Help Center/ Cloud Container Engine/ User Guide/ Nodes/ Node Specifications

Updated on 2025-09-15 GMT+08:00

View PDF

Node Specifications

This section describes the node flavors supported by CCE and typical node features.

Node flavors vary depending on regions, and some flavors may be unavailable. To obtain available node flavors, log in to the CCE console and switch to the page for creating nodes.

In CCE Turbo clusters, pods need to occupy network interfaces or supplementary network interfaces of nodes. The number of pods you can create on a node depends on how many network interfaces that node has. You need to evaluate the number of network interfaces that can be attached to each node. For details, see Maximum Number of Pods That Can Be Created on a Node.

Node Type	Description	Flavors
ECS (VM)	ECS running on a KVM or QingTian hypervisor. Multiple flavors are offered to meet your compute and storage needs for various applications.	x86: General Computing-plus ECSs General Computing ECSs Memory-optimized ECSs General Computing-Basic ECSs Disk-intensive ECSs Ultra-high I/O ECSs FlexusX Instances Arm (Kunpeng): Kunpeng General Computing-plus ECSs Kunpeng Memory-optimized ECSs Kunpeng Ultra-high I/O ECSs Heterogeneous: GPU-accelerated ECSs AI-accelerated ECSs
ECS (PM)	ECS running on a QingTian-backed bare-metal hypervisor. PM resources and VM resources of such an ECS are in the same resource pool and can be dynamically scheduled.	General Computing-plus ECSs
BMS	BMS offering high-performance computing at low latency for containers.	Instance Family

Node Type

Description

Flavors

ECS (VM)

ECS running on a KVM or QingTian hypervisor. Multiple flavors are offered to meet your compute and storage needs for various applications.

x86:

General Computing-plus ECSs
General Computing ECSs
Memory-optimized ECSs
General Computing-Basic ECSs
Disk-intensive ECSs
Ultra-high I/O ECSs
FlexusX Instances

Arm (Kunpeng):

Kunpeng General Computing-plus ECSs
Kunpeng Memory-optimized ECSs
Kunpeng Ultra-high I/O ECSs

Heterogeneous:

GPU-accelerated ECSs
AI-accelerated ECSs

ECS (PM)

ECS running on a QingTian-backed bare-metal hypervisor. PM resources and VM resources of such an ECS are in the same resource pool and can be dynamically scheduled.

General Computing-plus ECSs

BMS

BMS offering high-performance computing at low latency for containers.

Instance Family

General Computing-plus ECSs

General computing-plus ECSs use dedicated vCPUs and next-generation network acceleration engines to provide powerful compute and network performance.

**Table 1** General computing-plus ECS features
ECS Type	Compute	Network	Supported Cluster Type
C9	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 192 Base/Turbo frequency: 3.1 GHz/3.6 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 40,000,000 Maximum intranet bandwidth: 120 Gbit/s Maximum number of network connections: 10,000,000	CCE standard clusters CCE Turbo clusters
aC8	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 192 Base/Turbo frequency: 2.4 GHz/3.7 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 20,000,000 Maximum intranet bandwidth: 120 Gbit/s Maximum number of network connections: 10,000,000	CCE standard clusters CCE Turbo clusters
aC7	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 232 Base/Turbo frequency: 2.45 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 20,000,000 Maximum intranet bandwidth: 100 Gbit/s	CCE standard clusters CCE Turbo clusters
C7t	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 128 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.4 GHz and 2.8 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 30,000,000 Maximum intranet bandwidth: 90 Gbit/s Maximum number of network connections: 10,000,000	CCE standard clusters CCE Turbo clusters
C7	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 128 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.5 GHz and 2.8 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 12,000,000 Maximum intranet bandwidth: 42 Gbit/s	CCE standard clusters CCE Turbo clusters
C6ne	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 64 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 10,000,000 Maximum intranet bandwidth: 40 Gbit/s	CCE Turbo clusters
C6s	vCPU-to-memory ratio: 1:2 The number of vCPUs: 2 to 64 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 8,500,000 Maximum intranet bandwidth: 30 Gbit/s	CCE standard clusters
C6	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 88 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 12,000,000 Maximum intranet bandwidth: 44 Gbit/s	CCE standard clusters
aC6	vCPU-to-memory ratio: 1:1, 1:2, or 1:4 The number of vCPUs: 8 to 224 Base/Turbo frequency: 2.25 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 8,000,000 Maximum intranet bandwidth: 100 Gbit/s	CCE standard clusters
C3	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 60 Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 5,000,000 Maximum intranet bandwidth: 16 Gbit/s	CCE standard clusters

