chevron_leftBack

US - English China - 简体中文 India - English Japan - 日本語 Malaysia - English Singapore - English Taiwan – 繁體中文

  • Data center storage insights center

    美光中心提供数据中心存储的技术和工作负载见解-直接来自我们的技术专家.

  • test alt

Storage technology insights zone

阅读这些博客，从美光存储技术专家那里获得最新的数据中心存储见解.

SEE STORAGE INSIGHTS

+

Storage technical resources zone

Whether you build on leading-edge open-source or industry titan software, these technical briefs, 白皮书和其他文件按工作负载类型进行分类，以帮助您快速准备好生产.
SEE TECH RESOURCES

+



Storage technology insights zone

博客直接从美光的数据中心存储技术专家的见解.

WEKA storage with the Micron 6500 ION SSD supports 256 AI accelerators
Most AI training data lives on shared storage. 本文描述了我们在AI工作负载中使用MLPerf storage基准测试WEKA存储集群中30TB Micron 6500 ION NVMe ssd时观察到的惊人结果.
SC'23 Amazing AI Scaling: Weka + Micron 6500 ION >
+
A person using a laptop in a server room

FMS ’23: ATS/ATC support for Performance SSDs
地址转换服务(ATS)和地址转换缓存(ATC)是很容易理解的系统概念, but their usage in NVMe storage is still novel. 本博客讨论了美光在高性能NVMe固态硬盘中使用ATS/ATC的一些策略.
See how ATS/ATC supports SSD performance >
+
Multiple racks of servers in a dim room with indicator lights

Scanning the horizon for the peak of AI performance
当人工智能(AI)专家将创新的软件定义存储(SDS)与高性能NVMe ssd相结合时会发生什么? Good things. Fast and scalable things.
Scanning AI performance peak >
+
A man examining the cords in a server room with a clipboard on his hand.

Micron 7500: Mainstream NVMe for the real world
The Micron 7500 NVMe™ SSD has low and consistent latency, which enables rapid, reliable responsiveness for demanding data center workloads. 本博客讨论了Micron 7500在各种混合读写工作负载和块大小下的性能和延迟, demonstrating its best-in-class QoS versus competitive SSDs.
Micron 7500 SSD mixed workload QoS and performance >
+
A man in a white jacket is reading from his tablet in server room

现实生活中的工作负载允许更高效的数据粒度，并支持非常大的SSD容量
不断增长的SSD容量和不断增加的功耗需要更现代的数据映射粒度. 在过去，业界对于做出任何会对SSD寿命产生负面影响的改变都很犹豫. 来自实际应用程序的最新重点数据表明，情况可能并非如此，而且有一条通往更有效映射的道路.
Understanding large indirection units for high-capacity SSDs >
+
Server room in pink and blue hue

The Micron 9400 NVMe SSD is the top PCIe Gen4 SSD for AI storage
表征AI训练系统的存储工作量面临MLPerf存储基准套件旨在解决的两个独特挑战:AI加速器的成本和可用数据集的小尺寸. This blog shows how the MLPerf Storage benchmark addresses both.
MLPerf storage benchmark on Micron 9400 NVMe >
+
Blue rendering lines to show data stream

Storage for AI training: MLPerf Storage on the Micron® 9400 NVMe™ SSD
MLPerf Storage工具通过再现真实的AI工作负载，对各种模型的存储进行基准测试非常有帮助. 阅读这篇博客，了解美光如何使用MLPerf测试AI工作负载的存储
Benchmarking AI storage with MLPerf >
+
Blue, orange and red rendering lines to show rapid speed

Identifying latency outliers in workload testing
当运行和收集RocksDB的工作负载跟踪时，我们有时会看到较大的延迟峰值. With this blog, 我们将讨论用于识别混合读写工作负载中延迟峰值的根本原因的方法. 
How to identify latency outliers >
+
MORE ON PAGE 2

Storage technical resource zone

美光在解决方案设计和与行业软件领导者的工作负载协作方面拥有数十年的经验. Whether you build on leading-edge open-source or industry titan software, these resources get you production-ready in a flash.

