Modifying an Instance Attribute.

Characteristics

An instance type determines the hardware characteristics of your instance, including:

  • Compute characteristics: CPU generation, number of vCPUs, amount of memory, and performance flag.

  • Network characteristics: maximum number of FNIs you can attach, and maximum number of private IPs you can assign per FNI. For more information, see About FNIs.

The CPU generation is a value that corresponds to a family of Intel CPUs:

  • v2: Intel Ivy Bridge

  • v3: Intel Haswell

  • v4: Intel Broadwell

  • v5: Intel Skylake

For more information about these CPU families, see the official Intel documentation.

The performance flag is a value that determines the overall performance of the vCPUs:

  • 1 (highest): Maximum compute capacity on all the vCPUs of the instance. This value is ideal for intensive compute applications.

  • 2 (high): High compute capacity on all the vCPUs of the instance. This value offers small performance variability and applies to most cases.

  • 3 (medium): Performance with significant variability on all the vCPUs of the instance. This option does not guarantee consistent performance, but is ideal for machines with low CPU demand.

Available Instance Types

TINA Type

The TINA type enables you to fully customize the compute characteristics of your instance.

When you specify a TINA type for an instance, you must follow the tinavW.cXrYpZ format, where W is the CPU generation, X is the number of vCPUs, Y is the amount of memory, and Z is the performance flag.

Specifying the CPU generation, number of vCPUs and amount of memory of the instance is mandatory. Only the performance flag is optional.

Name CPU Generation vCPUs Memory (in GiB) Performance Flag Max. Number of FNIs (Including Primary Network Interface) Max. Number of Private IPs Per FNI (Including Primary IP)

tinavW.cXrYpZ

A value between 2 and 5

A value between 1 and 78

A value between 1 and 1039

A value between 1 and 3

8

10

AWS Types

AWS types are predefined instance types. They include:

  • Instance types named after those of AWS. They are provided for compatibility purposes.

  • Instance types unique to 3DS OUTSCALE, but named according to a similar naming pattern as AWS. In the table below, their family is marked with an asterisk (*).

Each AWS type has a corresponding TINA type in terms of compute characteristics, but they differ in terms of FNI characteristics.

The following table presents the available AWS types. You can also list them by API using the ReadVmTypes or DescribeInstanceTypes methods.

Name Family Corresponding TINA Type CPU Generation vCPUs Memory (in GiB) Performance Flag Max. Number of FNIs (Including Primary Network Interface) Max. Number of Private IPs per FNI (Including Primary IP) Notes