**Table 2** aC8 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	Max. Network PPS (10,000)	Network Connections (10,000)	Max. NIC Queues	Max. NICs	Max. Supplementary NICs	Max./Assured EVS Disk Bandwidth (Gbit/s)	Max. EVS Disk IOPS (10,000)	Virtualization
ac8.large.2	2	4	20/2	90	75	2	3	8	6/0.3	5	QingTian
ac8.xlarge.2	4	8	20/3	100	150	4	3	16	6/0.5	5
ac8.2xlarge.2	8	16	20/5	160	200	8	4	32	6/1.1	5
ac8.4xlarge.2	16	32	30/8	200	300	16	4	64	6/2.1	5
ac8.8xlarge.2	32	64	40/12	350	400	32	4	128	6/4.3	5
ac8.12xlarge.2	48	96	50/18	500	500	48	4	256	8/6.4	7
ac8.16xlarge.2	64	128	60/25	700	600	64	4	256	11/8.5	10
ac8.24xlarge.2	96	192	80/35	1,000	700	64	8	256	16/13	13
ac8.32xlarge.2	128	256	100/45	1,400	800	64	10	256	21/17	17
ac8.48xlarge.2	192	384	120/65	2,000	1,000	64	12	256	32/26	25
ac8.large.4	2	8	20/2	90	75	2	3	8	6/0.3	5	QingTian
ac8.xlarge.4	4	16	20/3	100	150	4	3	16	6/0.5	5
ac8.2xlarge.4	8	32	20/5	160	200	8	4	32	6/1.1	5
ac8.4xlarge.4	16	64	30/8	200	300	16	4	64	6/2.1	5
ac8.8xlarge.4	32	128	40/12	350	400	32	4	128	6/4.3	5
ac8.12xlarge.4	48	192	50/18	500	500	48	4	256	8/6.4	7
ac8.16xlarge.4	64	256	60/25	700	600	64	4	256	11/8.5	10
ac8.24xlarge.4	96	384	80/35	800	700	64	8	256	16/13	13
ac8.32xlarge.4	128	512	100/45	1,400	800	64	10	256	21/17	17
ac8.48xlarge.4	192	768	120/65	2,000	1,000	64	12	256	32/26	25

**Table 3** C7t ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	Max. Network PPS (10,000)	Network Connections (10,000)	Max. NIC Queues	Max. NICs	Max. Supplementary NICs	Max./Assured EVS Disk Bandwidth (Gbit/s)	Max. EVS Disk IOPS (10,000)	Virtualization
c7t.large.4	2	8	10/1.28	80	50	2	2	16	6/1.5	16	QingTian
c7t.xlarge.4	4	16	16/2.56	150	50	2	3	32	6/2	16
c7t.2xlarge.4	8	32	20/4	200	100	4	4	64	6/3	16
c7t.3xlarge.4	12	48	34/6.4	300	150	4	6	96	6/4	16
c7t.4xlarge.4	16	64	40/8	400	150	8	8	128	6/5	16
c7t.6xlarge.4	24	96	50/14.4	600	200	8	8	192	None/6	16
c7t.8xlarge.4	32	128	60/16	800	300	16	8	256	None/8	16
c7t.12xlarge.4	48	192	70/28.8	1,200	400	16	8	256	None/10	16
c7t.16xlarge.4	64	256	72/32	1,500	500	28	8	256	None/16	16
c7t.24xlarge.4	96	384	80/40	2,400	800	32	8	256	None/20	22.5
c7t.32xlarge.4	128	512	90/48	3,000	1,000	32	8	256	None/24	30

**Table 4** C6ne ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces
c6ne.large.2	2	4	4/1.2	40	2	2	16
c6ne.large.4	2	8	4/1.2	40	2	2	16
c6ne.xlarge.2	4	8	8/2.4	80	2	3	32
c6ne.xlarge.4	4	16	8/2.4	80	2	3	32
c6ne.2xlarge.2	8	16	15/4.5	150	4	4	64
c6ne.2xlarge.4	8	32	15/4.5	150	4	4	64
c6ne.3xlarge.2	12	24	17/7	200	4	6	96
c6ne.3xlarge.4	12	48	17/7	200	4	6	96
c6ne.4xlarge.2	16	32	20/9	280	8	8	128
c6ne.4xlarge.4	16	64	20/9	280	8	8	128
c6ne.6xlarge.2	24	48	25/14	400	8	8	192
c6ne.6xlarge.4	24	96	25/14	400	8	8	192
c6ne.8xlarge.2	32	64	30/18	550	16	8	256
c6ne.8xlarge.4	32	128	30/18	550	16	8	256
c6ne.12xlarge.2	48	96	35/27	750	16	8	256
c6ne.12xlarge.4	48	192	35/27	750	16	8	256
c6ne.16xlarge.2	64	128	40/36	1,000	32	8	256
c6ne.16xlarge.4	64	256	40/36	1,000	32	8	256

**Table 5** C6s ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	Max. Network PPS (10,000)	Network Connections (10,000)	Max. NIC Queues	Max. NICs	Max. EVS Disk Bandwidth (Gbit/s)	Max. EVS Disk IOPS (10,000)	Virtualization
c6s.large.2	2	4	1/1	30	50	2	2	1.5	1	KVM
c6s.xlarge.2	4	8	2/2	60	50	2	3	2	2	KVM
c6s.2xlarge.2	8	16	4/4	120	100	4	4	3	2.5	KVM
c6s.3xlarge.2	12	24	5.5/5.5	180	150	4	6	4	3	KVM
c6s.4xlarge.2	16	32	7.5/7.5	240	150	8	8	5	4	KVM
c6s.6xlarge.2	24	48	11/11	350	200	8	8	7.5	5	KVM
c6s.8xlarge.2	32	64	15/15	450	300	16	8	10	6	KVM
c6s.12xlarge.2	48	96	22/22	650	400	16	8	15	10	KVM
c6s.16xlarge.2	64	128	30/30	850	500	32	8	20	15	KVM
c6s.large.4	2	8	1/1	30	50	2	2	1.5	1	KVM
c6s.xlarge.4	4	16	2/2	60	50	2	3	2	2	KVM
c6s.2xlarge.4	8	32	4/4	120	100	4	4	3	2.5	KVM
c6s.3xlarge.4	12	48	5.5/5.5	180	150	4	6	4	3	KVM
c6s.4xlarge.4	16	64	7.5/7.5	240	150	8	8	5	4	KVM
c6s.6xlarge.4	24	96	11/11	350	200	8	8	7.5	5	KVM
c6s.8xlarge.4	32	128	15/15	450	300	16	8	10	6	KVM
c6s.12xlarge.4	48	192	22/22	650	400	16	8	15	10	KVM
c6s.16xlarge.4	64	256	30/30	850	500	32	8	20	15	KVM

