Free application: "Infrastructure in the Era of Big Models: A Guide to Building a Big Model Computing Center", the author will help answer questions

EEWORLD社区

Free application: "Infrastructure in the Era of Big Models: A Guide to Building a Big Model Computing Center", the author will help answer questions [Copy link]

 

Big models have been one of the hot topics in recent years. The basis for the booming development of big models is the computing power and infrastructure designed for their needs. This book explains in depth how to build infrastructure for big models, and analyzes the design solutions of well-known cases in the industry based on the TOGAF methodology.

"Infrastructure in the Big Model Era: A Guide to Building a Big Model Computing Center" has a total of 13 chapters. Chapter 1 explains the demand for infrastructure in the era of AI and large models; Chapter 2 explains the combination of software programs and dedicated hardware, involving GPU parallel computing libraries, machine learning program development frameworks, and distributed AI training; Chapter 3 analyzes the hardware architecture of GPUs, involving the overall design of GPUs, the Nvidia GH100 chip architecture, and GPUs with other Hopper architectures; Chapter 4 explains the design and implementation of GPU servers; Chapter 5 explains the I/O framework system on which machine learning relies; Chapter 6 explains the network design and implementation of GPU clusters; Chapter 7 explains the GPU board computing power scheduling technology; Chapter 8 explains the GPU virtualization scheduling solution; Chapter 9 explains the network virtualization design and implementation of GPU clusters; Chapter 10 explains the storage design and implementation of GPU clusters; Chapter 11 explains how to design and implement a better development and operation platform for machine learning applications based on cloud native technology; Chapter 12 explains the management and operation of cloud-based GPU clusters, involving cloud operation and maintenance platforms, cloud operation platforms, and cloud audit platforms; Chapter 13 is based on a GPU computing platform landing case that serves machine learning, showing how to perform demand analysis, design, and implementation for machine learning applications.

Whether you are an undergraduate or graduate student majoring in computer science and artificial intelligence in colleges and universities, or a researcher or engineer interested in parallel computing technology, cloud computing technology, high-performance storage and high-performance network technology, you can refer to and read this book. This event can invite the author to participate in the discussion of #没读懂# and #聊一聊#, helping readers of the book gain more. Interested netizens are welcome to participate.

how to apply

>> Click I want to apply , carefully fill in the reason for application, reading sharing plan, etc., and you will have the opportunity to get the book for free. There is 1 quota for evaluation coupons in this event, first come first served.

Event Schedule

Application period: July 2 to July 18
Selection announcement: All shortlisted candidates will be announced before July 22
Reading period: July 29 to September 30
Awarding time: within two weeks after the event ends

Reading requirements

(1) After receiving the book, please post your reading experience with a title of your own in the special edition of the EEWorld Evaluation Center . During the event period, complete the reading sharing plan you submitted and share at least 3 experiences with substantive content . The progress is to submit at least one article every 15 days. 100% original first release is required. Plagiarism will be banned. The experience can include: evaluation of the book content, understanding, and trying the experimental process in the book, etc.;

(2) Initiate #Ididn’tUnderstand# and #Let’s Chat# discussions (authors can be invited to participate regularly): During the reading process, each netizen can initiate a discussion post with #Ididn’tUnderstand# or #Let’s Chat# + a self-made topic based on his or her own reading comprehension. Each post with substantive discussion can be awarded 2 to 10 points depending on the content and popularity of the discussion. The maximum score for this part can be 20 points.

Activity exchange group

Add "helloeeworld" as a friend on WeChat, and start the conversation: Join the AI technology exchange group and join the group to communicate.

Scan the QR code and start chatting: join the AI technology exchange group.

For more event information, click here to go to the event page .

hustzxx

Thanks for sharing! I hope I can apply for it!