您现在访问的是微软AZURE全球版技术文档网站，若需要访问由世纪互联运营的MICROSOFT AZURE中国区技术文档网站，请访问 https://docs.azure.cn.
Lsv2 系列Lsv2-series
Lsv2 系列具有高吞吐量、低延迟、在 AMD EPYCTM7551 处理器上运行的直接映射的本地 NVMe 存储，且所有核心提升 2.55GHz，最大提升 3.0GHz。The Lsv2-series features high throughput, low latency, directly mapped local NVMe storage running on the AMD EPYCTM 7551 processor with an all core boost of 2.55GHz and a max boost of 3.0GHz. Lsv2 系列 VM 提供同时进行的多线程配置中的 8 到 80 个 vCPU。The Lsv2-series VMs come in sizes from 8 to 80 vCPU in a simultaneous multi-threading configuration. 每个 vCPU 有 8 GiB 内存，每 8 个 vCPU 有一个 1.92TB 的 NVMe SSD M.2 设备，并且在 L80s v2 上可以使用最多 19.2TB (10x1.92TB)。There is 8 GiB of memory per vCPU, and one 1.92TB NVMe SSD M.2 device per 8 vCPUs, with up to 19.2TB (10x1.92TB) available on the L80s v2.
备注
优化 Lsv2 系列 VM 以使用直接连接到 VM 的节点上的本地磁盘，而不是使用持久数据磁盘。The Lsv2-series VMs are optimized to use the local disk on the node attached directly to the VM rather than using durable data disks. 这可实现工作负载的更大的 IOPS/吞吐量。This allows for greater IOPs / throughput for your workloads. Lsv2 和 Ls 系列不支持通过创建本地缓存来增加持久数据磁盘可实现的 IOPS。The Lsv2 and Ls-series do not support the creation of a local cache to increase the IOPs achievable by durable data disks.
本地磁盘的高吞吐量和 IOPS 使得 Lsv2 系列 VM 非常适用于 NoSQL 存储（例如 Apache Cassandra 和 MongoDB）。这些存储将数据复制到多个 VM 上，以便在单个 VM 发生故障时实现持久性。The high throughput and IOPs of the local disk makes the Lsv2-series VMs ideal for NoSQL stores such as Apache Cassandra and MongoDB which replicate data across multiple VMs to achieve persistence in the event of the failure of a single VM.
若要了解详细信息，请参阅：针对 Windows 或 Linux 优化 Lsv2 系列虚拟机上的性能。To learn more, see Optimize performance on the Lsv2-series virtual machines for Windows or Linux.
ACU：150-175ACU: 150-175
高级存储：支持Premium Storage: Supported
高级存储缓存：不支持Premium Storage caching: Not Supported
实时迁移：不支持Live Migration: Not Supported
内存保留更新：不支持Memory Preserving Updates: Not Supported
VM 代系支持：第 1 代和第 2 代VM Generation Support: Generation 1 and 2
突发：支持Bursting: Supported
加速网络：支持Accelerated Networking: Supported
临时 OS 磁盘：不支持Ephemeral OS Disks: Not Supported
| 大小Size | vCPUvCPU | 内存 (GiB)Memory (GiB) | 临时磁盘1 (GiB)Temp disk1 (GiB) | NVMe 磁盘2NVMe Disks2 | NVMe 磁盘吞吐量3（读取 IOPS/MBps）NVMe Disk throughput3 (Read IOPS/MBps) | 非缓存数据磁盘吞吐量 (IOPS/MBps)4Uncached data disk throughput (IOPs/MBps)4 | 最大突发非缓存数据磁盘吞吐量 (IOPS/MBps)5Max burst uncached data disk throughput (IOPs/MBps)5 | 最大数据磁盘Max Data Disks | 最大 NIC 数Max NICs | 预期的网络带宽 (Mbps)Expected network bandwidth (Mbps) |
|---|---|---|---|---|---|---|---|---|---|---|