**Table 6** aC6 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Virtualization
ac6.2xlarge.1	8	8	4.0/1.4	80	4	4	KVM
ac6.2xlarge.2	8	16	4.0/1.4	80	4	4	KVM
ac6.2xlarge.4	8	32	4.0/1.4	80	4	4	KVM
ac6.4xlarge.1	16	16	7.5/2.8	100	8	8	KVM
ac6.4xlarge.2	16	32	7.5/2.8	100	8	8	KVM
ac6.4xlarge.4	16	64	7.5/2.8	100	8	8	KVM
ac6.6xlarge.2	24	48	11.0/4.2	150	8	8	KVM
ac6.6xlarge.4	24	96	11.0/4.2	150	8	8	KVM
ac6.8xlarge.2	32	64	15.5/5.5	200	16	8	KVM
ac6.8xlarge.4	32	128	15.5/5.5	200	16	8	KVM
ac6.12xlarge.2	48	96	24.0/8.0	250	16	8	KVM
ac6.12xlarge.4	48	192	24.0/8.0	250	16	8	KVM
ac6.16xlarge.2	64	128	28.0/11.0	300	16	8	KVM
ac6.16xlarge.4	64	256	28.0/11.0	300	16	8	KVM
ac6.24xlarge.2	96	192	30.0/16.0	400	32	16	KVM
ac6.24xlarge.4	96	384	30.0/16.0	400	32	16	KVM
ac6.32xlarge.2	128	256	35.0/22.0	600	32	16	KVM
ac6.32xlarge.4	128	256	35.0/22.0	600	32	16	KVM
ac6.48xlarge.2	192	384	38.0/30.0	700	32	16	KVM
ac6.48xlarge.4	192	768	38.0/30.0	700	32	16	KVM
ac6.56xlarge.2	224	432	40.0/35.0	800	32	16	KVM
ac6.56xlarge.2	224	924	40.0/35.0	800	32	16	KVM

General Computing ECSs

General computing ECSs provide a balance of compute, memory, and networking resources and a baseline level of vCPU performance with the ability to burst above the baseline.

**Table 7** General computing ECS features
ECS Type	Compute	Network	Supported Cluster Type
S7n	vCPU-to-memory ratio: 1:1, 1:2, or 1:4 The number of vCPUs: 1 to 8 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.4 GHz	IPv6 access Higher compute specifications for better network performance Maximum PPS: 500,000 Maximum intranet bandwidth: 3 Gbit/s Maximum number of network connections: 300,000	CCE standard clusters CCE Turbo clusters
S7	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 8 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.8 GHz/3.5 GHz	IPv6 access Higher compute specifications for better network performance Maximum PPS: 500,000 Maximum intranet bandwidth: 3 Gbit/s	CCE standard clusters CCE Turbo clusters
aS7	vCPU-to-memory ratio: 1:2 The number of vCPUs: 2 to 16 Base/Turbo frequency: 2.45 GHz/3.5 GHz	IPv6 access Higher compute specifications for better network performance Maximum PPS: 800,000 Maximum intranet bandwidth: 8 Gbit/s	CCE standard clusters
S6	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 8 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	IPv6 access Higher compute specifications for better network performance Maximum PPS: 500,000 Maximum intranet bandwidth: 3 Gbit/s	CCE standard clusters
Sn3	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 16 Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.2 GHz/3.0 GHz	Higher compute specifications for better network performance Maximum PPS: 500,000 Maximum intranet bandwidth: 3 Gbit/s	CCE standard clusters
S3	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 16 Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.2 GHz/3.0 GHz	Higher compute specifications for better network performance Maximum PPS: 300,000 Maximum intranet bandwidth: 4 Gbit/s	CCE standard clusters
S2	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 32 Intel® Xeon® Processor E5 v4 Family Base/Turbo frequency: 2.4 GHz/3.3 GHz	Higher compute specifications for better network performance Maximum PPS: 500,000 Maximum intranet bandwidth: 6 Gbit/s	CCE standard clusters

**Table 8** S7n ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Network Connections (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
s7n.large.2	2	4	1.5/0.2	15	30	1	2	8	KVM
s7n.xlarge.2	4	8	2/0.35	25	30	1	2	16	KVM
s7n.2xlarge.2	8	16	3/0.75	50	30	2	2	32	KVM
s7n.large.4	2	8	1.5/0.2	15	30	1	2	8	KVM
s7n.xlarge.4	4	16	2/0.35	25	30	1	2	16	KVM
s7n.2xlarge.4	8	32	3/0.75	50	30	2	2	32	KVM

**Table 9** S7 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
s7.large.2	2	4	1.5/0.2	15	1	2	8	KVM
s7.xlarge.2	4	8	2/0.35	25	1	2	16	KVM
s7.2xlarge.2	8	16	3/0.75	50	2	2	32	KVM
s7.large.4	2	8	1.5/0.2	15	1	2	8	KVM
s7.xlarge.4	4	16	2/0.35	25	1	2	16	KVM
s7.2xlarge.4	8	32	3/0.75	50	2	2	32	KVM

**Table 10** aS7 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
as7.large.2	2	4	1.5/0.3	15	1	2	8	KVM
as7.xlarge.2	4	8	2/0.5	25	1	2	16	KVM
as7.2xlarge.2	8	16	4/1	40	2	2	32	KVM
as7.3xlarge.2	12	24	6/1.5	60	2	2	64	KVM
as7.4xlarge.2	16	32	8/2	80	4	2	64	KVM