c1.medium

Compute optimized

tinav1.c2r2p1

v1

2

2

1

2

6

c1.xlarge

Compute optimized

tinav1.c8r7p1

v1

8

7

1

4

15

c3.2xlarge

Compute optimized

tinav3.c8r15p1

v3

8

15

1

4

15

c3.4xlarge

Compute optimized

tinav3.c16r31p1

v3

16

31

1

8

30

c3.8xlarge

Compute optimized

tinav3.c32r61p1

v3

32

61

1

8

30

c3.large

Compute optimized

tinav3.c2r4p1

v3

2

4

1

3

10

c3.xlarge

Compute optimized

tinav3.c4r8p1

v3

4

8

1

4

15

c4.2xlarge

Compute optimized

tinav4.c8r15p1

v4

8

15

1

4

15

c4.4xlarge

Compute optimized

tinav4.c16r31p1

v4

16

31

1

8

30

c4.8xlarge

Compute optimized

tinav4.c36r61p1

v4

36

61

1

8

30

c4.large

Compute optimized

tinav4.c2r4p1

v4

2

4

1

3

10

c4.xlarge

Compute optimized

tinav4.c4r8p1

v4

4

8

1

4

15

c5.18xlarge

Compute optimized

tinav5.c72r147p1

v5

72

147

1

8

30

c5.2xlarge

Compute optimized

tinav5.c8r16p1

v5

8

16

1

4

15

c5.4xlarge

Compute optimized

tinav5.c16r33p1

v5

16

33

1

8

30

c5.8xlarge

Compute optimized

tinav5.c36r74p1

v5

36

74

1

8

30

c5.9xlarge

Compute optimized

tinav5.c36r72p1

v5

36

72

1

8

30

c5.large

Compute optimized

tinav5.c2r4p1

v5

2

4

1

3

10

c5.xlarge

Compute optimized

tinav5.c4r8p1

v5

4

8

1

4

15

cc1.4xlarge

Compute optimized

tinav1.c6r25p1

v1

6

25

1

8

30

cc2.8xlarge

Compute optimized

tinav2.c32r61p1

v2

32

61

1

8

30

cr1.8xlarge

Memory optimized

tinav1.c32r250p2

v1

32

250

2

8

30

g2.2xlarge

Accelerated computing

tinav2.c8r15p1

v2

8

15

1

4

15

Provides 1 dedicated GPU (nvidia-k2)

g2.8xlarge

Accelerated computing

tinav2.c32r60p1

v2

32

60

1

4

15

Provides 4 dedicated GPUs (nvidia-k2)

g3.16xlarge

Accelerated computing

tinav3.c64r500p1

v3

64

500

1

15

50

Provides 4 dedicated GPUs (nvidia-k2)

g3.4xlarge

Accelerated computing

tinav3.c16r125p1

v3

16

125

1

8

30

Provides 1 dedicated GPU (nvidia-k2)

g3.8xlarge

Accelerated computing

tinav3.c32r250p1

v3

32

250

1

8

30

Provides 2 dedicated GPUs (nvidia-k2)