Technical assets by workload

技术简报、白皮书等为您的下一个项目提供了一个经过实验室验证的起点.
Ceph workloads logo image
Ceph is a scalable, simple, 开源存储软件包，用于现代以数据为中心的应用程序，如人工智能, machine learning, data analytics and emerging cloud solutions.

了解美光如何将Ceph的软件定义存储创新与美光的先进技术相结合. Our testing found the Micron 6500 ION SSD provides a true alternative to HDDs for Ceph deployments. We also found that erasure coding paired with Micron 7450 SSDs 提供令人印象深刻的性能和低成本，以提高您的下一个Ceph部署的TCO.
weka logo image
美光和WEKA将美光NVMe™ssd的性能与WekaFS™结合在一起，为您的数据密集型应用程序提供动力. We have written this technical brief 以帮助您了解哪种美光NVMe SSD和服务器配置是您部署的正确选择.

Also read this blog 以了解更多与美光合作对您的业务的好处, WEKA, AMD and Supermicro.
v m ware logo image
VMware® provides a set of server design recommendations for VMware® vSAN ReadyNode™ configurations. 这些全闪存VMware vSAN平台设计提供易于部署和管理的可预测的高性能.

Micron data center SSDs offer ReadyNode-certified storage options 为VMware vSAN解决方案提供最先进的数据中心效率. Also read how Micron SSDs 使用使用VMware vSAN ESA的联想®ThinkAgile™VX VX系列服务器提高存储性能.
Artificial intelligence icon image

美光科技使人工智能成为可能，为新一代提供更快的技术, more intelligent global infrastructures. Read how the Micron 9400 and NVIDIA GDS 共同努力提高性能和响应时间，以充分利用GPU部署.

database icon image

SQL

SQL Server工作负载需要具有快速输入/输出(I/O)性能的存储, 快速吞吐量和高耐用性，以管理当今现代应用的需求. Micron’s portfolio of SSDs enable efficient, optimized analytics platforms. 

We tested a wide variety of SQL Server configurations and built a guide to 使用美光ssd、DRAM和软件配置优化Microsoft SQL Server分析工作负载 to maximize your return on investment (ROI).

Also, read how the Micron 6500 ION and XTR SSDs optimize SQL Server analytics workloads with unbeatable value and performance.
database icon image

NoSQL

NoSQL workloads such as social media platforms, 内容管理系统和实时分析需要可扩展的存储, high performing and reliable.

We tested NoSQL analytics workloads with the Micron 6500 ION SSD and Cassandra. 我们将其与竞争对手进行了比较，我们的测试显示出更好的结果.

We also tested the Micron 9400 SSD with the popular NoSQL Aerospike database and found it a great fit for demanding workloads like Aerospike.
Noisy neighbor workloads icon image

In a cloud multi-tenant architecture, a tenant that monopolizes shared resources like storage, network or compute is called a “noisy neighbor.”

Read how the Micron 9400 SSD is the best performance SSD for noisy neighbor environments, so workloads perform their best no matter the neighborhood

rocks DB logo image
RocksDB is a high-performance embedded database key-value store, 针对多个CPU内核进行了优化，并针对I/ o绑定工作负载有效地使用了快速存储(如Micron ssd).

Read how the Micron 9400 SSD enables RocksDB performance for the most demanding workloads. Also, learn how the 美光7450 SSD为混合RocksDB工作负载提供了令人印象深刻的性能 while outperforming leading competitors.

Helpful links

Micron customer and partner storage success
美光与我们的客户和合作伙伴合作，提供高性能存储解决方案，丰富生活 for all. See how we are transforming storage through the development of innovative, cooperatively designed solutions.
+

Micron data center SSDs
Made for the big demands of data center workloads, our SSDs deliver high performance, reliability, superior data protection, and optimal endurance to support your data center infrastructure.
+

Micron storage blog
Read about Micron’s storage innovation, partner updates and industry stories that impact your next deployment.
+
+