**Table 11** S6 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Virtualization
s6.large.2	2	4	1.5/0.2	15	1	2	KVM
s6.xlarge.2	4	8	2/0.35	25	1	2	KVM
s6.2xlarge.2	8	16	3/0.75	50	2	2	KVM
s6.large.4	2	8	1.5/0.2	15	1	2	KVM
s6.xlarge.4	4	16	2/0.35	25	1	2	KVM
s6.2xlarge.4	8	32	3/0.75	50	2	2	KVM

**Table 12** Sn3 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Virtualization
sn3.large.2	2	4	1.5/0.35	15	1	2	KVM
sn3.xlarge.2	4	8	2/0.7	25	1	2	KVM
sn3.2xlarge.2	8	16	3/1.3	50	2	2	KVM
sn3.4xlarge.2	16	32	6/2.5	100	4	2	KVM
sn3.large.4	2	8	1.5/0.35	15	1	2	KVM
sn3.xlarge.4	4	16	2/0.7	25	1	2	KVM
sn3.2xlarge.4	8	32	3/1.3	50	2	2	KVM
sn3.4xlarge.4	16	64	6/2.5	100	4	2	KVM

**Table 13** S3 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Virtualization
s3.large.2	2	4	0.8/0.2	10	1	KVM
s3.xlarge.2	4	8	1.5/0.4	15	1	KVM
s3.2xlarge.2	8	16	3/0.8	20	2	KVM
s3.4xlarge.2	16	32	4/1.5	30	4	KVM
s3.large.4	2	8	0.8/0.2	10	1	KVM
s3.xlarge.4	4	16	1.5/0.4	15	1	KVM
s3.2xlarge.4	8	32	3/0.8	20	2	KVM
s3.4xlarge.4	16	64	4/1.5	30	4	KVM

**Table 14** S2 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Virtualization
s2.large.2	2	4	0.8/0.2	10	1	KVM
s2.xlarge.2	4	8	1.5/0.4	15	1	KVM
s2.2xlarge.2	8	16	3/0.8	20	2	KVM
s2.4xlarge.2	16	32	4/1.5	30	4	KVM
s2.8xlarge.2	32	64	6/3	50	8	KVM
s2.large.4	2	8	0.8/0.2	10	1	KVM
s2.xlarge.4	4	16	1.5/0.4	15	1	KVM
s2.2xlarge.4	8	32	3/0.8	20	2	KVM
s2.4xlarge.4	16	64	4/1.5	30	4	KVM
s2.8xlarge.4	32	128	6/3	50	8	KVM

Memory-optimized ECSs

Memory-optimized ECSs have a large memory and provide high memory performance. They are designed for memory-intensive applications that process a large amount of data.

**Table 15** Memory-optimized ECS features
ECS Type	Compute	Network	Supported Cluster Type
M7	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 128 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 12,000,000 Maximum intranet bandwidth: 42 Gbit/s	CCE standard clusters CCE Turbo clusters
aM7	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 232 Base/Turbo frequency: 2.45 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 20,000,000 Maximum intranet bandwidth: 100 Gbit/s Maximum number of network connections: 20,000,000	CCE standard clusters CCE Turbo clusters
M7n	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 96 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.4 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 8,500,000 Maximum intranet bandwidth: 40 Gbit/s Maximum number of network connections: 8,000,000	CCE standard clusters CCE Turbo clusters
M6ne	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 64 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 10,000,000 Maximum intranet bandwidth: 40 Gbit/s	CCE Turbo clusters
M6	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 64 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 10,000,000 Maximum intranet bandwidth: 40 Gbit/s	CCE standard clusters
M3	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 60 Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 5,000,000 Maximum intranet bandwidth: 17 Gbit/s	CCE standard clusters

**Table 16** M7 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
m7.large.8	2	16	4/0.8	40	2	2	16	QingTian-backed simplified, proprietary virtualization
m7.xlarge.8	4	32	8/1.6	80	2	3	32
m7.2xlarge.8	8	64	15/3	150	4	4	64
m7.3xlarge.8	12	96	17/5	200	4	6	96
m7.4xlarge.8	16	128	20/6	280	8	8	128
m7.6xlarge.8	24	192	25/9	400	8	8	192
m7.8xlarge.8	32	256	30/12	550	16	8	256
m7.12xlarge.8	48	384	35/18	750	16	8	256
m7.16xlarge.8	64	512	36/24	1000	28	8	256
m7.24xlarge.8	96	768	40/36	1100	32	8	256
m7.32xlarge.8	128	1024	42/40	1200	32	8	256

**Table 17** aS7 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Network Connections (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
am7.large.8	2	16	2/1	40	75	2	2	16	KVM
am7.xlarge.8	4	32	3/1.5	60	150	2	3	32
am7.2xlarge.8	8	64	4/2.5	100	200	4	4	64
am7.4xlarge.8	16	128	8/5	200	300	8	8	128
am7.8xlarge.8	32	256	15/8	300	400	16	8	256
am7.12xlarge.8	48	384	22/12	400	500	16	8	256
am7.16xlarge.8	64	512	28/16	550	800	24	12	256
am7.24xlarge.8	96	768	40/25	800	1000	24	12	256
am7.29xlarge.8	116	934	50/30	950	1200	32	16	256
am7.32xlarge.8	128	1024	55/35	1000	1300	32	16	256
am7.48xlarge.8	192	1536	100/80	1600	1800	32	16	256
am7.58xlarge.8	232	1868	120/100	2000	2000	32	16	256

