ជំហានដំបូងក្នុងបន្ទះឈីប និងហេដ្ឋារចនាសម្ព័ន្ធ AI
វិស័យនេះគ្របដណ្តប់លើបន្ទះឈីប (ឧបករណ៍ពាក់កណ្តាលចម្លង) ដែលគណនា និងចងចាំ ព្រមទាំងមជ្ឈមណ្ឌលទិន្នន័យ និងអគ្គិសនីដែលដំណើរការ AI លើបន្ទះឈីបទាំងនោះ.
ហេតុអ្វីវាសំខាន់នៅពេលនេះ
AI runs on chips in the end. Training and using a large AI takes huge numbers of AI chips and fast memory, so who designs the chips and who makes them has become the center of the AI race.
A data center holding tens of thousands of chips can use as much electricity as a small city. That is why chip news now also covers factories and equipment, export controls between countries, and power grids and cooling.
របៀបដែលវាដំណើរការ
Design · Draw the chip's circuits. A company that only designs and leaves production to others is called fabless.
Production · A foundry etches very fine circuits onto a wafer to make the chips. Equipment and yield decide how good it is.
Packaging · Fast memory is attached next to the computing chip and the two are packaged as one.
Data center · Tens of thousands of servers with chips installed are gathered in one place to train and run AI.
Power and cooling · AI infrastructure also means cooling the heat the chips give off and supplying electricity that never cuts out.
How one chip ends up running AI, shortened to five steps.
គោលគំនិតដើម្បីចាប់ផ្តើម
ឧបករណ៍ពាក់កណ្តាលចម្លងវត្ថុធាតុដែលអាចចម្លងអគ្គិសនី ឬរារាំងវា។ បន្ទះឈីបតូចៗដែលធ្វើពីវាក៏ត្រូវបានគេហៅថាឧបករណ៍ពាក់កណ្តាលចម្លងផងដែរ។.GPU (graphics processing unit)បន្ទះឈីបដែលធ្វើការគណនាដូចគ្នាក្នុងចំនួនដ៏ច្រើនក្នុងពេលតែមួយ។ វាគឺជាបន្ទះឈីបដែលត្រូវបានប្រើប្រាស់ច្រើនបំផុតដើម្បីបណ្តុះបណ្តាលនិងដំណើរការ AI.HBM (high bandwidth memory)អង្គចងចាំ AI ដែលបង្កើតឡើងដោយការដាក់ជង់បន្ទះឈីបអង្គចងចាំជាស្រទាប់ៗ ដើម្បីឱ្យទិន្នន័យផ្លាស់ទីចេញចូលបានយ៉ាងលឿន.រោងចក្រផលិតរោងចក្រដែលផលិតឧបករណ៍ពាក់កណ្តាលចម្លងដែលរចនាដោយក្រុមហ៊ុនផ្សេងទៀត ឬអាជីវកម្មនោះ.Process nodeឈ្មោះសម្រាប់ភាពស្តើង និងការវេចខ្ចប់យ៉ាងតឹងណែននៃសៀគ្វីរបស់បន្ទះឈីប។ លេខកាន់តែតូច បច្ចេកវិទ្យាកាន់តែទំនើប.មជ្ឈមណ្ឌលទិន្នន័យ AIកន្លែងដ៏ធំមួយដែលប្រមូលផ្តុំបន្ទះឈីបដែលមានប្រសិទ្ធភាពខ្ពស់រាប់ម៉ឺនទៅរាប់សែនបន្ទះនៅកន្លែងតែមួយដើម្បីបណ្តុះបណ្តាល និងដំណើរការ AI.ការយល់ខុសទូទៅ
"3 nm" means the circuits are really 3 nm wide.
These days it is closer to a label that names a generation. Each company uses different standards, so it is hard to compare by the number alone.
With a good chip alone, you can run AI.
Memory that feeds data to the chip quickly, links between chips, electricity and cooling all have to support it. That is why memory and data center companies get attention too.
One country or one company makes all of a chip.
Design, equipment, materials, production and packaging are spread across many countries and companies. If even one is blocked, the whole chain shakes, so supply chains and export controls are often in the news.
High Bandwidth MemoryAI 반도체 곁에서 데이터를 빠르게 공급하는 적층 메모리예요.NVIDIAGPU와 AI 컴퓨팅·로봇 개발 플랫폼을 만드는 곳이에요.Samsung Electronics메모리·반도체와 휴대폰·가전을 만드는 곳이에요.SK hynixHBM을 비롯한 메모리 반도체를 만드는 곳이에요.Jensen HuangNVIDIA를 공동 창업한 인물이에요.Lisa SuAMD의 반도체·AI 컴퓨팅 소식에 등장하는 인물이에요.បន្ទាត់ពេលវេលាបន្ទះឈីប និងហេដ្ឋារចនាសម្ព័ន្ធFrom 1947, when the transistor was first demonstrated, to today, when the chips and memory that run AI are at the center of the industry.Read more chip conceptsKeep reading, from fabless and yield to EUV and advanced packaging.Data centers and powerRead about the power grids and cooling that support AI infrastructure.Chip newsGet news about chips and data centers in easy summaries.