tencent cloud

Feedback

Use Limits

Last updated: 2024-05-16 10:55:46

    Account-level Limits for Purchasing CVM Instances

    You need to sign up for a Tencent Cloud account. For more information, see Signing up for a Tencent Cloud Account.
    If you create a pay-as-you-go CVM, the system will freeze the cost of one-hour CVM usage. Make sure that your account has sufficient balance for the order.

    CVM Instance Use Limits

    Virtualized software cannot be installed or re-virtualized (such as installing VMware or Hyper-V).
    You cannot use sound cards or mount external hardware devices (such as USB flash drives, external disks, and U-keys).
    Only Linux CVMs can act as a public gateway.

    CVM Instance Purchase Limits

    The purchase limit of pay-as-you-go CVM instances for each user in each AZ is between 30 and 60.
    For more information, see Purchase Limits.

    Image Limits

    Public images: No use limits.
    Custom images: Each region supports a maximum of 500 custom images.
    Shared images: Each custom image can be shared with a maximum of 500 Tencent Cloud users. Custom images can only be shared with accounts in the same region as the source account.
    For more information, see Image Types.

    ENI Limits

    Based on CPU and memory configurations, the number of ENIs bound to a CVM instance differs from the number of private IPs bound to an ENI. The quotes are as shown below:
    Note:
    The number of IP addresses bound to a single ENI indicates the maximum number allowed. The EIP quota is not provided based on this upper limit but based on EIP use limits.
    ENIs per CVM instance
    Private IPs per ENI
    Model
    Instance Type
    Number of ENIs
    CPU: 1 core
    CPU: 2 cores
    CPU: 4 cores
    CPU: 6 cores
    CPU: 8 cores
    CPU: 10 cores
    CPU: 12 cores
    CPU: 14 cores
    CPU: 16 cores
    CPU: >16 cores
    Standard
    Standard S5
    2
    4
    4
    -
    6
    -
    -
    -
    8
    8
    Standard Storage Optimized S5se
    -
    -
    4
    -
    6
    -
    -
    -
    8
    8
    Standard SA2
    2
    4
    4
    -
    6
    -
    -
    -
    8
    8
    Standard S4
    2
    4
    4
    -
    6
    -
    -
    -
    8
    8
    Standard Network-optimized SN3ne
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Standard S3
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Standard SA1
    2
    2
    4
    -
    6
    -
    -
    -
    8
    8
    Standard S2
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Standard S1
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    High IO
    High IO IT5
    -
    -
    -
    -
    -
    -
    -
    -
    8
    8
    High IO IT3
    -
    -
    -
    -
    -
    -
    -
    -
    8
    8
    Memory Optimized
    Memory Optimized M5
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Memory Optimized M4
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Memory Optimized M3
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Memory Optimized M2
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Memory Optimized M1
    2
    4
    4
    -
    6
    -
    8
    -
    8
    8
    Compute
    Compute Optimized C4
    -
    -
    4
    -
    6
    -
    -
    -
    8
    8
    Compute Network-optimized CN3
    -
    -
    4
    -
    6
    -
    -
    -
    8
    8
    Compute C3
    -
    -
    4
    -
    6
    -
    -
    -
    8
    8
    Compute C2
    -
    -
    4
    -
    6
    -
    -
    -
    8
    8
    GPU-based
    GPU Compute GN6
    -
    -
    -
    -
    -
    -
    -
    -
    -
    8
    GPU Compute GN6S
    -
    -
    4
    -
    6
    -
    -
    -
    -
    -
    GPU Compute GN7
    -
    -
    4
    -
    6
    -
    -
    -
    -
    8
    GPU Compute GN8
    -
    -
    -
    4
    -
    -
    -
    8
    -
    8
    GPU Compute GN10X
    -
    -
    -
    -
    6
    -
    -
    -
    -
    8
    GPU Compute GN10Xp
    -
    -
    -
    -
    -
    6
    -
    -
    -
    8
    FPGA-based
    FPGA Accelerated FX4
    -
    -
    -
    -
    -
    6
    -
    -
    -
    8
    Big Data
    Big Data D3
    -
    -
    -
    -
    6
    -
    -
    -
    8
    8
    Big Data D2
    -
    -
    -
    -
    6
    -
    -
    -
    8
    8
    Big Data D1
    -
    -
    -
    -
    6
    -
    -
    -
    -
    8
    CPM
    Not supported
    