**Table 18** M7n ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Network Connections (10,000)	Max. Network Interface Queues	Max. Network Interfaces	EVS Basic Bandwidth/Burst Bandwidth (Gbit/s)	Virtualization
m7n.large.8	2	16	4/0.8	40	50	2	2	1.5/6	KVM
m7n.xlarge.8	4	32	8/1.6	80	50	2	3	2/6	KVM
m7n.2xlarge.8	8	64	15/3	150	100	4	4	3/6	KVM
m7n.3xlarge.8	12	96	17/5	200	150	4	6	4/6	KVM
m7n.4xlarge.8	16	128	20/6	280	150	8	8	5/6	KVM
m7n.6xlarge.8	24	192	25/9	400	200	8	8	6/None	KVM
m7n.8xlarge.8	32	256	30/12	550	300	16	8	8/None	KVM
m7n.12xlarge.8	48	384	35/18	750	400	16	8	12/None	KVM
m7n.16xlarge.8	64	512	36/24	800	500	28	8	16/None	KVM
m7n.24xlarge.8	96	768	40/36	850	800	32	8	24/None	KVM

**Table 19** M6ne ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces
m6ne.large.8	2	16	4/1.2	40	2	2	16
m6ne.xlarge.8	4	32	8/2.4	80	2	3	32
m6ne.2xlarge.8	8	64	15/4.5	150	4	4	64
m6ne.3xlarge.8	12	96	17/7	200	4	6	96
m6ne.4xlarge.8	16	128	20/9	280	8	8	128
m6ne.6xlarge.8	24	192	25/14	400	8	8	192
m6ne.8xlarge.8	32	256	30/18	550	16	8	256
m6ne.16xlarge.8	64	512	40/36	1000	32	8	256

**Table 20** M6 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	PPS (10,000)	Network Connections (10,000)	Max. NIC Queues	Max. NICs	Max./Assured EVS Disk Bandwidth (Gbit/s)	Max. EVS Disk IOPS (10,000)	Virtualization
m6.large.8	2	16	4/1.2	40	50	2	2	5/1	1	KVM
m6.xlarge.8	4	32	8/2.4	80	50	2	3	5/1.5	2	KVM
m6.2xlarge.8	8	64	15/4.5	150	100	4	4	5/2	2.5	KVM
m6.3xlarge.8	12	96	17/7	200	150	4	6	5/2.5	3	KVM
m6.4xlarge.8	16	128	20/9	280	150	8	8	5/3.5	4	KVM
m6.6xlarge.8	24	192	25/14	400	200	8	8	5/4	5	KVM
m6.8xlarge.8	32	256	30/18	550	300	16	8	10/7	6	KVM
m6.16xlarge.8	64	512	40/36	1,000	500	32	8	None/20	15	KVM

General Computing-Basic ECSs

General computing-basic ECSs are suitable for scenarios that require moderate CPU performance generally but burstable high performance occasionally while keeping costs low. The performance of such ECSs is constrained by the baseline performance and CPU credits. There are no additional charges for using CPU credits. For details, see CPU Credits.

**Table 21** General computing-basic ECS features
ECS Type	Compute	Network	Supported Cluster Type
T6	vCPU-to-memory ratio: 1:1, 1:2, or 1:4 The number of vCPUs: 2 to 16 Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.2 GHz/3.0 GHz	Higher compute specifications for better network performance Maximum PPS: 600,000 Maximum intranet bandwidth: 3 Gbit/s	CCE standard clusters

**Table 22** T6 ECS specifications
Flavor	vCPUs	Memory (GiB)	CPU Baseline (%)	Average CPU Baseline (%)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interfaces	Virtualization
t6.xlarge.1	4	4	80	20	1/0.2	20	2	KVM
t6.2xlarge.1	8	8	120	15	2/0.4	40	2	KVM
t6.4xlarge.1	16	16	240	15	3/0.8	60	2	KVM
t6.large.2	2	4	40	20	0.5/0.1	10	1	KVM
t6.xlarge.2	4	8	80	20	1/0.2	20	2	KVM
t6.2xlarge.2	8	16	120	15	2/0.4	40	2	KVM
t6.4xlarge.2	16	32	240	15	3/0.8	60	2	KVM
t6.large.4	2	8	40	20	0.5/0.1	10	1	KVM
t6.xlarge.4	4	16	80	20	1/0.2	20	2	KVM
t6.2xlarge.4	8	32	120	15	2/0.4	40	2	KVM

GPU-accelerated ECSs

GPU-accelerated ECSs provide outstanding floating-point computing capabilities. They are suitable for applications that require real-time, highly concurrent massive computing.

GPU-accelerated ECSs are classified as G series and P series.

G series: Graphics-accelerated ECSs, which are suitable for 3D animation rendering and CAD
P series: Computing-accelerated or inference-accelerated ECSs, which are suitable for deep learning, scientific computing, and CAE