hi1.xlarge

Hi I/O

tinav3.c4.r430p1

v3

4

430

1

4

15

i2.2xlarge

Storage optimized

tinav4.c8r62p2

v4

8

62

2

4

15

i2.4xlarge

Storage optimized

tinav4.c16r125p2

v4

16

125

2

8

30

i2.8xlarge

Storage optimized

tinav4.c32r250p2

v4

32

250

2

8

30

i2.xlarge

Storage optimized

tinav4.c4r30p2

v4

4

30

2

4

15

io5.12xlarge

Big data usage*

tinav4.c36r492p2

v4

36

492

2

8

30

io5.18xlarge

Big data usage*

tinav4.c54r737p2

v4

54

737

2

8

30

io5.2xlarge

Big data usage*

tinav4.c6r82p2

v4

6

82

2

4

15

io5.4xlarge

Big data usage*

tinav4.c12r164p2

v4

12

164

2

8

30

io5.6xlarge

Big data usage*

tinav4.c18r246p2

v4

18

246

2

8

30

io5.8xlarge

Big data usage*

tinav4.c24r328p2

v4

24

328

2

8

30

m1.large

General purpose

tinav1.c2r8p2

v1

2

8

2

3

10

m1.medium

General purpose

tinav1.c1r4p2

v1

1

4

2

2

6

m1.small

General purpose

tinav1.c1r2p2

v1

1

2

2

2

4

m1.xlarge

General purpose

tinav1.c4r15p2

v1

4

15

2

4

15

m2.2xlarge

Memory optimized

tinav2.c4r35p2

v2

4

35

2

4

30

m2.4xlarge

Memory optimized

tinav2.c8r70p2

v2

8

70

2

8

30

m2.xlarge

Memory optimized

tinav2.c2r18p2

v2

2

18

2

4

15

m3.2xlarge

General purpose

tinav3.c8r31p2

v3

8

31

2

4

30

m3.large

General purpose

tinav3.c2r8p2

v3

2

8

2

3

10

m3.medium

General purpose

tinav3.c1r4p2

v3

1

4

2

2

6

m3.xlarge

General purpose

tinav3.c4r15p2

v3

4

15

2

4

15

m4.10xlarge

General purpose

tinav4.c40r164p2

v4

40

164

2

8

30

m4.2xlarge

General purpose

tinav4.c8r33p2

v4

8

33

2

4

15

m4.4xlarge

General purpose

tinav4.c16r66p2

v4

16

66

2

8

30

m4.large

General purpose

tinav4.c2r8p2

v4

2

8

2

3

10

m4.xlarge

General purpose

tinav4.c4r16p2

v4

4

16

2

4

15

m5.12xlarge

General purpose

tinav5.c48r197p2

v5

48

197

2

8

30

m5.24xlarge

General purpose

tinav5.c96r384p2

v5

96

384

2

8

30

m5.2xlarge

General purpose

tinav5.c8r33p2

v5

8

33

2

4

15

m5.4xlarge

General purpose

tinav5.c16r66p2

v5

16

66

2

8

30

m5.large

General purpose

tinav5.c2r8p2

v5

2

8

2

3

10

m5.xlarge

General purpose

tinav5.c4r16p2

v5

4

16

2

4

15

mv3.2xlarge

General purpose*

tinav4.c8r31p2

v4

8

31

2

4

30

Provides 1 dedicated GPU (nvidia-k2)

mv3.large

General purpose*

tinav4.c2r8p2

v4

2

8

2

3

10

Provides 1 dedicated GPU (nvidia-k2)

mv3.xlarge

General purpose*

tinav4.c4r15p2

v4

4

15

2

4

15

Provides 1 dedicated GPU (nvidia-k2)

nv1.large

Remote visualization*

tinav4.c2r8p2

v4

2

8

2

3

10

Provides 1 dedicated GPU (nvidia-k2)

nv1.medium

Remote visualization*

tinav4.c1r4p2

v4

1

4

2

2

6

Provides 1 dedicated GPU (nvidia-k2)

nv1.small

Remote visualization*

tinav4.c1r2p2

v4

1

2

2

2

6

Provides 1 dedicated GPU (nvidia-k2)

nv1.xlarge

Remote visualization*

tinav4.c4r15p2

v4

4

15

2

4

15

Provides 1 dedicated GPU (nvidia-k2)

nv2.2xlarge

Remote visualization*

tinav4.c4r35p2

v4

4

35

2

4

15

Provides 1 dedicated GPU (nvidia-k2)

nv2.4xlarge

Remote visualization*

tinav4.c8r70p2

v4

8

70

2

8

30

Provides 1 dedicated GPU (nvidia-k2)

nv2.xlarge

Remote visualization*

tinav4.c2r18p2

v4

2

18

2

4

15

Provides 1 dedicated GPU (nvidia-k2)

oc1.4xlarge

General purpose*

tinav1.c6r25p1

v1

6

25

1

8

30

oc2.8xlarge

General purpose*

tinav2.c15r66p1

v2

15

66

1

8

30

oc5.12xlarge

Compute optimized*

tinav4.c36r492p1

v4

36

492

1

8

30

oc5.18xlarge

Compute optimized*

tinav4.c54r737p1

v4

54

737

1

8

30

oc5.2xlarge

Compute optimized*

tinav4.c6r82p1

v4

6

82

1

4

15

oc5.4xlarge

Compute optimized*

tinav4.c12r164p1

v4

12

164

1

4

15

oc5.6xlarge

Compute optimized*

tinav4.c18r246p1

v4

18

246

1

8

30

oc5.8xlarge

Compute optimized*

tinav4.c24r328p1

v4

24

328

1

8

30

og3.2xlarge

Remote visualization*

tinav4.c16r61p2

v4

16

61

2

4

15

Provides 1 dedicated GPU (nvidia-k2)

og4.2xlarge

Remote visualization*

tinav4.c16r123p1

v4

16

123

1

4

15

Provides 2 dedicated GPUs (nvidia-k2)

og4.4xlarge

Remote visualization*

tinav4.c24r184p1

v4

24

184

1

8

30

Provides 3 dedicated GPUs (nvidia-k2)

