• Bare Metal Server

bms
  1. Help Center
  2. Bare Metal Server
  3. User Guide
  4. Service Overview
  5. Instance
  6. Instance Family

Instance Family

Overview

An instance is a BMS. Different instance types provide varied computing capabilities, storage space, and network performance. You can select a BMS type that suits your service requirements.

BMS Types

The cloud platform provides a variety of BMS flavors. You can choose BMS flavors that best suit your needs.

  • Compute-optimized

    This type of BMS uses the SDI iNIC and supports OLTP databases with higher performance, higher memory ratio, and transactional workloads cache.

    Table 1 Compute-optimized BMS specifications

    Flavor Name/ID

    CPU

    Memory

    Local Disk

    Extended Configuration

    physical.o2.medium

    2 * 8 Core Intel Xeon E5-2667 V4 (3.2 GHz)

    256 DDR4 RAM (GB)

    2 * 800GB SAS SSD RAID 1 + NVMe SSD Card 1.6 TB

    2 x 2 * 10GE

  • High-performance computing

    High-performance BMSs provide a large number of CPU cores, large memory size, and high throughput. These BMSs are suitable for high-performance processor applications restricted by computing performance. This flavor uses the V5 CPU server and InfiniBand NIC to support quick BMS provisioning.

    Table 2 High-performance computing BMS specifications

    Flavor Name/ID

    CPU

    Memory

    Local Disk

    Extended Configuration

    physical.h2.large

    2 * 18 Core Intel Skylake 6151 V5 (3.0 GHz)

    192 DDR4 RAM (GB)

    1 * 1.6TB NVMe SSD

    1 * 100G IB + 2 * 10GE

  • GPU-accelerated

    GPU-accelerated BMSs provide outstanding floating-point computing capabilities. They are suitable for scenarios that require real-time, highly concurrent massive computing, such as deep learning, scientific computing, CAE, 3D animation rendering, and CAD. This flavor uses the G560 server, InfiniBand NIC, and the industry-leading NVIDIA Tesla P100 GPU and Tesla V100 GPU to provide you with excellent performance and cost-effectiveness.

    Table 3 GPU-accelerated BMS specifications

    Flavor Name/ID

    CPU

    Memory

    Local Disk

    Extended Configuration

    physical.p1.large

    2 * 14 Core Intel Xeon E5-2690 V4 (2.6 GHz)

    512 DDR4 RAM (GB)

    2 * 600GB SAS HDD + 6 * 800GB NVMe SSD

    NIC: 1 x 100 Gbit/s IB + 2 x 10GE

    GPU: 8 x Tesla P100

    Graphics card memory: 16 GB

    physical.p2.large

    2 * 14 Core Intel Xeon E5-2690 V4 (2.6 GHz)

    512 DDR4 RAM (GB)

    2 * 600GB SAS HDD + 6 * 800GB NVMe SSD

    NIC: 1 x 100 Gbit/s IB + 2 x 10GE

    GPU: 8 x Tesla V100

    Graphics card memory: 16 GB

    NOTE:

    physical.p1.large and physical.p2.large BMSs only support Ubuntu 16.04 and CentOS 7.4 private images.

  • Memory-optimized

    This type of BMS has DIMM memory which has quick read and write speed and high density, and is used in SAP HANA and memory databases.

    Table 4 Memory-optimized BMS specifications

    Flavor Name/ID

    CPU

    Memory

    Local Disk

    Extended Configuration

    physical.m2.medium

    4 * 24 Core Broadwell EX Xeon E7-8890 V4 (2.2 GHz)

    2048 DDR4 RAM (GB)

    2 * 600GB SAS HDD RAID 1 + 7 * 1.8TB SAS HDD RAID 5 + 2 * 1.6TB NVMe SSD Card

    2 x 2 * 10GE

    physical.m2.xlarge

    4 * 24 Core Broadwell EX Xeon E7-8890 V4 (2.2 GHz)

    4096 DDR4 RAM (GB)

    2 * 600GB SAS HDD RAID 1 + 14 * 1.8TB SAS HDD RAID 50 + 2 * 1.6TB NVMe SSD Card

    2 x 2 * 10GE