**Table 23** GPU-accelerated ECSs
Classification	ECS Type	GPU	CUDA Cores per GPU	Single-GPU Performance	Application	Supported Cluster Type
Graphics-accelerated	G6ne	NVIDIA T4 (GPU passthrough)	2560	8.1 TFLOPS of single-precision floating-point computing 130 INT8 TOPS 260 INT4 TOPS	Cloud desktop, image rendering, 3D visualization, and heavy-load graphics design	CCE Turbo clusters
Graphics-accelerated	G6	NVIDIA T4 (GPU passthrough)	2560	8.1 TFLOPS of single-precision floating-point computing 130 INT8 TOPS 260 INT4 TOPS	Cloud desktop, image rendering, 3D visualization, and heavy-load graphics design	CCE standard clusters
Graphics-accelerated	G5	NVIDIA V100 (GPU passthrough)	5120	14 TFLOPS of single-precision floating-point computing 7 TFLOPS of double-precision floating-point computing 112 TFLOPS Tensor Cores for deep learning acceleration	Cloud desktop, image rendering, 3D visualization, and heavy-load graphics design	CCE standard clusters
Computing-accelerated	P2s	NVIDIA V100	5120	14 TFLOPS of single-precision floating-point computing 7 TFLOPS of double-precision floating-point computing 112 TFLOPS Tensor Cores for deep learning acceleration	AI deep learning training, scientific computing, computational fluid dynamics, computational finance, seismic analysis, molecular modeling, and genomics	CCE standard clusters
Computing-accelerated	P2v	NVIDIA V100 NVLink (GPU passthrough)	5120	15.7 TFLOPS of single-precision floating-point computing 7.8 TFLOPS of double-precision floating-point computing 125 TFLOPS Tensor Cores for deep learning acceleration 300 GiB/s NVLink	Machine learning, deep learning, inference training, scientific computing, seismic analysis, computational finance, rendering, multimedia encoding and decoding	CCE standard clusters
Inference-accelerated	Pi2ne	NVIDIA T4 (GPU passthrough)	2560	8.1 TFLOPS of single-precision floating-point computing 130 INT8 TOPS 260 INT4 TOPS	Machine learning, deep learning, inference training, scientific computing, seismic analysis, computational finance, rendering, multimedia encoding and decoding	CCE Turbo clusters
Inference-accelerated	Pi2	NVIDIA T4 (GPU passthrough)	2560	8.1 TFLOPS of single-precision floating-point computing 130 INT8 TOPS 260 INT4 TOPS	Machine learning, deep learning, inference training, scientific computing, seismic analysis, computational finance, rendering, multimedia encoding and decoding	CCE standard clusters
Inference-accelerated	Pi1	NVIDIA P4 (GPU passthrough)	2560	5.5 TFLOPS of single-precision floating-point computing	Machine learning, deep learning, inference training, scientific computing, seismic analysis, computational finance, rendering, multimedia encoding and decoding	CCE standard clusters

**Table 24** P2v ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	Max. Network PPS (10,000)	Max. NIC Queues	Max. NICs	GPUs	GPU Connection	GPU Memory (GiB)	Virtualization	Hardware
p2v.2xlarge.8	8	64	10/4	50	4	4	1 × V100	N/A	1 × 16 GiB	KVM	CPU: Intel® Xeon® Skylake-SP Gold 6151 v5
p2v.4xlarge.8	16	128	15/8	100	8	8	2 × V100	NVLink	2 × 16 GiB	KVM
p2v.8xlarge.8	32	256	25/15	200	16	8	4 × V100	NVLink	4 × 16 GiB	KVM
p2v.16xlarge.8	64	512	30/30	400	32	8	8 × V100	NVLink	8 × 16 GiB	KVM

Disk-intensive ECSs

Disk-intensive ECSs are delivered with local disks for high storage bandwidth and IOPS. Local disks are more cost-effective in massive data storage scenarios.

**Table 25** Disk-intensive ECS features
ECS Type	Compute	Network	Supported Cluster Type
D7	vCPU-to-memory ratio: 1:4 The number of vCPUs: 4 to 72 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 8,500,000 Maximum intranet bandwidth: 42 Gbit/s	CCE standard clusters CCE Turbo clusters
D6	vCPU-to-memory ratio: 1:4 The number of vCPUs: 4 to 72 Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 9,000,000 Maximum intranet bandwidth: 44 Gbit/s	CCE standard clusters CCE Turbo clusters
D3	vCPU-to-memory ratio: 1:8 or 1:10 The number of vCPUs: 4 to 56 Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 5,000,000 Maximum intranet bandwidth: 40 Gbit/s	CCE standard clusters CCE Turbo clusters

**Table 26** D6 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	Max. Network PPS (10,000)	Network Connections (10,000)	Max. NIC Queues	Max. NICs	Local Disks (GiB)	Virtualization
d6.xlarge.4	4	16	5/2	60	50	2	3	2 × 3,600	KVM
d6.2xlarge.4	8	32	10/4	120	100	4	4	4 × 3,600	KVM
d6.4xlarge.4	16	64	20/7.5	240	150	8	8	8 × 3,600	KVM
d6.6xlarge.4	24	96	25/11	350	200	8	8	12 × 3,600	KVM
d6.8xlarge.4	32	128	30/15	450	300	16	8	16 × 3,600	KVM
d6.12xlarge.4	48	192	40/22	650	400	16	8	24 × 3,600	KVM
d6.16xlarge.4	64	256	42/30	850	500	32	8	32 × 3,600	KVM

Kunpeng General Computing-plus ECSs

Kunpeng general computing-plus ECSs use Kunpeng processors to provide powerful compute and high-performance network, meeting enterprise requirements for cost-effective, secure, reliable cloud services.

