登录/注册
您的应用程序运行在什么读写混合状态下? 块大小? Those simple questions have extremely complex answers in a data center running AI training, 关系型和NoSQL数据库, 虚拟化/容器化堆栈, 等等....... 数据中心 固态硬盘 specification sheets typically show performance at 4KB 100% random read, 写, 队列深度(QD)为256，顺序读写为128KB，为70/30. 但是在16KB 80/20下会发生什么呢? 当您的应用程序在驱动器上没有达到256队列深度时会发生什么? 最重要的是, how do 固态硬盘 vendors compare at more realistic block sizes and when latency is considered?
In this blog we’ll look at performance data for 微米’s new 7500 NVMe™ 固态硬盘 — a mainstream drive we built with real world performance at the forefront. 我们将检查4KB的混合工作负载性能, 8KB, and 16KB block sizes and dive into 6x9’s QoS latency in an 80/20 workload at 16KB block size. We’ll compare performance data to 微米’s previous-gen mainstream NVMe 固态硬盘 as well as two data center competitor NVMe 固态硬盘s.
The results are proof that 微米 built our 7500 to not only hit the max spec sheet 4KB numbers, but to be best in class in the types of 混合 workloads (both read/写 mix and block size) common in today’s data center.
乱成一团
让我们检查队列深度为32的4KB随机工作负载, 表示驱动器在正常压力水平下工作. These range from 写 heavy at 50% read / 50% 写, to read intensive at 90% read / 10% 写.
Here we see 微米’s 7500 固态硬盘 slightly outperforming the previous gen and 清晰的ly outperforming both competitors.
那太好了, but it’s not much of an improvement for the 7500; what happens when we move to larger block sizes like 8KB and 16KB?
在8 kb, the 7500 gains a measurable lead over previous gen and a commanding lead over the competition.
在16 kb, that delta widens a little further and 清晰的ly positions the 7500 as the best-in-class mainstream NVMe 固态硬盘 for 混合 workloads.
质量与数量
Another critical performance metric for 存储 is quality of service (QoS); how long does it take for your 固态硬盘 to return a request? QoS以百分位数来衡量, 例如, a 99% percentile latency of 1 millisecond means 1% of your IOs took longer than 1ms. In large data centers with many thousands of drives pushing many millions of IOs per second, 6x9的延迟变得至关重要.
网上交易处理, 推荐引擎, 实时分析, 人工智能, content distribution and financial trading are examples of workloads that benefit from low latency. Not just for hyperscalers, but also traditional enterprise data centers running databases. Both must deliver on service level agreements (SLAs) with their customers — milliseconds matter.
Let’s examine an 80% read / 20% 写 workload with 16KB IOs at queue depths ranging from 1 to 32.
该图显示了6x9在y轴上的读取延迟和在x轴上的IOPS. Higher performance is characterized by being lower on the Y-axis and further to the right on the X-axis. 该线上的每个点表示从1到32的队列深度. We set max latency on the graph to 4ms to increase the resolution of the comparison. 这切断了竞争者B的QD32度量(8ms).
Here we see significant decrease in 6x9’s latency from 微米’s previous gen to the 7500. We also see that the 7500 now beats the previous best-in-class drive at every queue depth.
Competitor B has slightly lower latency below QD8, but that is a symptom of much lower performance. 在QD8之上，竞争者B看到QoS延迟急剧增加.
现实世界的主流NVMe
The 美光7500固态硬盘 is the most advanced mainstream PCIe® Gen4 data center 固态硬盘 and first with 200+ layer NAND, 提供优于竞争对手的QoS和性能1.
美光7500 NVMe数据中心固态硬盘
创建一个新类别的sub-1毫秒驱动器
The 微米 7500 NVMe 固态硬盘 is the first mainstream data center 固态硬盘 to use 200+ layer NAND, and this leading technology helps to deliver excellent performance and QoS on a broad range of 混合 workloads and IO sizes.
美光制造的7500固态硬盘采用了世界上最先进的232层NAND, mass production NAND in the world today — and combined it with our advanced controller and firmware to deliver incredible results.
美光7500固态硬盘满足了业界领先的QoS需求.9999% (6x9s) 4K read-intensive 4K workload latencies under 1 millisecond while delivering over 1 million IOPS.
美光7500也有一些共同点, 混合, 随机工作负载，帮助驱动器提高流行数据库的性能, 包括甲骨文, MySQL, RocksDB, 和Microsoft SQL Server. 所有数据库都是多线程的, and that means multiple queues go out and are waiting for the slowest operation before results can be delivered. 这就是一致的延迟对用户产生巨大影响的地方.
欲知详情，请浏览 7500沙巴体育结算平台页面 and contact your sales representative to get them in your lab — and then into production!
附加信息:
注1:对于gen4u.2/U.3 mainstream competitive 固态硬盘 suppliers with at least 10% data center 固态硬盘 market share as of August ‘23 as noted in Forward Insights analyst report 固态硬盘 Supplier Status_Q2/Q3.
Note 2: Queue depth (QD) is the number of outstanding I/O requests per target per thread. In this case, QD 128 is used to as a common comparable QD across competitive drives. For example Microsoft generally uses QD 128 for benchmark performance analysis and to stress the system CPU under test with higher IOPS, 更多信息请访问:http://learn.微软.com/en-us/azure/virtual-machines/disks-benchmarks
Note 3: All results from internal testing in 微米’s data center workload engineering (DCWE) labs.