og4.8xlarge

Remote visualization*

tinav4.c32r246p1

v4

32

246

1

8

30

Provides 4 dedicated GPUs (nvidia-k2)

og4.xlarge

Remote visualization*

tinav4.c8r61p1

v4

8

61

1

4

15

Provides 1 dedicated GPU (nvidia-k2)

om5.12xlarge

Memory optimized*

tinav4.c36r983p2

v4

36

983

2

8

30

om5.18xlarge

Memory optimized*

tinav4.c54r1475p2

v4

54

1475

2

8

30

om5.2xlarge

Memory optimized*

tinav4.c6r164p2

v4

6

164

2

4

15

om5.4xlarge

Memory optimized*

tinav4.c12r328p2

v4

12

328

2

4

15

om5.6xlarge

Memory optimized*

tinav4.c18r492p2

v4

18

492

2

8

30

om5.8xlarge

Memory optimized*

tinav4.c24r655p2

v4

24

655

2

8

30

os1.2xlarge

General purpose*

tinav4.c4r66p2

v1

4

66

2

8

30

os1.4xlarge

General purpose*

tinav1.c8r131p2

v1

8

131

2

8

30

os1.8xlarge

General purpose*

tinav1.c15r250p2

v1

15

250

2

8

30

os3.2xlarge

General purpose*

tinav3.c16r66p2

v3

16

66

2

8

30

p100.2xlarge

GPU-based*

tinav5.c8r61p1

v5

8

61

1

4

15

Provides 1 dedicated GPU (nvidia-p100)

p100.4xlarge

GPU-based*

tinav5.c16r122p1

v5

16

122

1

4

15

Provides 1 dedicated GPU (nvidia-p100)

p100.8xlarge

GPU-based*

tinav5.c32r244p1

v5

32

244

1

4

15

Provides 2 dedicated GPUs (nvidia-p100)

p3.2xlarge

GPU-based*

tinav5.c8r61p1

v5

8

61

1

4

15

Provides 1 dedicated GPU (nvidia-p100)

p6.8xlarge

GPU-based*

tinav5.c32r244p1

v5

32

244

1

4

15

Provides 1 dedicated GPU (nvidia-p6)

r3.2xlarge

Memory optimized

tinav3.c8r62p2

v3

8

62

2

4

15

r3.4xlarge

Memory optimized

tinav3.c16r125p2

v3

16

125

2

8

30

r3.8xlarge

Memory optimized

tinav3.c32r250p2

v3

32

250

2

8

30

r3.large

Memory optimized

tinav3.c2r15p2

v3

2

15

2

3

10

r3.xlarge

Memory optimized

tinav3.c4r31p2

v3

4

31

2

4

15

r4.16xlarge

Memory optimized

tinav4.c64r500p2

v4

64

500

2

8

30

r4.2xlarge

Memory optimized

tinav4.c8r62p2

v4

8

62

2

4

15

r4.4xlarge

Memory optimized

tinav4.c16r125p2

v4

16

125

2

8

30

r4.8xlarge

Memory optimized

tinav4.c32r250p2

v4

32

250

2

8

30

r4.large

Memory optimized

tinav4.c2r16p2

v4

2

16

2

3

10

r4.xlarge

Memory optimized

tinav4.c4r31p2

v4

4

31

2

4

15

t1.micro

Micro

tinav1.c1r1p3

v1

1

1

3

2

2

t2.large

General purpose

tinav2.c2r8p3

v2

2

8

3

3

6

t2.medium

General purpose

tinav2.c2r4p3

v2

2

4

3

3

6

t2.micro

General purpose

tinav2.c1r1p3

v2

1

1

3

2

2

t2.nano

General purpose

tinav2.c1r1p3

v2

1

1

3

2

2

t2.small

General purpose

tinav2.c1r2p3

v2

1

2

3

2

4

Related Pages

AWS™ and Amazon Web Services™ are trademarks of Amazon Technologies, Inc or its affiliates in the United States and/or other countries.