**Table 27** Kunpeng general computing-plus ECS features
ECS Type	Compute	Network	Supported Cluster Type
kC2	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 1 to 160 New Kunpeng 920 processor Base frequency: 2.9 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 40,000,000 Maximum intranet bandwidth: 102 Gbit/s	CCE standard clusters CCE Turbo clusters
kC1	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 60 Kunpeng 920 processor Base frequency: 2.6 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 4,000,000 Maximum intranet bandwidth: 30 Gbit/s	CCE standard clusters
kC1n	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 8 to 48 Kunpeng 920 processor Base frequency: 2.6 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 3,500,000 Maximum intranet bandwidth: 25 Gbit/s	CCE standard clusters CCE Turbo clusters

**Table 28** kC2 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
kc2.large.2	2	4	10/2	90	2	3	16	QingTian-backed simplified virtualization
kc2.xlarge.2	4	8	10/4	100	4	3	32
kc2.2xlarge.2	8	16	15/8	200	8	4	64
kc2.3xlarge.2	12	24	20/10	300	12	4	96
kc2.4xlarge.2	16	32	30/15	400	16	6	128
kc2.6xlarge.2	24	48	40/20	600	24	6	192
kc2.8xlarge.2	32	64	50/25	800	32	6	256
kc2.10xlarge.2	40	80	60/30	1,000	40	6	256
kc2.12xlarge.2	48	96	80/40	1,200	48	6	256
kc2.16xlarge.2	64	128	90/50	1,600	64	6	256
kc2.20xlarge.2	80	160	102/60	2,000	64	8	256
kc2.24xlarge.2	96	192	102/80	2,400	64	8	256
kc2.32xlarge.2	128	256	102/90	3,200	64	8	256
kc2.40xlarge.2	160	320	102/102	4,000	64	15	256

Kunpeng Memory-optimized ECSs

Kunpeng memory-optimized ECSs use Kunpeng 920 processors and 25GE high-speed intelligent network interfaces to provide up to 480-GiB DDR4 memory with high network performance for large in-memory datasets.

**Table 29** Kunpeng memory-optimized ECS features
ECS Type	Compute	Network	Supported Cluster Type
kM2	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 96 Kunpeng 920 processor Base frequency: 2.9 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 24,000,000 Maximum intranet bandwidth: 102 Gbit/s	CCE standard clusters CCE Turbo clusters
kM1	vCPU-to-memory ratio: 1:8 The number of vCPUs: 2 to 60 Kunpeng 920 processor Base frequency: 2.6 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 4,000,000 Maximum intranet bandwidth: 30 Gbit/s	CCE standard clusters

Kunpeng Ultra-high I/O ECSs

Kunpeng ultra-high I/O ECSs use Kunpeng 920 processors and 25GE high-speed intelligent NICs to provide up to 480 GiB DDR4-based memory with high network performance for large in-memory datasets.

**Table 30** Kunpeng ultra-high I/O ECS features
ECS Type	Compute	Network	Supported Cluster Type
kI1	vCPU-to-memory ratio: 1:4 The number of vCPUs: 8 to 64 Kunpeng 920 processor Base frequency: 2.6 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 4,000,000 Maximum intranet bandwidth: 30 Gbit/s	CCE standard clusters

**Table 31** kI1 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Network Bandwidth (Gbit/s)	Max. Network PPS (10,000)	Max. NICs	Max. NIC Queues	Local Disks	Virtualization
ki1.2xlarge.4	8	32	7/3	80	4	4	1 × 3,200 GiB	KVM
ki1.4xlarge.4	16	64	12/6	140	6	4	2 × 3,200 GiB	KVM
ki1.6xlarge.4	24	96	15/8.5	200	6	8	3 × 3,200 GiB	KVM
ki1.8xlarge.4	32	128	18/10	260	6	8	4 × 3,200 GiB	KVM
ki1.12xlarge.4	48	192	25/16	350	6	16	6 × 3,200 GiB	KVM
ki1.16xlarge.4	64	228	30/20	400	6	16	8 × 3,200 GiB	KVM

Ultra-high I/O ECSs

Ultra-high I/O ECSs use high-performance local NVMe SSDs to provide high storage IOPS and low read/write latency.

**Table 32** Ultra-high I/O ECS features
ECS Type	Compute	Network	Supported Cluster Type
Ir7	vCPU-to-memory ratio: 1:4 The number of vCPUs: 2 to 64 3rd Gen Intel® Xeon® Platinum Scalable Processor Base/Turbo frequency: 3.0 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 6,000,000 Maximum intranet bandwidth: 40 Gbit/s	CCE standard clusters CCE Turbo clusters
I7	vCPU-to-memory ratio: 1:4 The number of vCPUs: 8 to 96 3rd Gen Intel® Xeon® Platinum Scalable Processor Base/Turbo frequency: 3.0 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 8,000,000 Maximum intranet bandwidth: 44 Gbit/s	CCE standard clusters CCE Turbo clusters
Ir3	vCPU-to-memory ratio: 1:4 The number of vCPUs: 2 to 32 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 5,500,000 Maximum intranet bandwidth: 30 Gbit/s	CCE standard clusters
I3	vCPU-to-memory ratio: 1:8 The number of vCPUs: 8 to 64 Intel® Xeon® Scalable Processor Base/Turbo frequency: 3.0 GHz/3.4 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 5,000,000 Maximum intranet bandwidth: 25 Gbit/s	CCE standard clusters

FlexusX Instances

FlexusX instances are next-generation cloud servers that offer flexible computing for small and medium-sized enterprises and developers. They allow for flexible configuration of vCPU-to-memory ratios, so you can choose the appropriate flavors based on your service resource needs to minimize resource overhead.

