#linuxtools

[govindhtech]

Micron SSD: Gen5 NVMe SSDs for Dell PowerEdge Servers

Micron SSDs

Gen5 NVMe SSDs

Micron presented its industry-leading research on AI training model offload to NVMe, collaborating with teams at Dell and NVIDIA. In a Dell PowerEdge R7625 server equipped with Micron’s upcoming high-performance Gen5 E3.S NVMe SSD, the Data Centre Workload Engineering team at Micron tested Big Accelerator Memory (BaM) with GPU-initiated direct storage (GIDS) on the NVIDIA H100 Tensor Core GPU with assistance from Dell’s Technical Marketing Lab and NVIDIA’s storage software development team.

More Memory using NVMe?

The standard procedure for training huge models whose sizes are increasing quickly is to use as much HBM as possible on the GPU, followed by as much system DRAM. If a model cannot fit in HBM + DRAM, it is then parallelized over many NVIDIA GPU systems.

The cost of parallelizing training over numerous servers is high since data must travel over system and network links, which can quickly become bottlenecks. This is especially true for GPU utilisation and efficiency.

What if NVMe could be used as a third tier of “slow” memory by Micron to avoid having to divide an AI training job across many GPU systems? Exactly that is what BaM with GIDS accomplishes. It transfers the data and control routes to the GPU by replacing and streamlining the Gen5 NVMe SSD driver. How does that perform then?

Results of Baseline Performance

The open-source BaM implementation mentioned above includes the BaM Graph Neural Network (GNN) benchmark, which was used to execute all of the test results displayed.

This initial test illustrates the results with and without BaM when GIDS is turned on. As a test example without particular storage software, a common implementation of Linux mmap was used to fault memory accesses through the CPU to storage.Image Credit to Micron

Using a Micron 9400 Gen4 NVMe SSD and an NVIDIA A100 80GB Tensor Core GPU, the mmap test took 19 minutes. It took 42 seconds with BaM and GIDS deployed, a 26x increase in performance. The benchmark’s feature aggregation component, which depends on storage performance, shows that performance improvement.

Dell Laboratories’ Gen5 Performance

Micron aimed to demonstrate at GTC how successfully their future Gen5 NVMe SSD performed AI model offload. In order to obtain access to a Dell PowerEdge R7625 server with an NVIDIA H100 80GB PCIe GPU (Gen5x16), Micron teamed up with Dell’s Technical Marketing Labs. With their outstanding help, Micron successfully completed testing.

SSD performance affects feature aggregation. Its execution duration accounts for 80% of the whole runtime, and it improves by twice between Gen4 and Gen5 NVMe SSD. Training and sampling are dependent on the GPU; an NVIDIA A100 to an H100 Tensor Core GPU can enhance training performance five times. For this use case, high-performance Gen5 NVMe SSDs are necessary, and a pre-production sample of Micron SSD i.e. Gen5 NVMe SSD exhibits roughly double the performance of Gen4.GNN WORKLOAD PERFORMANCEMICRON GEN5 H100MICRON GEN4 A100GEN5 VS GEN4 PERFORMANCEFeature Aggregation (NVMe)18s25s2xTraining (GPU)0.73s3.6s5xSampling3s4.6s1.5xEnd-to-End time (Total of Feature Aggregation + Training + Sampling)22.4s43.2s2xGIDS + BaM Accesses/s2.87M1.5M2x

What Is Micron SSD Being Affected by BaM With GIDS?

The typical Linux tools to view the IO metrics (IOPs, latency, etc.) are inoperable since BaM with GIDS substitutes the Gen5 NVMe SSD driver. After tracing the BaM using GIDS GNN training workload, Micron discovered some astonishing findings.

BaM with GIDS operates at almost the drive’s maximum input/output speed.

For GNN training, the IO profile is 99% tiny block reads.

The SSD queue depth is 10-100 times greater than what Micron anticipates from a “typical” data centre CPU demand.

This is a new workload designed to maximise Gen5 NVMe SSD performance. Multiple streams can be managed by a GPU in parallel, and the BaM with GIDS software will optimise and manage latency, resulting in a workload profile that might not even be feasible to execute on a CPU.