    Model
    Instance Type
    Private IPs bound to a single ENI
    CPU: 1 core
    CPU: 2 cores
    CPU: 4 cores
    CPU: 6 cores
    CPU: 8 cores
    CPU: 10 cores
    CPU: 12 cores
    CPU: 14 cores
    CPU: 16 cores
    CPU: >16 cores
    Standard
    Standard S5
    6
    10
    10
    -
    20
    -
    -
    -
    30
    30
    Standard Storage Optimized S5se
    -
    -
    20
    -
    20
    -
    -
    -
    30
    30
    Standard SA2
    6
    10
    10
    -
    20
    -
    -
    -
    30
    30
    Standard S4
    6
    10
    10
    -
    20
    -
    -
    -
    30
    30
    Standard Network-optimized SN3ne
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Standard S3
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Standard SA1
    1 GB memory: 2>1 GB memory: 6
    10
    8 GB memory: 1016 GB memory: 20
    -
    20
    -
    -
    -
    30
    30
    Standard S2
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Standard S1
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    High IO
    High IO IT5
    -
    -
    -
    -
    -
    -
    -
    -
    30
    30
    High IO IT3
    -
    -
    -
    -
    -
    -
    -
    -
    30
    30
    Memory Optimized
    Memory Optimized M5
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Memory Optimized M4
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Memory Optimized M3
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Memory Optimized M2
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Memory Optimized M1
    6
    10
    10
    -
    20
    -
    30
    -
    30
    30
    Compute
    Compute Optimized C4
    -
    -
    10
    -
    20
    -
    -
    -
    30
    30
    Compute Network-optimized CN3
    -
    -
    10
    -
    20
    -
    -
    -
    30
    30
    Compute C3
    -
    -
    10
    -
    20
    -
    -
    -
    30
    30
    Compute C2
    -
    -
    10
    -
    20
    -
    -
    -
    30
    30
    GPU-based
    GPU Compute GN2
    -
    -
    -
    -
    -
    -
    -
    -
    -
    30
    GPU Compute GN6
    -
    -
    -
    -
    -
    -
    -
    -
    -
    30
    GPU Compute GN6S
    -
    -
    10
    -
    20
    -
    -
    -
    -
    -
    GPU Compute GN7
    -
    -
    10
    -
    20
    -
    -
    -
    -
    30
    GPU Compute GN8
    -
    -
    -
    10
    -
    -
    -
    30
    -
    30
    GPU Compute GN10X
    -
    -
    -
    -
    20
    -
    -
    -
    -
    30
    GPU Compute GN10Xp
    -
    -
    -
    -
    -
    20
    -
    -
    -
    30
    FPGA-based
    FPGA Accelerated FX4
    -
    -
    -
    -
    -
    20
    -
    -
    -
    30
    Big Data
    Big Data D3
    -
    -
    -
    -
    20
    -
    -
    -
    30
    30
    Big Data D2
    -
    -
    -
    -
    20
    -
    -
    -
    30
    30
    Big Data D1
    -
    -
    -
    -
    20
    -
    -
    -
    -
    30
    CPM
    Not supported
    

    Bandwidth Limits

    Maximum outbound bandwidth (downstream bandwidth)
    The following rules apply to instances created after 00:00, February 24, 2020:
    Network Billing Method
    Instance
    Maximum Bandwidth Range (Mbps)
    Instance Billing Method
    Instance Configuration
    Bill-by-traffic
    Pay-as-you-go instances
    All
    0-100
    Bill-by-bandwidth
    Pay-as-you-go instances
    All
    0-100
    Bandwidth package
    All
    0-2000
    The following rules apply to instances created before 00:00, February 24, 2020:
    Network Billing Method
    Instance
    Range of Bandwidth Cap (Mbps)
    Instance Billing Method
    Instance Configuration
    Bill-by-traffic
    Pay-as-you-go instances
    All
    0-100
    Bill-by-bandwidth
    Pay-as-you-go instances
    All
    0-100
    Bandwidth package
    All
    0-2000
    Maximum inbound bandwidth (upstream bandwidth)
    Purchased fixed bandwidth > 10 Mbps:Tencent Cloud will assign a public network inbound bandwidth equals to the purchased bandwidth.
    Purchased fixed bandwidth < 10 Mbps, Tencent Cloud will assign 10-Mbps public network inbound bandwidth.

    Disk Limits

    Limitations
    Description
    Elastic cloud disk capability
    Starting from May 2018, all data disks purchased with CVM instances are elastic cloud disks, which can be unmounted from and remounted to CVM instances. This feature is supported in all availability zones.
    Cloud disk performance
    I/O specification applies to both input and output performance at the same time.For example, if a 1-TB SSD has a maximum random IOPS of 26,000, it means that both its read and write performance can reach this value. Due to performance limits, if the block size in this example is 4 KB or 8 KB, the maximum IOPS can be reached. If the block size is 16 KB, the maximum IOPS cannot be reached (throughput has already reached the limit of 260 MB/s).
    Elastic cloud disks per CVM
    A maximum of 20
    Snapshots per region
    64 + Number of cloud disks in the region x 64
    Attaching cloud disks to a CVM
    The CVM instance and cloud disks must be in the same availability zone.
    Snapshot rollback
    Snapshot data can only be rolled back to the cloud disk where the snapshot was created.
    Creating cloud disks using snapshot - Type limit
    Only snapshots of data disks can be used to create new elastic cloud disks.
    Creating cloud disks using snapshot - Size limit
    The capacity of new cloud disk must be larger than the source disk of the snapshot.

    Security Group Limits

    Security groups are region-specific. A CVM instance can only be bound to security groups in the same region.
    Security groups are applicable to CVM instances in any network environment.
    Each user can configure a maximum of 50 security groups for each project in a region.
    A maximum of 100 inbound or outbound rules can be configured for a security group.
    One CVM instance can be associated with multiple security groups, and a security group can be associated with multiple CVM instances.
    Security groups associated with CVM instances on the classic networkcannot filter packets from or to TencentDB (MySQL, MariaDB, SQL Server, or PostgreSQL) and NoSQL (Redis or Memcached) databases. Instead, you can use iptables or purchase CFW to filter traffic for such instances.
    The quotas are as shown below:
    Item
    Limit
    Security groups
    50 per region
    Rules in a security group
    100 for inbound rules and 100 for outbound rules
    CVM instances associated with a security group
    2,000
    Security groups associated with a CVM instance
    5
    Security groups referenced by a security group
    10

    VPC Limits

    Resource
    Limit
    VPCs per region per account
    20
    Subnets per VPC
    100
    Classic network-based CVMs associated with each VPC
    100
    Route tables per VPC
    10
    Route tables associated with each subnet
    1
    Routes per route table
    50
    HAVIPs per VPC
    10
    
    Contact Us

    Contact our sales team or business advisors to help your business.

    Technical Support

    Open a ticket if you're looking for further assistance. Our Ticket is 7x24 avaliable.

    7x24 Phone Support