Tesla details how it finds punishing defective cores on its million-core Dojo supercomputers — a single error can ruin a weeks-long AI training run

Tesla's Stress tool detects and disables faulty cores in Dojo wafer-scale processors, which power Dojo clusters with millions of cores, without interrupting AI training.

https://www.tomshardware.com/pc-components/cpus/tesla-details-how-it-finds-punishing-defective-cores-on-its-million-core-dojo-supercomputers-a-single-error-can-ruin-a-weeks-long-ai-training-run

Utworzony 2d | 7 cze 2025, 14:10:05


Zaloguj się, aby dodać komentarz

Inne posty w tej grupie

 How To Find Files in the Linux terminal

In this how-to we will look at the find command and a range of additional arguments which will give us a variety of approaches to finding files and directories.

9 cze 2025, 17:10:05 | tomshardware.com
 How To Copy Files and Directories in the Linux terminal

Copying files and directories using the Linux command line is efficient and can be used while sitting at the machine or when accessing a system remotely.

9 cze 2025, 17:10:05 | tomshardware.com
 How To Kill a Process in the Linux terminal

Sometimes things go wrong and we have to kill a running process. In this how-to we show you how easy this can be via the terminal.

9 cze 2025, 17:10:04 | tomshardware.com
 Nvidia RTX 5050 won't use GDDR7 memory— entry-level GPU tipped to use Samsung & SK hynix GDDR6 modules instead

A media outlet has reportedly received confirmation from Nvidia's board partners that the upcoming RTX 5050 will receive GDDR6 memory rather than GDDR7.

9 cze 2025, 17:10:03 | tomshardware.com
 Micron seeks reversal of court order to share 73 pages of sensitive data with China's banned YMTC chipmaker — Micron strives to protect IP from Chinese chip firm on the entity list

Micron has petitioned the Supreme Court to reverse earlier rulings that granted YMTC’s legal team access to 73 pages of confidential 3D NAND documents, citing national se

9 cze 2025, 14:40:08 | tomshardware.com
 Crucial's 2TB T500, one of the fastest PCIe Gen 4 SSDs available at 7400MB/s, is now only $132

Crucial's T500 2TB SSD is on sale for just $132. A great price for one of the fastest Gen 4 drives on the market.

9 cze 2025, 14:40:08 | tomshardware.com
 ChatGPT 'got absolutely wrecked' by Atari 2600 in beginner's chess match — OpenAI's newest model bamboozled by 1970s logic

It is claimed that OpenAI’s ChatGPT 4o model “got absolutely wrecked on the beginner level” of Atari Chess on an Atari 2600 console from the 1970s.

9 cze 2025, 14:40:08 | tomshardware.com