**Table 33** FlexusX instance features
Flavor	Compute	Network	Supported Cluster Type
FlexusX1	vCPU-to-memory ratio: user-defined The number of vCPUs: 1 to 16 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.8 GHz/3.5 GHz	Higher compute specifications for better network performance Maximum PPS: 1,000,000 Maximum intranet bandwidth: 12 Gbit/s Maximum number of network connections: 1,000,000	CCE standard clusters CCE Turbo clusters
FlexusX1e	vCPU-to-memory ratio: user-defined The number of vCPUs: 2 to 32 Base/Turbo frequency: 2.45 GHz/3.5 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 3,000,000 Maximum intranet bandwidth: 16 Gbit/s	CCE standard clusters CCE Turbo clusters
FlexusX2e	vCPU-to-memory ratio: user-defined The number of vCPUs: 2 to 128 Base/Turbo frequency: 2.4 GHz/3.7 GHz	IPv6 access Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 10,000,000 Maximum intranet bandwidth: 55 Gbit/s	CCE standard clusters CCE Turbo clusters

**Table 34** X1 ECS specifications
vCPU	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Network Connections (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
2	2/0.2	30	50	2	2	8	KVM
3	2/0.2	30	50	2	2	8	KVM
4	3/0.4	50	50	2	2	16	KVM
5	3/0.4	50	50	2	2	16	KVM
6	4/0.6	60	50	2	2	24	KVM
7	4/0.6	60	50	2	2	24	KVM
8	6/0.8	80	100	2	2	32	KVM
9	6/0.8	80	100	2	2	32	KVM
12	8/1.2	90	100	4	3	48	KVM
16	12/1.6	100	100	4	3	64	KVM

**Table 35** X1e ECS specifications
vCPU	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
2	2/1	40	2	2	16	KVM
4	3/1.5	60	2	3	32	KVM
8	6/2.5	100	4	4	64	KVM
12	8/4	150	4	6	96	KVM
16	12/5	200	8	8	128	KVM
20	13/5	220	8	8	128	KVM
24	14/6	250	8	8	192	KVM
28	15/6	280	8	8	192	KVM
32	16/8	300	16	8	256	KVM

**Table 36** X2e ECS specifications
vCPU	Max./Assured Bandwidth (Gbit/s)	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Max. Supplementary Network Interfaces	Virtualization
2	2/1	40	2	2	8	QingTian-backed simplified, proprietary virtualization
4	3/1.5	60	2	3	16
8	6/2.5	100	4	4	32
16	12/5	200	8	4	64
32	16/8	300	16	4	128
48	22/12	400	16	4	256
64	28/16	550	28	4	256
96	40/25	800	32	8	256
128	55/35	1000	32	10	256

AI-accelerated ECSs

AI-accelerated ECSs are dedicated for accelerating AI applications. These ECSs use Ascend processors that feature low power consumption, high compute capabilities, and significantly improved energy efficiency for AI inference and video encoding and decoding.

AI-accelerated ECSs are classified as kAi series and Ai series ECSs.

kAi series: Arm ECSs, which use Kunpeng 920 processors
AI series: x86 ECSs, which use Intel Xeon processors

**Table 37** AI-accelerated ECS features
ECS Type	Compute	Network	Supported Cluster Type
kAi1s	vCPU-to-memory ratio: 1:1 or 1:2 The number of vCPUs: 4 to 48 Kunpeng 920 processor Base frequency: 2.6 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 4,000,000 Maximum intranet bandwidth: 20 Gbit/s	CCE standard clusters
Ai1s	vCPU-to-memory ratio: 1:2 or 1:4 The number of vCPUs: 2 to 32 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 2,000,000 Maximum intranet bandwidth: 25 Gbit/s	CCE standard clusters
Ai2	vCPU-to-memory ratio: 1:4 The number of vCPUs: 4 to 96 3rd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 8,500,000 Maximum intranet bandwidth: 40 Gbit/s	CCE standard clusters CCE Turbo clusters
Ai1	vCPU-to-memory ratio: 1:4 The number of vCPUs: 2 to 32 2nd Gen Intel® Xeon® Scalable Processor Base/Turbo frequency: 2.6 GHz/3.5 GHz	Ultra-high PPS throughput Higher compute specifications for better network performance Maximum PPS: 2,000,000 Maximum intranet bandwidth: 25 Gbit/s	CCE standard clusters

**Table 38** kAi1s ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth	Max. PPS (10,000)	Max. Network Interface Queues	Max. Network Interfaces	Ascend 310	Virtualization
kai1s.xlarge.1	4	4	3/0.8	20	2	2	1	KVM
kai1s.2xlarge.1	8	8	4/1.5	40	2	3	2	KVM
kai1s.4xlarge.1	16	16	6/3	80	4	4	4	KVM
kai1s.3xlarge.2	12	24	8/4	100	4	4	4	KVM
kai1s.4xlarge.2	16	32	10/6	140	4	5	6	KVM
kai1s.6xlarge.2	24	48	12/8	200	8	6	8	KVM
kai1s.9xlarge.2	36	72	16/12	280	8	6	12	KVM
kai1s.12xlarge.2	48	96	20/16	400	16	6	12	KVM

**Table 39** Ai2 ECS specifications
Flavor	vCPUs	Memory (GiB)	Max./Assured Bandwidth	Max. PPS (10,000)	Ascend Chip	Ascend RAM (GiB)	Max. Network Interface Queues	Max. Network Interfaces	Virtualization
ai2.xlarge.4	4	16	8/1.6	80	1	24	3	3	KVM
ai2.2xlarge.4	8	32	15/3	150	1	24	4	4	KVM
ai2.4xlarge.4	16	64	20/6	280	1	24	8	8	KVM
ai2.8xlarge.4	32	128	30/12	550	2	48	8	8	KVM
ai2.16xlarge.4	64	256	36/24	800	4	96	16	8	KVM
ai2.24xlarge.4	96	384	40/36	850	6	144	32	8	KVM