其實有點可惜沒有講到關於應用上MI300有多大的門檻要跨這件事。 看新聞,就知道所有在做AI的大公司都瘋狂的在搶NVIDIA的AI運算硬體,我都開玩笑說NVIDIA的產量限制了整個人類的進步XD 那為麼有更好的硬體不去用,還要跟人去搶NVIDIA的產品? 有原因的嘛~ 真的光是硬體設計夠優秀是沒有用的,沒辦法把軟體開發環境等整個生態系建立起來,搞得大家用起來門檻很高很麻煩,沒辦法“It just works",那就是失敗。大部份會去做AI Training的人剛開始入門的時候用就是學的某個框架,之後要大量使用的時候,也都會直覺得待在這個生態系。因為我只是要training model,專注的是思考新的訓練的方法,參數怎麼調等等,不太想花時間在轉換生態系這件事上面。舉個例子,大家用Windows,是生態系的關係,不是因為Windows有多好。Linux即使再好,再安全,再開放而且都能客製化還免費,這跟「一般使用者」沒有什麼關係,他們還是不想花時間在轉換生態系這件事上。再舉個例子,寫程式不就是有個基本的文字編輯器(notepad或vi)跟terminal就好了嗎?幹嘛要用IDE還被綁住? 如果今天是學術機構(不像企業有那麼強的時間跟競爭壓力),或有天才型工程師,能不受框架限制把所有硬體都運用自如,那很好啊?!就可以去用MI300或其它的硬體,沒必要被NVIDIA綁架。不過很可惜的是,真正需要大量AI硬體的大企業,他們有時間跟競爭的壓力,他們就是那個會被生態系綁架的那個「一般使用者」。
@judahxiiiyoung7320 Жыл бұрын
我靠,你終於回來了!你不要走啊! #好了評論完了可以開始看影片了
@limitli1117 Жыл бұрын
Twitter有人推荐。看完感觉太厉害了。知识量强大。讲述清晰。❤
@dare-to-come-down-ice-bird Жыл бұрын
感動 竟然回來了!!!!
@Tech4AllYall Жыл бұрын
我剛剛連你名字一起看,看成還敢回來啊冰鳥
@takahirokan Жыл бұрын
感謝分享,每次都能學到很多。之前還以為以後都看不到您的影片了🥲
@isthiswhat3961 Жыл бұрын
WOW 歡迎回歸! 還以為你不拍了
@petercandylee Жыл бұрын
From Tom’s Hardware The MI300 3D design allows for incredible data throughput between the CPU, GPU and memory dies while also allowing the CPU and GPU to work on the same data in memory simultaneously (zero-copy), which saves power, boosts performance, and simplifies programming.
There are a couple of reasons why MI300 is not selling 1. It is not ready - it won't be ready until next year 2024. 2. The older versions (MI250, MI100) are not selling well because the supporting software is not mature. But this will change because large software houses Microsoft, Pytorch, and Hugging Face are helping AMD to optimize its software. Large tech companies want to have a second source.
NVLink is a connection between the CPUs and GPUs , so between sockets. The Infinity Fabric is many things more as it's build within the CPU/GPU, provides a link between dies and across sockets. Infinity Fabric isn't a kind of hypertransport but a superset of it. NVLink 是 CPU 和 GPU之间的连接,也是插槽之间的连接。 Infinity Fabric (AMD) 具有更多功能,因为它构建在 CPU/GPU 内,提供芯片之间和跨插槽的链接。 Infinity Fabric 不是一种超传输,而是它的超集
@Baliyoho Жыл бұрын
強勢回歸!
@sjcabbw11 ай бұрын
SOC, system on chip, 系統一直在被集成一個單一晶片. 所以根本無所謂 系統重要或晶片重要 , 因為今天的系統可能就是明天的晶片.
@張硯棠 Жыл бұрын
非常感謝解說如此清楚😀 最近IBM New Analogue Chip看來是很前瞻的設計,是否能講解一下😂 存算合一使是否才是打破馮諾依曼瓶頸?
Actually, in LLM, higher memory is way more important. If you have to sync intermediate values, it's a lot slower. So if the chip can hold the entire model, that would be the fastest, with data sharding only.