| Standard_L8s_v2Standard_L8s_v2 | 88 | 6464 | 8080 | 1x1.92 TB1x1.92 TB | 400000/2000400000/2000 | 8000/1608000/160 | 8000/12808000/1280 | 1616 | 22 | 32003200 |
| Standard_L16s_v2Standard_L16s_v2 | 1616 | 128128 | 160160 | 2x1.92 TB2x1.92 TB | 800000/4000800000/4000 | 16000/32016000/320 | 16000/128016000/1280 | 3232 | 44 | 64006400 |
| Standard_L32s_v2Standard_L32s_v2 | 3232 | 256256 | 320320 | 4x1.92 TB4x1.92 TB | 1.5M/80001.5M/8000 | 32000/64032000/640 | 32000/128032000/1280 | 3232 | 88 | 1280012800 |
| Standard_L48s_v2Standard_L48s_v2 | 4848 | 384384 | 480480 | 6x1.92 TB6x1.92 TB | 2.2M/140002.2M/14000 | 48000/96048000/960 | 48000/200048000/2000 | 3232 | 88 | 16000 +16000+ |
| Standard_L64s_v2Standard_L64s_v2 | 6464 | 512512 | 640640 | 8x1.92 TB8x1.92 TB | 2.9M/160002.9M/16000 | 64000/128064000/1280 | 64000/200064000/2000 | 3232 | 88 | 16000 +16000+ |
| Standard_L80s_v26Standard_L80s_v26 | 8080 | 640640 | 800800 | 10x1.92TB10x1.92TB | 3.8M/200003.8M/20000 | 80000/140080000/1400 | 80000/200080000/2000 | 3232 | 88 | 16000 +16000+ |
1 Lsv2 系列 VM 具有标准的基于 SCSI 的临时资源磁盘，用于 OS 分页/交换文件的使用（对于 Windows，为 D:；对于 Linux，为 /dev/sdb）。1 Lsv2-series VMs have a standard SCSI based temp resource disk for OS paging/swap file use (D: on Windows, /dev/sdb on Linux). 此磁盘为每 8 个 vCPU 提供 80 GiB 的存储空间、4000 IOPS 和 80 MBps 的传输速率（例如，Standard_L80s_v2 在 40000 IOPS 和 800 MBps 下提供 800 GiB）。This disk provides 80 GiB of storage, 4,000 IOPS, and 80 MBps transfer rate for every 8 vCPUs (e.g. Standard_L80s_v2 provides 800 GiB at 40,000 IOPS and 800 MBPS). 这确保 NVMe 驱动器可以完全专用于应用程序的使用。This ensures the NVMe drives can be fully dedicated to application use. 此磁盘是临时的，停止/解除分配时将丢失所有数据。This disk is Ephemeral, and all data will be lost on stop/deallocate.
2 本地 NVMe 磁盘是临时的，如果停止/解除分配 VM，这些磁盘上的数据都将丢失。2 Local NVMe Disks are ephemeral, data will be lost on these disks if you stop/deallocate your VM.
3 Hyper-V NVMe Direct 技术提供对安全映射到来宾 VM 空间的本地 NVMe 驱动器的无限制访问。3 Hyper-V NVMe Direct technology provides unthrottled access to local NVMe drives mapped securely into the guest VM space. 实现最大性能需要使用最新的 WS2019 版本或者 Azure 市场上的 Ubuntu 18.04 或 16.04。Achieving maximum performance requires using either the latest WS2019 build or Ubuntu 18.04 or 16.04 from the Azure Marketplace. 写入性能因 IO 大小、驱动器负载和容量利用率而异。Write performance varies based on IO size, drive load, and capacity utilization.
4 Lsv2 系列 VM 不为数据磁盘提供主机缓存，因为这不会让 Lsv2 工作负荷受益。4 Lsv2-series VMs do not provide host cache for data disk as it does not benefit the Lsv2 workloads.
5 Lsv2 系列 VM 可以通过突发方式提高其磁盘性能，一次长达 30 分钟。5 Lsv2-series VMs can burst their disk performance for up to 30 minutes at a time.