In summary

As the AI sector develops, clever solutions for GPU system efficiency and utilisation become increasingly crucial. Larger AI issue sets can be solved more effectively with the help of software like BaM with GIDS, which will increase the efficiency of AI system resources. Extending model storage to Gen5 NVMe SSD will have an impact on training times, but this trade-off will enable larger, less time-sensitive training jobs to be completed on fewer GPU systems, hence increasing the effectiveness and total cost of ownership (TCO) of deployed AI gear.

Specifics of the Hardware and Software:

Workload: Complete Training for IGBH and GIDS.

The Data Centre Workload Engineering team at Micron measured the Gen5 NVMe SSD performance, whereas the NVIDIA storage software team measured the baseline (mmap) performance on a system that was comparable.

Systems being evaluated:

Gen4: NVIDIA A100-80GB GPU, Ubuntu 20.04 LTS (5.4.0-144), NVIDIA Driver 535.129.03, CUDA 12.3, DGL 2.0.0, Dual AMD EPYC 7713 64-core, 1TB DDR4, Micron 9400 PRO 8TB

GL 2.0.0, CUDA 12.3, NVIDIA H100-80GB GPU, Ubuntu 20.04 LTS (5.4.0-144), NVIDIA Driver 535.129.03, Dell R7625, 2x AMD EPYC 9274F, 24-core, 1TB DDR5, Micron Gen5 NVMe SSD

Work based on the publication “Introduction of GPU-Initiated High-Throughput Storage Access in the BaM System Architecture”

Read more on govindgtech.com

[rainsoft]

Web Developer Tools: A Great Suite of Linux Tools in a Single Interface for Web Developers and More - Rainsoft

[dailytech-info]

Any Linux users . . 🌐www.techyrick.com . . #linuxuser #linuxindonesia #firefoxbikes #linuxubuntu #linuxfoundation #firefox #archlinux #linuxwindows #debianlinux #linux #linuxmint #linuxterminal #ubuntulinux #firefoxcycles #firefoxhusky #fedoralinux #linuxlover #linuxos #linuxmasterrace #linuxtools #linuxsetup #firefoxos #linuxtips #linuxisbetter #kalilinuxtools #linuxhosting #kalilinux #gnulinux #linuxadmin #linuxfan https://www.instagram.com/p/CQi6XPeBGK5/?utm_medium=tumblr

[stockbitcoin]

Great Digital Signage Content Creation Tools

Choose the right digital signage content creation tools

Digital signage content can help you get your message out, but it only has an impact on your business and team if it’s relevant and compelling. Luckily, turning your content strategy into memorable visuals that get noticed doesn’t have to cost a lot or take hours of time. Want to know the secret to creating great content, even if you don’t have a graphic designer on staff? Use the right tools. That’s why our list of 27 digital signage content creation tools is a must-have for anyone managing screen content.  Read the full article

[ssiddique]

3 methods to Install and Configure Nano on Windows

3 methods to Install and Configure #Nano on #Windows #linux #linuxtool #codeeditor

This tutorial explains how you can install the nano text editor on a Windows PC. I will show two different methods. In the first method, we will use Chocolatey to install nano. If you have not installed Chocolatey, open PowerShell and run the following command to install Chocolatey. [Method 1] : Use Chocolatey to install Nano on Windows Step 1: Install Chocolatey using Powershell…

[hacknews]

Official Evernote Linux Client Is Coming; GParted 1.1.0 Also Released #evernote #evernotelinuxclient #gparted #gpartedlive #linux #linuxtools #opensource #hacking #hacker #cybersecurity #hack #ethicalhacking #hacknews

[dailytech-info]

🌐www.techyrick.com . . #linuxtutorial #linuxcnc #linuxindonesia #linuxmasterrace #linuxhosting #linuxmint #linuxadministrator #ubuntulinux #embeddedlinux #linuxkernel #firefoxhusky #linuxusers #firefoxos #linuxubuntu #linuxlover #linuxfan #linux #linuxfoundation #linuxsetup #linuxtools #kalilinux #firefoxcycles #linuxmemes #firefoxbikes #archlinux #linuxteach #linuxcommands #linuxwindows #firefox #linuxgaming https://www.instagram.com/p/CQYdu9Nhex8/?utm_medium=tumblr