6 具有 64 个以上 vCPU 的 VM 需要以下受支持的来宾操作系统之一：6 VMs with more than 64 vCPUs require one of these supported guest operating systems:
- Windows Server 2016 或更高版本Windows Server 2016 or later
- Ubuntu 16.04 LTS 或更高版本，带 Azure 优化内核（4.15 内核或更高版本）Ubuntu 16.04 LTS or later, with Azure tuned kernel (4.15 kernel or later)
- SLES 12 SP2 或更高版本SLES 12 SP2 or later
- RHEL 或 CentOS 版本6.7 至6.10，使用 Microsoft 提供的 .LIS 包 4.3.1 (或更高版本) 安装RHEL or CentOS version 6.7 through 6.10, with Microsoft-provided LIS package 4.3.1 (or later) installed
- RHEL 或 CentOS 版本7.3，其中安装了 Microsoft 提供的 .LIS 包 4.2.1 (或更高版本) RHEL or CentOS version 7.3, with Microsoft-provided LIS package 4.2.1 (or later) installed
- RHEL 或 CentOS 版本7.6 或更高版本RHEL or CentOS version 7.6 or later
- UEK4 或更高版本的 Oracle LinuxOracle Linux with UEK4 or later
- Debian 9，带有向后移植内核 Debian 10 或更高版本Debian 9 with the backports kernel, Debian 10 or later
- 带有 4.14 内核或更高版本的 CoreOSCoreOS with a 4.14 kernel or later
大小表定义Size table definitions
- 存储容量的单位为 GiB 或 1024^3 字节。Storage capacity is shown in units of GiB or 1024^3 bytes. 比较以 GB（1000^3 字节）为单位的磁盘和以 GiB（1024^3 字节）为单位的磁盘时，请记住以 GiB 为单位的容量数显得更小。When comparing disks measured in GB (1000^3 bytes) to disks measured in GiB (1024^3) remember that capacity numbers given in GiB may appear smaller. 例如，1023 GiB = 1098.4 GBFor example, 1023 GiB = 1098.4 GB
- 磁盘吞吐量的单位为每秒输入/输出操作数 (IOPS) 和 Mbps，其中 Mbps = 10^6 字节/秒。Disk throughput is measured in input/output operations per second (IOPS) and MBps where MBps = 10^6 bytes/sec.
- 若要获得 VM 的最佳性能，应将数据磁盘数限制为每 vCPU 2 个磁盘。If you want to get the best performance for your VMs, you should limit the number of data disks to 2 disks per vCPU.
- 预期的网络带宽 是指跨所有 NIC 为每个 VM 类型分配的最大聚合带宽，适用于所有目标。Expected network bandwidth is the maximum aggregated bandwidth allocated per VM type across all NICs, for all destinations. 不保证能够达到上述预期上限，但目的是提供相关指导，以便为目标应用程序选择适当的 VM 类型。Upper limits are not guaranteed, but are intended to provide guidance for selecting the right VM type for the intended application. 实际的网络性能将取决于许多因素，包括网络拥塞、应用程序负载和网络设置。Actual network performance will depend on a variety of factors including network congestion, application loads, and network settings. 有关如何优化网络吞吐量的信息，请参阅为 Windows 和 Linux 优化网络吞吐量。For information on optimizing network throughput, see Optimizing network throughput for Windows and Linux. 若要在 Linux 或 Windows 中达到预期的网络性能，可能需要选择特定版本，或者需要优化 VM。To achieve the expected network performance on Linux or Windows, it may be necessary to select a specific version or optimize your VM. 有关详细信息，请参阅如何针对虚拟机吞吐量进行可靠的测试。For more information, see How to reliably test for virtual machine throughput.
其他大小和信息Other sizes and information
- 常规用途General purpose
- 内存优化Memory optimized
- 存储优化Storage optimized
- GPU 优化GPU optimized
- 高性能计算High performance compute
- 前几代Previous generations
定价计算器：定价计算器Pricing Calculator: Pricing Calculator
有关磁盘类型的详细信息：磁盘类型More information on Disks Types : Disk Types
后续步骤Next steps
了解有关 Azure 计算单元 (ACU) 如何帮助跨 Azure SKU 比较计算性能的详细信息。Learn more about how Azure compute units (ACU) can help you compare compute performance across Azure SKUs.